Sponsors
Sponsor Products
VFX Reference Platform - CY2016 Draft published
posted by Nick Cannon  on June 1, 2015, 3:55 p.m. (5 years, 2 months, 2 days ago)
25 Responses     0 Plus One's     0 Comments  
The Working Group behind the VFX Reference Platform, the target platform for building software for VFX, has just announced the draft of the Platform for next year, CY2016 at http://vfxplatform.com. This represents the set of components that all major software releases are expected to be built against from January 2016. This Draft version is still a work in progress and all feedback will be considered as we work towards the Final version of CY2016 which will be published around SIGGRAPH. Please send feedback privately to feedback@vfxplatform.com or publicly to the vfx-platform-discuss Google Group. The major change for CY2016 is a move to Qt 5 which requires a port of PySide (already in progress) and modifications to vanilla Qt to resolve issues that impact some vendors' tools. Our aim is for the final modified Qt to be available to anyone who wishes to target the CY2016 Platform, and for all major software vendors to build their CY2016 releases against the same modified Qt. More details will be shared in due course. For those attending SIGGRAPH, keep an eye out for our Birds Of A Feather session. Nick To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

Thread Tags:
  discuss-at-studiosysadmins 

Response from Nick Cannon @ July 28, 2015, 11:35 a.m.
Hi John,
No movement on the Windows side for now I'm afraid, the Platform is very much focused on Linux because that is the primary platform in VFX for creating integrated pipelines. There is the added complication that the Games industry tends to be the one leading the software vendors when it comes to the Windows platform so the VFX industry has slightly less leverage there than we do with Linux.
That said, the Working Group continues to discuss addressing Windows and Mac OS in the future and the information you provided to to us last month was both useful and very much appreciated.
My advice to anyone who feels that their platform needs are not being addressed by the VFX Reference Platform is to make sure you are letting us know at feedback@vfxplatform.com (which John has done in some style) and then also make sure the right people at the relevant software vendors know what you need.
Cheers,Nick

On 27 July 2015 at 22:14, John Burnett <studiosysadmins@johnburnett.com> wrote:
Hi Nick,

Shoot - double booked here with the BoF's meeting time (with Autodesk, coincidentally). My main question would be on the Windows side - is there any movement on nailing down better Visual Studio agreement across vendors?
On Sun, Jul 26, 2015 at 2:24 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
Thanks to the feedback received so far, the draft VFX Reference Platform for next year has been recently updated and you can see the latest athttp://vfxplatform.com. The main updates are around clarifying the gcc version and adding further information about the proposed Qt modifications.
There is still a couple of weeks left to give feedback before the final version is locked down in time for SIGGRAPH so if you have any further comments please send them along tofeedback@vfxplatform.com.
For those of you attending SIGGRAPH, please come along to the VFX Reference Platform BoF on the Wednesday at 12pm in room 503 to hear the latest news, talk directly with the working group of software vendors and discuss strategy for CY2017. There was talk on this list about wanting a more open dialog around this process so please do join us at the BoF if you are at SIGGRAPH.
Nick
On 1 June 2015 at 20:49, Nick Cannon <nick.cannon@gmail.com> wrote:
The Working Group behind the VFX Reference Platform, the target
platform for building software for VFX, has just announced the draft
of the Platform for next year, CY2016 at http://vfxplatform.com. This
represents the set of components that all major software releases are
expected to be built against from January 2016.

This Draft version is still a work in progress and all feedback will
be considered as we work towards the Final version of CY2016 which
will be published around SIGGRAPH. Please send feedback privately to
feedback@vfxplatform.com or publicly to the vfx-platform-discuss
Google Group.

The major change for CY2016 is a move to Qt 5 which requires a port of
PySide (already in progress) and modifications to vanilla Qt to
resolve issues that impact some vendors' tools. Our aim is for the
final modified Qt to be available to anyone who wishes to target the
CY2016 Platform, and for all major software vendors to build their
CY2016 releases against the same modified Qt. More details will be
shared in due course. For those attending SIGGRAPH, keep an eye out
for our Birds Of A Feather session.

Nick


To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


0 Plus One's     0 Comments  
   

Response from Anonymous @ July 27, 2015, 5:20 p.m.
Hi Nick,

Shoot - double booked here with the BoF's meeting time (with Autodesk, coincidentally). My main question would be on the Windows side - is there any movement on nailing down better Visual Studio agreement across vendors?
On Sun, Jul 26, 2015 at 2:24 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
Thanks to the feedback received so far, the draft VFX Reference Platform for next year has been recently updated and you can see the latest athttp://vfxplatform.com. The main updates are around clarifying the gcc version and adding further information about the proposed Qt modifications.
There is still a couple of weeks left to give feedback before the final version is locked down in time for SIGGRAPH so if you have any further comments please send them along tofeedback@vfxplatform.com.
For those of you attending SIGGRAPH, please come along to the VFX Reference Platform BoF on the Wednesday at 12pm in room 503 to hear the latest news, talk directly with the working group of software vendors and discuss strategy for CY2017. There was talk on this list about wanting a more open dialog around this process so please do join us at the BoF if you are at SIGGRAPH.
Nick
On 1 June 2015 at 20:49, Nick Cannon <nick.cannon@gmail.com> wrote:
The Working Group behind the VFX Reference Platform, the target
platform for building software for VFX, has just announced the draft
of the Platform for next year, CY2016 at http://vfxplatform.com. This
represents the set of components that all major software releases are
expected to be built against from January 2016.

This Draft version is still a work in progress and all feedback will
be considered as we work towards the Final version of CY2016 which
will be published around SIGGRAPH. Please send feedback privately to
feedback@vfxplatform.com or publicly to the vfx-platform-discuss
Google Group.

The major change for CY2016 is a move to Qt 5 which requires a port of
PySide (already in progress) and modifications to vanilla Qt to
resolve issues that impact some vendors' tools. Our aim is for the
final modified Qt to be available to anyone who wishes to target the
CY2016 Platform, and for all major software vendors to build their
CY2016 releases against the same modified Qt. More details will be
shared in due course. For those attending SIGGRAPH, keep an eye out
for our Birds Of A Feather session.

Nick


To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


0 Plus One's     0 Comments  
   

Response from Nick Cannon @ July 26, 2015, 5:30 p.m.
Thanks to the feedback received so far, the draft VFX Reference Platform for next year has been recently updated and you can see the latest athttp://vfxplatform.com. The main updates are around clarifying the gcc version and adding further information about the proposed Qt modifications.
There is still a couple of weeks left to give feedback before the final version is locked down in time for SIGGRAPH so if you have any further comments please send them along tofeedback@vfxplatform.com.
For those of you attending SIGGRAPH, please come along to the VFX Reference Platform BoF on the Wednesday at 12pm in room 503 to hear the latest news, talk directly with the working group of software vendors and discuss strategy for CY2017. There was talk on this list about wanting a more open dialog around this process so please do join us at the BoF if you are at SIGGRAPH.
Nick
On 1 June 2015 at 20:49, Nick Cannon <nick.cannon@gmail.com> wrote:
The Working Group behind the VFX Reference Platform, the target
platform for building software for VFX, has just announced the draft
of the Platform for next year, CY2016 at http://vfxplatform.com. This
represents the set of components that all major software releases are
expected to be built against from January 2016.

This Draft version is still a work in progress and all feedback will
be considered as we work towards the Final version of CY2016 which
will be published around SIGGRAPH. Please send feedback privately to
feedback@vfxplatform.com or publicly to the vfx-platform-discuss
Google Group.

The major change for CY2016 is a move to Qt 5 which requires a port of
PySide (already in progress) and modifications to vanilla Qt to
resolve issues that impact some vendors' tools. Our aim is for the
final modified Qt to be available to anyone who wishes to target the
CY2016 Platform, and for all major software vendors to build their
CY2016 releases against the same modified Qt. More details will be
shared in due course. For those attending SIGGRAPH, keep an eye out
for our Birds Of A Feather session.

Nick


0 Plus One's     0 Comments  
   

Response from Dan Young @ June 10, 2015, 11:50 a.m.
Remember, RV is an Autodesk product now ;) DY On 06/09/2015 10:53 PM, Jimmy Christensen wrote: > Just wanted to give my 2 cents about this one too :) > > Recently I have been doing a few tools inside RV which requires more > and more UI customizations, which means I have to do it in "mu" :( > Ofcourse I "could" compile PyQt/PySide but I would have to recompile > it every few releases and not to mention I have to support atleast OSX > and Linux. So I looked at the version numbers of Qt inside RV and > compared it to Maya and Nuke which already comes with PySide > pre-compiled and Nuke seemed to match. Tried to import the PySide from > nuke (forgot which version) did a few tests and lo and behold, it > actually worked. No crashes or missing calls. > > The Maya version did not work, but it makes sense given the info about > the custom Qt version. > > Still I have only tested this so far on Linux, but I'm sure I can > contribute this success to the intend from both Tweak and Foundry to > keep the versions somewhat the same, to again make our lives easier. > > -Jimmy > > On Thu, Jun 4, 2015 at 2:57 AM, Kevin Constantine > wrote: >> Wow, this conversation devolved quickly. I wasn't expecting that. >> >> I, for one, am happy to see vendors and the studios attempting to >> collaborate on a reference platform. Up until now, Redhat shipped RHEL with >> some version of Qt. Autodesk was shipping Maya with their own proprietary >> build of Qt, different from Redhat. Houdini was compiled against yet >> another version, and it maybe agreed with what Nuke expected (or maybe not). >> Then you add in the differences in python version, which tended to be a >> smaller problem, but it just added to the issue. >> >> Trying to build our own set of software that could run from within Maya, >> Houdini, and Nuke meant, in some cases, 3 separate builds. Now there's the >> hope that I can write some piece of software and be able to run it from >> within Maya, and Nuke without having to reconfigure my environment pathing >> and I think that's pretty sweet. >> >> Do I wish that I had seen a bunch of discussion when I followed that >> google-groups link? Absolutely. However, ultimately these are not >> open-source projects that are being discussed, so I'm not quite sure what to >> expect. You could look at this phase of the process as the "open-dialog" >> portion. A draft of the reference was decided on, and now there's a chance >> for comment by the public. Perhaps that's an ok compromise between the >> needs to keep some discussions private for whatever reason, and the need for >> public discourse. Where it becomes a problem is when the public dialog is >> ignored or usurped by the closed-door discussions. Because this process is >> so new, there hasn't been nearly enough time to prove one way or the other >> how contentious topics will be handled (if there even are contentious >> topics). Will the draft be ratified anyway despite studio-outcry? Who >> knows. Hopefully not, but time will tell. >> >> So, if you believe the process can't be taken seriously, then we're in the >> same boat we've been in for years where the vendors will ship you something >> that you can choose to use or not. But even still, thanks to the >> closed-door discussions, the applications might play a little nicer >> together. If you do believe in the process, and you have a strong opinion >> on why one library should be used instead of another, then there's an outlet >> to voice that opinion. That seems to me like it is at least a little bit >> better than it used to be. >> >> -kevin >> >> On Wed, Jun 3, 2015 at 2:47 PM, Dan Young wrote: >>> This is pretty funny, seeing as how you know a sum-total zero things about >>> me (and seemingly about most things in general) - you might have made the >>> most incorrect judgement of my personality I've ever heard. I am constantly >>> fighting with vendors, I'm practically the number one vendor punisher - >>> including on this list, as a VAR, as an in-shop engineer - I've had vendors >>> threaten to sue me, some more than once (avid list-readers can probably >>> figure out who...) based on the kind of things I've said to/about them. >>> >>> That being said, attacking the only people who have been willing to put >>> clout behind platforming is probably the worst possible way of getting the >>> thing that you want, you dummy. >>> >>> On 06/03/2015 02:29 PM, Nathan Dunsworth wrote: >>>> Actually people like yourself, with the I'll take anything on my knees >>>> with a smile opinion, are why a vendor can get away with whatever they like >>>> and not suffer any consequences. >>> >>> To unsubscribe from the list send a blank e-mail to >>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >> >> >> To unsubscribe from the list send a blank e-mail to >> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe > To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

0 Plus One's     0 Comments  
   

Response from Jimmy Christensen @ June 10, 2015, 1:55 a.m.
Just wanted to give my 2 cents about this one too :) Recently I have been doing a few tools inside RV which requires more and more UI customizations, which means I have to do it in "mu" :( Ofcourse I "could" compile PyQt/PySide but I would have to recompile it every few releases and not to mention I have to support atleast OSX and Linux. So I looked at the version numbers of Qt inside RV and compared it to Maya and Nuke which already comes with PySide pre-compiled and Nuke seemed to match. Tried to import the PySide from nuke (forgot which version) did a few tests and lo and behold, it actually worked. No crashes or missing calls. The Maya version did not work, but it makes sense given the info about the custom Qt version. Still I have only tested this so far on Linux, but I'm sure I can contribute this success to the intend from both Tweak and Foundry to keep the versions somewhat the same, to again make our lives easier. -Jimmy On Thu, Jun 4, 2015 at 2:57 AM, Kevin Constantine wrote: > Wow, this conversation devolved quickly. I wasn't expecting that. > > I, for one, am happy to see vendors and the studios attempting to > collaborate on a reference platform. Up until now, Redhat shipped RHEL with > some version of Qt. Autodesk was shipping Maya with their own proprietary > build of Qt, different from Redhat. Houdini was compiled against yet > another version, and it maybe agreed with what Nuke expected (or maybe not). > Then you add in the differences in python version, which tended to be a > smaller problem, but it just added to the issue. > > Trying to build our own set of software that could run from within Maya, > Houdini, and Nuke meant, in some cases, 3 separate builds. Now there's the > hope that I can write some piece of software and be able to run it from > within Maya, and Nuke without having to reconfigure my environment pathing > and I think that's pretty sweet. > > Do I wish that I had seen a bunch of discussion when I followed that > google-groups link? Absolutely. However, ultimately these are not > open-source projects that are being discussed, so I'm not quite sure what to > expect. You could look at this phase of the process as the "open-dialog" > portion. A draft of the reference was decided on, and now there's a chance > for comment by the public. Perhaps that's an ok compromise between the > needs to keep some discussions private for whatever reason, and the need for > public discourse. Where it becomes a problem is when the public dialog is > ignored or usurped by the closed-door discussions. Because this process is > so new, there hasn't been nearly enough time to prove one way or the other > how contentious topics will be handled (if there even are contentious > topics). Will the draft be ratified anyway despite studio-outcry? Who > knows. Hopefully not, but time will tell. > > So, if you believe the process can't be taken seriously, then we're in the > same boat we've been in for years where the vendors will ship you something > that you can choose to use or not. But even still, thanks to the > closed-door discussions, the applications might play a little nicer > together. If you do believe in the process, and you have a strong opinion > on why one library should be used instead of another, then there's an outlet > to voice that opinion. That seems to me like it is at least a little bit > better than it used to be. > > -kevin > > On Wed, Jun 3, 2015 at 2:47 PM, Dan Young wrote: >> >> This is pretty funny, seeing as how you know a sum-total zero things about >> me (and seemingly about most things in general) - you might have made the >> most incorrect judgement of my personality I've ever heard. I am constantly >> fighting with vendors, I'm practically the number one vendor punisher - >> including on this list, as a VAR, as an in-shop engineer - I've had vendors >> threaten to sue me, some more than once (avid list-readers can probably >> figure out who...) based on the kind of things I've said to/about them. >> >> That being said, attacking the only people who have been willing to put >> clout behind platforming is probably the worst possible way of getting the >> thing that you want, you dummy. >> >> On 06/03/2015 02:29 PM, Nathan Dunsworth wrote: >>> >>> Actually people like yourself, with the I'll take anything on my knees >>> with a smile opinion, are why a vendor can get away with whatever they like >>> and not suffer any consequences. >> >> >> To unsubscribe from the list send a blank e-mail to >> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe > > > > To unsubscribe from the list send a blank e-mail to > mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

0 Plus One's     0 Comments  
   

Response from Kevin Constantine @ June 3, 2015, 9 p.m.
Wow, this conversation devolved quickly. I wasn't expecting that.

I, for one, am happy to see vendors and the studios attempting to collaborate on a reference platform. Up until now, Redhat shipped RHEL with some version of Qt. Autodesk was shipping Maya with their own proprietary build of Qt, different from Redhat. Houdini was compiled against yet another version, and it maybe agreed with what Nuke expected (or maybe not). Then you add in the differences in python version, which tended to be a smaller problem, but it just added to the issue.

Trying to build our own set of software that could run from within Maya, Houdini, and Nuke meant, in some cases, 3 separate builds. Now there's the hope that I can write some piece of software and be able to run it from within Maya, and Nuke without having to reconfigure my environment pathing and I think that's pretty sweet.

Do I wish that I had seen a bunch of discussion when I followed that google-groups link? Absolutely. However, ultimately these are not open-source projects that are being discussed, so I'm not quite sure what to expect. You could look at this phase of the process as the "open-dialog" portion. A draft of the reference was decided on, and now there's a chance for comment by the public. Perhaps that's an ok compromise between the needs to keep some discussions private for whatever reason, and the need for public discourse. Where it becomes a problem is when the public dialog is ignored or usurped by the closed-door discussions. Because this process is so new, there hasn't been nearly enough time to prove one way or the other how contentious topics will be handled (if there even are contentious topics). Will the draft be ratified anyway despite studio-outcry? Who knows. Hopefully not, but time will tell.

So, if you believe the process can't be taken seriously, then we're in the same boat we've been in for years where the vendors will ship you something that you can choose to use or not. But even still, thanks to the closed-door discussions, the applications might play a little nicer together. If you do believe in the process, and you have a strong opinion on why one library should be used instead of another, then there's an outlet to voice that opinion. That seems to me like it is at least a little bit better than it used to be.

-kevin

On Wed, Jun 3, 2015 at 2:47 PM, Dan Young <dyoung@tippett.com> wrote:
This is pretty funny, seeing as how you know a sum-total zero things about me (and seemingly about most things in general) - you might have made the most incorrect judgement of my personality I've ever heard. I am constantly fighting with vendors, I'm practically the number one vendor punisher - including on this list, as a VAR, as an in-shop engineer - I've had vendors threaten to sue me, some more than once (avid list-readers can probably figure out who...) based on the kind of things I've said to/about them.

That being said, attacking the only people who have been willing to put clout behind platforming is probably the worst possible way of getting the thing that you want, you dummy.

On 06/03/2015 02:29 PM, Nathan Dunsworth wrote:
Actually people like yourself, with the I'll take anything on my knees with a smile opinion, are why a vendor can get away with whatever they like and not suffer any consequences.

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


0 Plus One's     0 Comments  
   

Response from Dan Young @ June 3, 2015, 5:50 p.m.
This is pretty funny, seeing as how you know a sum-total zero things about me (and seemingly about most things in general) - you might have made the most incorrect judgement of my personality I've ever heard. I am constantly fighting with vendors, I'm practically the number one vendor punisher - including on this list, as a VAR, as an in-shop engineer - I've had vendors threaten to sue me, some more than once (avid list-readers can probably figure out who...) based on the kind of things I've said to/about them. That being said, attacking the only people who have been willing to put clout behind platforming is probably the worst possible way of getting the thing that you want, you dummy. On 06/03/2015 02:29 PM, Nathan Dunsworth wrote: > Actually people like yourself, with the I'll take anything on my knees > with a smile opinion, are why a vendor can get away with whatever they > like and not suffer any consequences. To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

0 Plus One's     0 Comments  
   

Response from Anonymous @ June 3, 2015, 5:35 p.m.
Actually people like yourself, with the I'll take anything on my knees with a smile opinion, are why a vendor can get away with whatever they like and not suffer any consequences.

On Wed, Jun 3, 2015 at 10:41 AM, Dan Young <dyoung@tippett.com> wrote:
Like I said - you don't have a choice. You work in a VFX shop, you facilitate production. Your production demands usage of certain tools. Those tools, for better or for worse, are not made by you - and can drastically impact your shop's pipeline. You feel entitlement to something that you don't deserve. Getting any transparency from a software vendor is good, full stop.

On 06/02/2015 06:36 PM, Nathan Dunsworth wrote:
If you find people/companies communicating about standards "incredible" you obviously live by low standards accepting whatever is thrown at you as some form of miracle and a favor by the giving parties. Plenty of other people however live/work to higher standards and expect/demand better.

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


0 Plus One's     0 Comments  
   

Response from Anonymous @ June 3, 2015, 5:35 p.m.
"Before the VFX Reference Platform initiative existed none of this information was ever discussed or coordinated publicly in any way shape or form, especially before actual vendor software was shipped. This is the ultimately one of the reasons why some of these discussions have to remain private since it also involves componentreleases that werent made publickly available at the time."
"With help from the VES Tech Committee coordinating this effort we now at least have platform to discuss this openly."
Quotes 1 and 2 cancel each other out "remain private" and "discuss this openly".
It makes no logical sense how Autodesk or any other vendor openly discussing with a spec group the upgrade of AwesomeSauceLib for the next cycleequatesto somehowexposingsecret world domination plans. "Hey everyone, what do you think of upgrading AwesomeSauceLib with the reference spec?".
This is no different then Google and Intel discussing with a spec group which version of OpenCL should be the stable branch. It in know way exposes there internal private plans.
You even further back up the idea that being private is pointless by stating its just a reference not something people have to follow. If its more guidelines then actual rules absolutely no reason for it being closed doors. A guideline is a broad stroke and thus why would you need to discuss component specific internal details?
On Wed, Jun 3, 2015 at 10:39 AM, Daniel Tutino <daniel.tutino@autodesk.com> wrote:
Hi everyone,
I represent one of those software vendors and Im an active participant in the VFX Reference platform working group. It sounds like some people arent aware of the process that goes into the platform spec and hopefully through positive discussion we can encourage more people to openly participate and contribute their requirements. Please keep in mind, like Nick mentioned already, the CY2016 platform spec is also draft at this stage which could still be amended based on feedback. Likewise, this whole process is less than a year old with Siggraph 2015 marking its first official year in effect.
All decisions on which component versions are based on end-user feedback either through the VES Tech committee, the reference platform email alias, or directly to the software vendors. We dont make it a habit to just update components blindly without considering the pipeline impact. Weve made this mistake in the past for Maya when updating python ahead of everyone else and ultimately it caused major disruption for studios adopting that release. Avoiding major disruption in software upgrades, or versionitis as weve come to call it, is the key spirit of the reference platform.
For better or for worse this could mean freezing a major component for several iterations/releases for stability and predictability; just as studios do with software versions while in production. If not, we could already be pushing out major updates for multiple core components every year. No matter how large of an organization you are that could introduce a major amount of disruption and risk to any environment. We obviously dont want to stifle innovation or hurt adoption of newer technologies. It comes down to what is the added user value for the new component upgrade and when is this right time to adopt it. Which is another important part to emphasize in the reference platform, just because a component is listed it doesnt mean that every vendor must absolutely use the component in their product; it means which reference version should be used. Before the reference platform is drafted, all participating vendors agree together; on what commitments we plan to make for the coming year (again based on user feedback).
Before the VFX Reference Platform initiative existed none of this information was ever discussed or coordinated publicly in any way shape or form, especially before actual vendor software was shipped. This is the ultimately one of the reasons why some of these discussions have to remain private since it also involves component releases that werent made publickly available at the time. Users would turn to forum/lists such as this one after the fact to discusswhat was changed. With help from the VES Tech Committee coordinating this effort we now at least have platform to discuss this openly .
If anyone has any question related Maya adoption of the platform, Qt modification or anything else please feel to ask.
Thanks Daniel

From: <studiosysadmins-discuss-bounces@studiosysadmins.com> on behalf of Nathan Dunsworth
Reply-To: "studiosysadmins-discuss@studiosysadmins.com"
Date: Wednesday, June 3, 2015 at 7:44 AM
To: "studiosysadmins-discuss@studiosysadmins.com"
Subject: Re: [SSA-Discuss] VFX Reference Platform - CY2016 Draft published

Your reasoning makes the draft being closed even more invalid.
The reference platform is a recommended list of library versions for software to use. You don't have to openly state why you dont want the spec to upgrade OmgLeetSauceApi from 1.0 to 6.0. Commercial entities are great at spinning why they do things. You could easily just say "We don't think its mature enough" or some other complete bs lie instead of the real reason. <sarcasm> Your 300 million dollar company lacks resources </sarcasm>.
At the end of the day you aren't hiding anything. If Autodesk wants spec 3.0 to upgrade Qt to version 69 but you lack the ability to push Nuke with it, when the new specs released and keeps Qt at version 21 Autodesk can easily infer who didn't want to. Its secrecy for stupids sake.

BTW Who appointed the software vendors the decision makers of which version of libraries people/software should be using? That is in serious need of being flipped the other way around.
On Wed, Jun 3, 2015 at 4:46 AM, Matt Plec <mplec@mplec.com> wrote:
In my experience the confidentiality issues were not about technical details like how we fixed a bug in Qt or which version we would like to use next, but rather studio and vendor business issues that drive why someone does or doesn't want changes to elements of the platform in this cycle. These platform changes can make a lot of work, so it often comes down to availability of resources to do that work and the reasons those resources will or won't be available in the time frame required by a platform change -- often not something people want to share publicly.


On Wed, Jun 3, 2015 at 2:26 PM, Nathan Dunsworth <nathandunsworth@gmail.com> wrote:
I'm sorry but your first example given doesn't really make much sense. Especially when you look at the reason why Qt is used by software vendors.
One of the main reasons software vendors use Qt is because of the widget toolkit. Yes Qt is much more than that, damn near a full blown self contained OS, but the widget toolkit is often why people use it.
There is nothing secretive about choosing which widget toolkit your application will use. Further more any issues discovered with your toolkit are often pushed back upstream to the toolkit developer and there are plenty of ways to do this while maintaining confidentiality. This is no different then say a hardware company running embedded Linux and working with the Linux kernel devs.
I cant think of a single logical reason that Qt and confidential should ever enter the same sentence when discussing a reference spec. The only thing confidential with Qt and a software vendor is there license agreement. Like not exposing the Qt API etc. Even then it still does not formulate a valid excuse as the one being given.
If The Foundry or Autodesk want confidentiality when discussing which version of Qt they will support then they need to get out and look at other industries. Choosing your widget toolkit version is not some secret formula that should be locked up in Ft. Knox. Its like those insane VFX companies that patent there file system structure. Worth a good laugh but that is about it.
The process being closed off is flat out ridiculous. It's akin to all the Linux distros suddenly keeping quiet about which version of XOrg they will support in the future.
On Tue, Jun 2, 2015 at 9:47 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
Thanks for the feedback Nathan. The reason for it not being an
entirely open process is that this involves coordination between
software vendors with commercial interests, sharing confidential
information and dealing with sometimes complex legal issues that would
be challenging to discuss in the public domain.

In answer to your questions, the modifications to Qt are to avoid
similar regressions to those that result in software products shipping
today with differently modified versions of Qt 4. While not ideal,
it's a step forward to at least make sure the modified version of Qt 5
shipped with products is identical between them.

The decision to port PySide to Qt 5 is the result of customer
feedback. Many studios have a significant investment in PySide and
indicated that a move to Qt 5 would be challenging if it also required
a migration effort to PyQt.

Nick


On 2 June 2015 at 19:08, Nathan Dunsworth <nathandunsworth@gmail.com> wrote:
> The moment I hear the word "committee" I'm immediately turned off.
>
> I could never take something like the VFX Reference Platform seriously. A
> reference spec directed towards the VFX industry decided by a "committee" of
> oil baron software vendors and VES nerds in private emails.
>
> These decisions should be decided/discussed in the open by users/software
> engineers/technical directors/software vendors.
>
> This "committee" already wreaks to high heaven with the Qt talk. Just what
> sort of modifications need to be made to Qt? You are porting PySide, a dead
> project for all practical purposes, for Qt 5? Who made that decision?
>
> If this were an open spec I wouldn't need to wait for "more details will be
> shared in due course". I should be able to goto a mailing list and see all
> the related discussions even chime in if I so feel inclined. Instead I'm
> only provided a barren Google Group link with less than 10 discussions.
>
> On Tue, Jun 2, 2015 at 4:49 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
>>
>> Thanks Dan but this is really a group voluntary effort from
>> individuals representing software vendors and the VES Technology
>> Committee. What we produce is only as good as the feedback we receive
>> from the community so please keep it coming. While we can't guarantee
>> a reply to everyone I can assure you it all gets read and actively
>> discussed.
>>
>> Nick
>>
>>
>> On 2 June 2015 at 17:07, Dan Young <dyoung@tippett.com> wrote:
>> > You're doing God's work, sir.
>> >
>> > DY
>> >
>> >
>> > On 06/02/2015 01:25 PM, Nick Cannon wrote:
>> >>
>> >> The PySide port is in progress with development resource assigned to
>> >> the work. This is largely the result of the VFX Reference Platform
>> >> effort giving confidence that there was demand for a port from the VFX
>> >> industry and it would be adopted by multiple software vendors. It's
>> >> also thanks to some investment being made which the Platform working
>> >> group is very grateful for.
>> >>
>> >> I can't say much more about the Qt modifications at this point because
>> >> there are still details to be worked out. Expect more specific details
>> >> to be shared when we announce the Final version at SIGGRAPH. The
>> >> current state of affairs with Qt 4 is that some vendors ship their own
>> >> modified versions with their product and these differently modified
>> >> versions of Qt contribute to the versionitis problem. We were hoping
>> >> that Qt 5 could be used unmodified but that's not the case so
>> >> modifications are needed. It's possible those modifications could make
>> >> it into mainline Qt but there is no guarantee of that. What we are
>> >> working towards is a commitment that the software vendors will use the
>> >> same modified version of Qt 5. As for details of how this will work,
>> >> watch this space...
>> >>
>> >> Nick
>> >>
>> >>
>> >> On 2 June 2015 at 15:21, Will Rosecrans <wrosecrans@gmail.com> wrote:
>> >>>
>> >>> I am also curious about the Qt modifications. Is that something that
>> >>> would
>> >>> ultimately be pushed back to mainline Qt? Or is maintaining a
>> >>> separate
>> >>> forked version a long term plan?
>> >>>
>> >>> Also, who is doing the Qt5 port of PySide? Last I heard, the PySide
>> >>> project
>> >>> had mostly dried up and wasn't making a lot of forward progress with
>> >>> Qt5.
>> >>> Support theoretically being RealSoonNow for the past three years. Is
>> >>> the
>> >>> Working Group dedicated some resources to helping revive it, or are we
>> >>> just
>> >>> sort of hoping that a PySide rennaissance happens on its own in the
>> >>> next
>> >>> six
>> >>> months?
>> >>>
>> >>> Aside from that, it looks solid. Keeping glibc and gcc stable should
>> >>> keep
>> >>> updating pretty straightforward. Thanks for posting, Nick.
>> >>>
>> >>> On Tue, Jun 2, 2015 at 10:14 AM, Kevin Constantine
>> >>> <kevin.constantine@disneyanimation.com> wrote:
>> >>>>
>> >>>> Nick-
>> >>>>
>> >>>> Can you provide any more information on the 'modifications' to Qt? I
>> >>>> don't see anything talking through the changes on vfxplatform.com,
>> >>>> but
>> >>>> perhaps I missed it.
>> >>>>
>> >>>> Thanks!
>> >>>>
>> >>>> -kevin
>> >>>>
>> >>>> On Mon, Jun 1, 2015 at 12:49 PM, Nick Cannon <nick.cannon@gmail.com>
>> >>>> wrote:
>> >>>>>
>> >>>>> The Working Group behind the VFX Reference Platform, the target
>> >>>>> platform for building software for VFX, has just announced the draft
>> >>>>> of the Platform for next year, CY2016 at http://vfxplatform.com.
>> >>>>> This
>> >>>>> represents the set of components that all major software releases
>> >>>>> are
>> >>>>> expected to be built against from January 2016.
>> >>>>>
>> >>>>> This Draft version is still a work in progress and all feedback will
>> >>>>> be considered as we work towards the Final version of CY2016 which
>> >>>>> will be published around SIGGRAPH. Please send feedback privately to
>> >>>>> feedback@vfxplatform.com or publicly to the vfx-platform-discuss
>> >>>>> Google Group.
>> >>>>>
>> >>>>> The major change for CY2016 is a move to Qt 5 which requires a port
>> >>>>> of
>> >>>>> PySide (already in progress) and modifications to vanilla Qt to
>> >>>>> resolve issues that impact some vendors' tools. Our aim is for the
>> >>>>> final modified Qt to be available to anyone who wishes to target the
>> >>>>> CY2016 Platform, and for all major software vendors to build their
>> >>>>> CY2016 releases against the same modified Qt. More details will be
>> >>>>> shared in due course. For those attending SIGGRAPH, keep an eye out
>> >>>>> for our Birds Of A Feather session.
>> >>>>>
>> >>>>> Nick
>> >>>>> To unsubscribe from the list send a blank e-mail to
>> >>>>>
>> >>>>>
>> >>>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >>>>
>> >>>>
>> >>>>
>> >>>> To unsubscribe from the list send a blank e-mail to
>> >>>>
>> >>>>
>> >>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >>>
>> >>>
>> >>>
>> >>> To unsubscribe from the list send a blank e-mail to
>> >>>
>> >>>
>> >>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >>
>> >> To unsubscribe from the list send a blank e-mail to
>> >>
>> >> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >
>> >
>> > To unsubscribe from the list send a blank e-mail to
>> >
>> > mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> To unsubscribe from the list send a blank e-mail to
>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>
>
>
> To unsubscribe from the list send a blank e-mail to
> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


0 Plus One's     0 Comments  
   

Response from Dan Young @ June 3, 2015, 11:45 a.m.
Like I said - you don't have a choice. You work in a VFX shop, you facilitate production. Your production demands usage of certain tools. Those tools, for better or for worse, are not made by you - and can drastically impact your shop's pipeline. You feel entitlement to something that you don't deserve. Getting any transparency from a software vendor is good, full stop. On 06/02/2015 06:36 PM, Nathan Dunsworth wrote: > If you find people/companies communicating about standards > "incredible" you obviously live by low standards accepting whatever is > thrown at you as some form of miracle and a favor by the giving > parties. Plenty of other people however live/work to higher standards > and expect/demand better. To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

0 Plus One's     0 Comments  
   

Response from Anonymous @ June 3, 2015, 7:50 a.m.
Your reasoning makes the draft being closed even more invalid.
The reference platform is a recommended list of library versions for software to use. You don't have to openly state why you dont want the spec to upgrade OmgLeetSauceApi from 1.0 to 6.0. Commercial entities are great at spinning why they do things. You could easily just say "We don't think its mature enough" or some other complete bs lie instead of the real reason. <sarcasm> Your 300 million dollar company lacks resources </sarcasm>.
At the end of the day you aren't hiding anything. If Autodesk wants spec 3.0 to upgrade Qt to version 69 but you lack the ability to push Nuke with it, when the new specs released and keeps Qt at version 21 Autodesk can easily infer who didn't want to. Its secrecy for stupids sake.

BTW Who appointed the software vendors the decision makers of which version of libraries people/software should be using? That is in serious need of being flipped the other way around.
On Wed, Jun 3, 2015 at 4:46 AM, Matt Plec <mplec@mplec.com> wrote:
In my experience the confidentiality issues were not about technical details like how we fixed a bug in Qt or which version we would like to use next, but rather studio and vendor business issues that drive why someone does or doesn't want changes to elements of the platform in this cycle. These platform changes can make a lot of work, so it often comes down to availability of resources to do that work and the reasons those resources will or won't be available in the time frame required by a platform change -- often not something people want to share publicly.


On Wed, Jun 3, 2015 at 2:26 PM, Nathan Dunsworth <nathandunsworth@gmail.com> wrote:
I'm sorry but your first example given doesn't really make much sense. Especially when you look at the reason why Qt is used by software vendors.
One of the main reasons software vendors use Qt is because of the widget toolkit. Yes Qt is much more than that, damn near a full blown self contained OS, but the widget toolkit is often why people use it.
There is nothing secretive about choosing which widget toolkit your application will use. Further more any issues discovered with your toolkit are often pushed back upstream to the toolkit developer and there are plenty of ways to do this while maintaining confidentiality. This is no different then say a hardware company running embedded Linux and working with the Linux kernel devs.
I cant think of a single logical reason that Qt and confidential should ever enter the same sentence when discussing a reference spec. The only thing confidential with Qt and a software vendor is there license agreement. Like not exposing the Qt API etc. Even then it still does not formulate a valid excuse as the one being given.
If The Foundry or Autodesk want confidentiality when discussing which version of Qt they will support then they need to get out and look at other industries. Choosing your widget toolkit version is not some secret formula that should be locked up in Ft. Knox. Its like those insane VFX companies that patent there file system structure. Worth a good laugh but that is about it.
The process being closed off is flat out ridiculous. It's akin to all the Linux distros suddenly keeping quiet about which version of XOrg they will support in the future.
On Tue, Jun 2, 2015 at 9:47 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
Thanks for the feedback Nathan. The reason for it not being an
entirely open process is that this involves coordination between
software vendors with commercial interests, sharing confidential
information and dealing with sometimes complex legal issues that would
be challenging to discuss in the public domain.

In answer to your questions, the modifications to Qt are to avoid
similar regressions to those that result in software products shipping
today with differently modified versions of Qt 4. While not ideal,
it's a step forward to at least make sure the modified version of Qt 5
shipped with products is identical between them.

The decision to port PySide to Qt 5 is the result of customer
feedback. Many studios have a significant investment in PySide and
indicated that a move to Qt 5 would be challenging if it also required
a migration effort to PyQt.

Nick


On 2 June 2015 at 19:08, Nathan Dunsworth <nathandunsworth@gmail.com> wrote:
> The moment I hear the word "committee" I'm immediately turned off.
>
> I could never take something like the VFX Reference Platform seriously. A
> reference spec directed towards the VFX industry decided by a "committee" of
> oil baron software vendors and VES nerds in private emails.
>
> These decisions should be decided/discussed in the open by users/software
> engineers/technical directors/software vendors.
>
> This "committee" already wreaks to high heaven with the Qt talk. Just what
> sort of modifications need to be made to Qt? You are porting PySide, a dead
> project for all practical purposes, for Qt 5? Who made that decision?
>
> If this were an open spec I wouldn't need to wait for "more details will be
> shared in due course". I should be able to goto a mailing list and see all
> the related discussions even chime in if I so feel inclined. Instead I'm
> only provided a barren Google Group link with less than 10 discussions.
>
> On Tue, Jun 2, 2015 at 4:49 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
>>
>> Thanks Dan but this is really a group voluntary effort from
>> individuals representing software vendors and the VES Technology
>> Committee. What we produce is only as good as the feedback we receive
>> from the community so please keep it coming. While we can't guarantee
>> a reply to everyone I can assure you it all gets read and actively
>> discussed.
>>
>> Nick
>>
>>
>> On 2 June 2015 at 17:07, Dan Young <dyoung@tippett.com> wrote:
>> > You're doing God's work, sir.
>> >
>> > DY
>> >
>> >
>> > On 06/02/2015 01:25 PM, Nick Cannon wrote:
>> >>
>> >> The PySide port is in progress with development resource assigned to
>> >> the work. This is largely the result of the VFX Reference Platform
>> >> effort giving confidence that there was demand for a port from the VFX
>> >> industry and it would be adopted by multiple software vendors. It's
>> >> also thanks to some investment being made which the Platform working
>> >> group is very grateful for.
>> >>
>> >> I can't say much more about the Qt modifications at this point because
>> >> there are still details to be worked out. Expect more specific details
>> >> to be shared when we announce the Final version at SIGGRAPH. The
>> >> current state of affairs with Qt 4 is that some vendors ship their own
>> >> modified versions with their product and these differently modified
>> >> versions of Qt contribute to the versionitis problem. We were hoping
>> >> that Qt 5 could be used unmodified but that's not the case so
>> >> modifications are needed. It's possible those modifications could make
>> >> it into mainline Qt but there is no guarantee of that. What we are
>> >> working towards is a commitment that the software vendors will use the
>> >> same modified version of Qt 5. As for details of how this will work,
>> >> watch this space...
>> >>
>> >> Nick
>> >>
>> >>
>> >> On 2 June 2015 at 15:21, Will Rosecrans <wrosecrans@gmail.com> wrote:
>> >>>
>> >>> I am also curious about the Qt modifications. Is that something that
>> >>> would
>> >>> ultimately be pushed back to mainline Qt? Or is maintaining a
>> >>> separate
>> >>> forked version a long term plan?
>> >>>
>> >>> Also, who is doing the Qt5 port of PySide? Last I heard, the PySide
>> >>> project
>> >>> had mostly dried up and wasn't making a lot of forward progress with
>> >>> Qt5.
>> >>> Support theoretically being RealSoonNow for the past three years. Is
>> >>> the
>> >>> Working Group dedicated some resources to helping revive it, or are we
>> >>> just
>> >>> sort of hoping that a PySide rennaissance happens on its own in the
>> >>> next
>> >>> six
>> >>> months?
>> >>>
>> >>> Aside from that, it looks solid. Keeping glibc and gcc stable should
>> >>> keep
>> >>> updating pretty straightforward. Thanks for posting, Nick.
>> >>>
>> >>> On Tue, Jun 2, 2015 at 10:14 AM, Kevin Constantine
>> >>> <kevin.constantine@disneyanimation.com> wrote:
>> >>>>
>> >>>> Nick-
>> >>>>
>> >>>> Can you provide any more information on the 'modifications' to Qt? I
>> >>>> don't see anything talking through the changes on vfxplatform.com,
>> >>>> but
>> >>>> perhaps I missed it.
>> >>>>
>> >>>> Thanks!
>> >>>>
>> >>>> -kevin
>> >>>>
>> >>>> On Mon, Jun 1, 2015 at 12:49 PM, Nick Cannon <nick.cannon@gmail.com>
>> >>>> wrote:
>> >>>>>
>> >>>>> The Working Group behind the VFX Reference Platform, the target
>> >>>>> platform for building software for VFX, has just announced the draft
>> >>>>> of the Platform for next year, CY2016 at http://vfxplatform.com.
>> >>>>> This
>> >>>>> represents the set of components that all major software releases
>> >>>>> are
>> >>>>> expected to be built against from January 2016.
>> >>>>>
>> >>>>> This Draft version is still a work in progress and all feedback will
>> >>>>> be considered as we work towards the Final version of CY2016 which
>> >>>>> will be published around SIGGRAPH. Please send feedback privately to
>> >>>>> feedback@vfxplatform.com or publicly to the vfx-platform-discuss
>> >>>>> Google Group.
>> >>>>>
>> >>>>> The major change for CY2016 is a move to Qt 5 which requires a port
>> >>>>> of
>> >>>>> PySide (already in progress) and modifications to vanilla Qt to
>> >>>>> resolve issues that impact some vendors' tools. Our aim is for the
>> >>>>> final modified Qt to be available to anyone who wishes to target the
>> >>>>> CY2016 Platform, and for all major software vendors to build their
>> >>>>> CY2016 releases against the same modified Qt. More details will be
>> >>>>> shared in due course. For those attending SIGGRAPH, keep an eye out
>> >>>>> for our Birds Of A Feather session.
>> >>>>>
>> >>>>> Nick
>> >>>>> To unsubscribe from the list send a blank e-mail to
>> >>>>>
>> >>>>>
>> >>>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >>>>
>> >>>>
>> >>>>
>> >>>> To unsubscribe from the list send a blank e-mail to
>> >>>>
>> >>>>
>> >>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >>>
>> >>>
>> >>>
>> >>> To unsubscribe from the list send a blank e-mail to
>> >>>
>> >>>
>> >>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >>
>> >> To unsubscribe from the list send a blank e-mail to
>> >>
>> >> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >
>> >
>> > To unsubscribe from the list send a blank e-mail to
>> >
>> > mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> To unsubscribe from the list send a blank e-mail to
>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>
>
>
> To unsubscribe from the list send a blank e-mail to
> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


0 Plus One's     0 Comments  
   

Response from Nick Cannon @ June 3, 2015, 7:25 a.m.
Hi Mark,
Yes, Autodesk is one of the vendors that identified modifications would still be needed to Qt 5 so those changes you point to are the kind of modifications you can expect. 
Nick



On 3 Jun 2015, at 01:10, Mark Streatfield <content@studiosysadmins.com> wrote:

Hi Nick,

Would it be fair to assume that at least some of the Qt modifications are based on those currently made by Autodesk for Maya?

For those interested, they publish a patch of the changes they make as part of the package they distribute.  For Maya 2016 this file can be found under support/opensource/Qt/adsk-qt486-patch.txt.  A download of the fully modified source is also available here: http://www.autodesk.com/company/legal-notices-trademarks/open-source-distribution.

Mark.

To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

0 Plus One's     0 Comments  
   

Response from Matt Plec @ June 3, 2015, 5:50 a.m.
In my experience the confidentiality issues were not about technical details like how we fixed a bug in Qt or which version we would like to use next, but rather studio and vendor business issues that drive why someone does or doesn't want changes to elements of the platform in this cycle. These platform changes can make a lot of work, so it often comes down to availability of resources to do that work and the reasons those resources will or won't be available in the time frame required by a platform change -- often not something people want to share publicly.


On Wed, Jun 3, 2015 at 2:26 PM, Nathan Dunsworth <nathandunsworth@gmail.com> wrote:
I'm sorry but your first example given doesn't really make much sense. Especially when you look at the reason why Qt is used by software vendors.
One of the main reasons software vendors use Qt is because of the widget toolkit. Yes Qt is much more than that, damn near a full blown self contained OS, but the widget toolkit is often why people use it.
There is nothing secretive about choosing which widget toolkit your application will use. Further more any issues discovered with your toolkit are often pushed back upstream to the toolkit developer and there are plenty of ways to do this while maintaining confidentiality. This is no different then say a hardware company running embedded Linux and working with the Linux kernel devs.
I cant think of a single logical reason that Qt and confidential should ever enter the same sentence when discussing a reference spec. The only thing confidential with Qt and a software vendor is there license agreement. Like not exposing the Qt API etc. Even then it still does not formulate a valid excuse as the one being given.
If The Foundry or Autodesk want confidentiality when discussing which version of Qt they will support then they need to get out and look at other industries. Choosing your widget toolkit version is not some secret formula that should be locked up in Ft. Knox. Its like those insane VFX companies that patent there file system structure. Worth a good laugh but that is about it.
The process being closed off is flat out ridiculous. It's akin to all the Linux distros suddenly keeping quiet about which version of XOrg they will support in the future.
On Tue, Jun 2, 2015 at 9:47 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
Thanks for the feedback Nathan. The reason for it not being an
entirely open process is that this involves coordination between
software vendors with commercial interests, sharing confidential
information and dealing with sometimes complex legal issues that would
be challenging to discuss in the public domain.

In answer to your questions, the modifications to Qt are to avoid
similar regressions to those that result in software products shipping
today with differently modified versions of Qt 4. While not ideal,
it's a step forward to at least make sure the modified version of Qt 5
shipped with products is identical between them.

The decision to port PySide to Qt 5 is the result of customer
feedback. Many studios have a significant investment in PySide and
indicated that a move to Qt 5 would be challenging if it also required
a migration effort to PyQt.

Nick


On 2 June 2015 at 19:08, Nathan Dunsworth <nathandunsworth@gmail.com> wrote:
> The moment I hear the word "committee" I'm immediately turned off.
>
> I could never take something like the VFX Reference Platform seriously. A
> reference spec directed towards the VFX industry decided by a "committee" of
> oil baron software vendors and VES nerds in private emails.
>
> These decisions should be decided/discussed in the open by users/software
> engineers/technical directors/software vendors.
>
> This "committee" already wreaks to high heaven with the Qt talk. Just what
> sort of modifications need to be made to Qt? You are porting PySide, a dead
> project for all practical purposes, for Qt 5? Who made that decision?
>
> If this were an open spec I wouldn't need to wait for "more details will be
> shared in due course". I should be able to goto a mailing list and see all
> the related discussions even chime in if I so feel inclined. Instead I'm
> only provided a barren Google Group link with less than 10 discussions.
>
> On Tue, Jun 2, 2015 at 4:49 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
>>
>> Thanks Dan but this is really a group voluntary effort from
>> individuals representing software vendors and the VES Technology
>> Committee. What we produce is only as good as the feedback we receive
>> from the community so please keep it coming. While we can't guarantee
>> a reply to everyone I can assure you it all gets read and actively
>> discussed.
>>
>> Nick
>>
>>
>> On 2 June 2015 at 17:07, Dan Young <dyoung@tippett.com> wrote:
>> > You're doing God's work, sir.
>> >
>> > DY
>> >
>> >
>> > On 06/02/2015 01:25 PM, Nick Cannon wrote:
>> >>
>> >> The PySide port is in progress with development resource assigned to
>> >> the work. This is largely the result of the VFX Reference Platform
>> >> effort giving confidence that there was demand for a port from the VFX
>> >> industry and it would be adopted by multiple software vendors. It's
>> >> also thanks to some investment being made which the Platform working
>> >> group is very grateful for.
>> >>
>> >> I can't say much more about the Qt modifications at this point because
>> >> there are still details to be worked out. Expect more specific details
>> >> to be shared when we announce the Final version at SIGGRAPH. The
>> >> current state of affairs with Qt 4 is that some vendors ship their own
>> >> modified versions with their product and these differently modified
>> >> versions of Qt contribute to the versionitis problem. We were hoping
>> >> that Qt 5 could be used unmodified but that's not the case so
>> >> modifications are needed. It's possible those modifications could make
>> >> it into mainline Qt but there is no guarantee of that. What we are
>> >> working towards is a commitment that the software vendors will use the
>> >> same modified version of Qt 5. As for details of how this will work,
>> >> watch this space...
>> >>
>> >> Nick
>> >>
>> >>
>> >> On 2 June 2015 at 15:21, Will Rosecrans <wrosecrans@gmail.com> wrote:
>> >>>
>> >>> I am also curious about the Qt modifications. Is that something that
>> >>> would
>> >>> ultimately be pushed back to mainline Qt? Or is maintaining a
>> >>> separate
>> >>> forked version a long term plan?
>> >>>
>> >>> Also, who is doing the Qt5 port of PySide? Last I heard, the PySide
>> >>> project
>> >>> had mostly dried up and wasn't making a lot of forward progress with
>> >>> Qt5.
>> >>> Support theoretically being RealSoonNow for the past three years. Is
>> >>> the
>> >>> Working Group dedicated some resources to helping revive it, or are we
>> >>> just
>> >>> sort of hoping that a PySide rennaissance happens on its own in the
>> >>> next
>> >>> six
>> >>> months?
>> >>>
>> >>> Aside from that, it looks solid. Keeping glibc and gcc stable should
>> >>> keep
>> >>> updating pretty straightforward. Thanks for posting, Nick.
>> >>>
>> >>> On Tue, Jun 2, 2015 at 10:14 AM, Kevin Constantine
>> >>> <kevin.constantine@disneyanimation.com> wrote:
>> >>>>
>> >>>> Nick-
>> >>>>
>> >>>> Can you provide any more information on the 'modifications' to Qt? I
>> >>>> don't see anything talking through the changes on vfxplatform.com,
>> >>>> but
>> >>>> perhaps I missed it.
>> >>>>
>> >>>> Thanks!
>> >>>>
>> >>>> -kevin
>> >>>>
>> >>>> On Mon, Jun 1, 2015 at 12:49 PM, Nick Cannon <nick.cannon@gmail.com>
>> >>>> wrote:
>> >>>>>
>> >>>>> The Working Group behind the VFX Reference Platform, the target
>> >>>>> platform for building software for VFX, has just announced the draft
>> >>>>> of the Platform for next year, CY2016 at http://vfxplatform.com.
>> >>>>> This
>> >>>>> represents the set of components that all major software releases
>> >>>>> are
>> >>>>> expected to be built against from January 2016.
>> >>>>>
>> >>>>> This Draft version is still a work in progress and all feedback will
>> >>>>> be considered as we work towards the Final version of CY2016 which
>> >>>>> will be published around SIGGRAPH. Please send feedback privately to
>> >>>>> feedback@vfxplatform.com or publicly to the vfx-platform-discuss
>> >>>>> Google Group.
>> >>>>>
>> >>>>> The major change for CY2016 is a move to Qt 5 which requires a port
>> >>>>> of
>> >>>>> PySide (already in progress) and modifications to vanilla Qt to
>> >>>>> resolve issues that impact some vendors' tools. Our aim is for the
>> >>>>> final modified Qt to be available to anyone who wishes to target the
>> >>>>> CY2016 Platform, and for all major software vendors to build their
>> >>>>> CY2016 releases against the same modified Qt. More details will be
>> >>>>> shared in due course. For those attending SIGGRAPH, keep an eye out
>> >>>>> for our Birds Of A Feather session.
>> >>>>>
>> >>>>> Nick
>> >>>>> To unsubscribe from the list send a blank e-mail to
>> >>>>>
>> >>>>>
>> >>>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >>>>
>> >>>>
>> >>>>
>> >>>> To unsubscribe from the list send a blank e-mail to
>> >>>>
>> >>>>
>> >>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >>>
>> >>>
>> >>>
>> >>> To unsubscribe from the list send a blank e-mail to
>> >>>
>> >>>
>> >>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >>
>> >> To unsubscribe from the list send a blank e-mail to
>> >>
>> >> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >
>> >
>> > To unsubscribe from the list send a blank e-mail to
>> >
>> > mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> To unsubscribe from the list send a blank e-mail to
>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>
>
>
> To unsubscribe from the list send a blank e-mail to
> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


0 Plus One's     0 Comments  
   

Response from Anonymous @ June 3, 2015, 5 a.m.
I'm sorry but your first example given doesn't really make much sense. Especially when you look at the reason why Qt is used by software vendors.
One of the main reasons software vendors use Qt is because of the widget toolkit. Yes Qt is much more than that, damn near a full blown self contained OS, but the widget toolkit is often why people use it.
There is nothing secretive about choosing which widget toolkit your application will use. Further more any issues discovered with your toolkit are often pushed back upstream to the toolkit developer and there are plenty of ways to do this while maintaining confidentiality. This is no different then say a hardware company running embedded Linux and working with the Linux kernel devs.
I cant think of a single logical reason that Qt and confidential should ever enter the same sentence when discussing a reference spec. The only thing confidential with Qt and a software vendor is there license agreement. Like not exposing the Qt API etc. Even then it still does not formulate a valid excuse as the one being given.
If The Foundry or Autodesk want confidentiality when discussing which version of Qt they will support then they need to get out and look at other industries. Choosing your widget toolkit version is not some secret formula that should be locked up in Ft. Knox. Its like those insane VFX companies that patent there file system structure. Worth a good laugh but that is about it.
The process being closed off is flat out ridiculous. It's akin to all the Linux distros suddenly keeping quiet about which version of XOrg they will support in the future.
On Tue, Jun 2, 2015 at 9:47 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
Thanks for the feedback Nathan. The reason for it not being an
entirely open process is that this involves coordination between
software vendors with commercial interests, sharing confidential
information and dealing with sometimes complex legal issues that would
be challenging to discuss in the public domain.

In answer to your questions, the modifications to Qt are to avoid
similar regressions to those that result in software products shipping
today with differently modified versions of Qt 4. While not ideal,
it's a step forward to at least make sure the modified version of Qt 5
shipped with products is identical between them.

The decision to port PySide to Qt 5 is the result of customer
feedback. Many studios have a significant investment in PySide and
indicated that a move to Qt 5 would be challenging if it also required
a migration effort to PyQt.

Nick


On 2 June 2015 at 19:08, Nathan Dunsworth <nathandunsworth@gmail.com> wrote:
> The moment I hear the word "committee" I'm immediately turned off.
>
> I could never take something like the VFX Reference Platform seriously. A
> reference spec directed towards the VFX industry decided by a "committee" of
> oil baron software vendors and VES nerds in private emails.
>
> These decisions should be decided/discussed in the open by users/software
> engineers/technical directors/software vendors.
>
> This "committee" already wreaks to high heaven with the Qt talk. Just what
> sort of modifications need to be made to Qt? You are porting PySide, a dead
> project for all practical purposes, for Qt 5? Who made that decision?
>
> If this were an open spec I wouldn't need to wait for "more details will be
> shared in due course". I should be able to goto a mailing list and see all
> the related discussions even chime in if I so feel inclined. Instead I'm
> only provided a barren Google Group link with less than 10 discussions.
>
> On Tue, Jun 2, 2015 at 4:49 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
>>
>> Thanks Dan but this is really a group voluntary effort from
>> individuals representing software vendors and the VES Technology
>> Committee. What we produce is only as good as the feedback we receive
>> from the community so please keep it coming. While we can't guarantee
>> a reply to everyone I can assure you it all gets read and actively
>> discussed.
>>
>> Nick
>>
>>
>> On 2 June 2015 at 17:07, Dan Young <dyoung@tippett.com> wrote:
>> > You're doing God's work, sir.
>> >
>> > DY
>> >
>> >
>> > On 06/02/2015 01:25 PM, Nick Cannon wrote:
>> >>
>> >> The PySide port is in progress with development resource assigned to
>> >> the work. This is largely the result of the VFX Reference Platform
>> >> effort giving confidence that there was demand for a port from the VFX
>> >> industry and it would be adopted by multiple software vendors. It's
>> >> also thanks to some investment being made which the Platform working
>> >> group is very grateful for.
>> >>
>> >> I can't say much more about the Qt modifications at this point because
>> >> there are still details to be worked out. Expect more specific details
>> >> to be shared when we announce the Final version at SIGGRAPH. The
>> >> current state of affairs with Qt 4 is that some vendors ship their own
>> >> modified versions with their product and these differently modified
>> >> versions of Qt contribute to the versionitis problem. We were hoping
>> >> that Qt 5 could be used unmodified but that's not the case so
>> >> modifications are needed. It's possible those modifications could make
>> >> it into mainline Qt but there is no guarantee of that. What we are
>> >> working towards is a commitment that the software vendors will use the
>> >> same modified version of Qt 5. As for details of how this will work,
>> >> watch this space...
>> >>
>> >> Nick
>> >>
>> >>
>> >> On 2 June 2015 at 15:21, Will Rosecrans <wrosecrans@gmail.com> wrote:
>> >>>
>> >>> I am also curious about the Qt modifications. Is that something that
>> >>> would
>> >>> ultimately be pushed back to mainline Qt? Or is maintaining a
>> >>> separate
>> >>> forked version a long term plan?
>> >>>
>> >>> Also, who is doing the Qt5 port of PySide? Last I heard, the PySide
>> >>> project
>> >>> had mostly dried up and wasn't making a lot of forward progress with
>> >>> Qt5.
>> >>> Support theoretically being RealSoonNow for the past three years. Is
>> >>> the
>> >>> Working Group dedicated some resources to helping revive it, or are we
>> >>> just
>> >>> sort of hoping that a PySide rennaissance happens on its own in the
>> >>> next
>> >>> six
>> >>> months?
>> >>>
>> >>> Aside from that, it looks solid. Keeping glibc and gcc stable should
>> >>> keep
>> >>> updating pretty straightforward. Thanks for posting, Nick.
>> >>>
>> >>> On Tue, Jun 2, 2015 at 10:14 AM, Kevin Constantine
>> >>> <kevin.constantine@disneyanimation.com> wrote:
>> >>>>
>> >>>> Nick-
>> >>>>
>> >>>> Can you provide any more information on the 'modifications' to Qt? I
>> >>>> don't see anything talking through the changes on vfxplatform.com,
>> >>>> but
>> >>>> perhaps I missed it.
>> >>>>
>> >>>> Thanks!
>> >>>>
>> >>>> -kevin
>> >>>>
>> >>>> On Mon, Jun 1, 2015 at 12:49 PM, Nick Cannon <nick.cannon@gmail.com>
>> >>>> wrote:
>> >>>>>
>> >>>>> The Working Group behind the VFX Reference Platform, the target
>> >>>>> platform for building software for VFX, has just announced the draft
>> >>>>> of the Platform for next year, CY2016 at http://vfxplatform.com.
>> >>>>> This
>> >>>>> represents the set of components that all major software releases
>> >>>>> are
>> >>>>> expected to be built against from January 2016.
>> >>>>>
>> >>>>> This Draft version is still a work in progress and all feedback will
>> >>>>> be considered as we work towards the Final version of CY2016 which
>> >>>>> will be published around SIGGRAPH. Please send feedback privately to
>> >>>>> feedback@vfxplatform.com or publicly to the vfx-platform-discuss
>> >>>>> Google Group.
>> >>>>>
>> >>>>> The major change for CY2016 is a move to Qt 5 which requires a port
>> >>>>> of
>> >>>>> PySide (already in progress) and modifications to vanilla Qt to
>> >>>>> resolve issues that impact some vendors' tools. Our aim is for the
>> >>>>> final modified Qt to be available to anyone who wishes to target the
>> >>>>> CY2016 Platform, and for all major software vendors to build their
>> >>>>> CY2016 releases against the same modified Qt. More details will be
>> >>>>> shared in due course. For those attending SIGGRAPH, keep an eye out
>> >>>>> for our Birds Of A Feather session.
>> >>>>>
>> >>>>> Nick
>> >>>>> To unsubscribe from the list send a blank e-mail to
>> >>>>>
>> >>>>>
>> >>>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >>>>
>> >>>>
>> >>>>
>> >>>> To unsubscribe from the list send a blank e-mail to
>> >>>>
>> >>>>
>> >>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >>>
>> >>>
>> >>>
>> >>> To unsubscribe from the list send a blank e-mail to
>> >>>
>> >>>
>> >>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >>
>> >> To unsubscribe from the list send a blank e-mail to
>> >>
>> >> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> >
>> >
>> > To unsubscribe from the list send a blank e-mail to
>> >
>> > mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>> To unsubscribe from the list send a blank e-mail to
>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>
>
>
> To unsubscribe from the list send a blank e-mail to
> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


0 Plus One's     0 Comments  
   

Response from Mark Streatfield @ June 3, 2015, 1:10 a.m.

Hi Nick,

Would it be fair to assume that at least some of the Qt modifications are based on those currently made by Autodesk for Maya?

For those interested, they publish a patch of the changes they make as part of the package they distribute.  For Maya 2016 this file can be found under support/opensource/Qt/adsk-qt486-patch.txt.  A download of the fully modified source is also available here: http://www.autodesk.com/company/legal-notices-trademarks/open-source-distribution.

Mark.


0 Plus One's     0 Comments  
   

Response from Ben De Luca @ June 3, 2015, 12:50 a.m.
Hey Nick, Qt5 :( It would be good to see what sort of changes to QT people are talking about in a public repo, and see where the pyside port is going? On 3 June 2015 at 10:47, Nick Cannon wrote: > Thanks for the feedback Nathan. The reason for it not being an > entirely open process is that this involves coordination between > software vendors with commercial interests, sharing confidential > information and dealing with sometimes complex legal issues that would > be challenging to discuss in the public domain. > > In answer to your questions, the modifications to Qt are to avoid > similar regressions to those that result in software products shipping > today with differently modified versions of Qt 4. While not ideal, > it's a step forward to at least make sure the modified version of Qt 5 > shipped with products is identical between them. > > The decision to port PySide to Qt 5 is the result of customer > feedback. Many studios have a significant investment in PySide and > indicated that a move to Qt 5 would be challenging if it also required > a migration effort to PyQt. > > Nick > > > On 2 June 2015 at 19:08, Nathan Dunsworth wrote: >> The moment I hear the word "committee" I'm immediately turned off. >> >> I could never take something like the VFX Reference Platform seriously. A >> reference spec directed towards the VFX industry decided by a "committee" of >> oil baron software vendors and VES nerds in private emails. >> >> These decisions should be decided/discussed in the open by users/software >> engineers/technical directors/software vendors. >> >> This "committee" already wreaks to high heaven with the Qt talk. Just what >> sort of modifications need to be made to Qt? You are porting PySide, a dead >> project for all practical purposes, for Qt 5? Who made that decision? >> >> If this were an open spec I wouldn't need to wait for "more details will be >> shared in due course". I should be able to goto a mailing list and see all >> the related discussions even chime in if I so feel inclined. Instead I'm >> only provided a barren Google Group link with less than 10 discussions. >> >> On Tue, Jun 2, 2015 at 4:49 PM, Nick Cannon wrote: >>> >>> Thanks Dan but this is really a group voluntary effort from >>> individuals representing software vendors and the VES Technology >>> Committee. What we produce is only as good as the feedback we receive >>> from the community so please keep it coming. While we can't guarantee >>> a reply to everyone I can assure you it all gets read and actively >>> discussed. >>> >>> Nick >>> >>> >>> On 2 June 2015 at 17:07, Dan Young wrote: >>> > You're doing God's work, sir. >>> > >>> > DY >>> > >>> > >>> > On 06/02/2015 01:25 PM, Nick Cannon wrote: >>> >> >>> >> The PySide port is in progress with development resource assigned to >>> >> the work. This is largely the result of the VFX Reference Platform >>> >> effort giving confidence that there was demand for a port from the VFX >>> >> industry and it would be adopted by multiple software vendors. It's >>> >> also thanks to some investment being made which the Platform working >>> >> group is very grateful for. >>> >> >>> >> I can't say much more about the Qt modifications at this point because >>> >> there are still details to be worked out. Expect more specific details >>> >> to be shared when we announce the Final version at SIGGRAPH. The >>> >> current state of affairs with Qt 4 is that some vendors ship their own >>> >> modified versions with their product and these differently modified >>> >> versions of Qt contribute to the versionitis problem. We were hoping >>> >> that Qt 5 could be used unmodified but that's not the case so >>> >> modifications are needed. It's possible those modifications could make >>> >> it into mainline Qt but there is no guarantee of that. What we are >>> >> working towards is a commitment that the software vendors will use the >>> >> same modified version of Qt 5. As for details of how this will work, >>> >> watch this space... >>> >> >>> >> Nick >>> >> >>> >> >>> >> On 2 June 2015 at 15:21, Will Rosecrans wrote: >>> >>> >>> >>> I am also curious about the Qt modifications. Is that something that >>> >>> would >>> >>> ultimately be pushed back to mainline Qt? Or is maintaining a >>> >>> separate >>> >>> forked version a long term plan? >>> >>> >>> >>> Also, who is doing the Qt5 port of PySide? Last I heard, the PySide >>> >>> project >>> >>> had mostly dried up and wasn't making a lot of forward progress with >>> >>> Qt5. >>> >>> Support theoretically being RealSoonNow for the past three years. Is >>> >>> the >>> >>> Working Group dedicated some resources to helping revive it, or are we >>> >>> just >>> >>> sort of hoping that a PySide rennaissance happens on its own in the >>> >>> next >>> >>> six >>> >>> months? >>> >>> >>> >>> Aside from that, it looks solid. Keeping glibc and gcc stable should >>> >>> keep >>> >>> updating pretty straightforward. Thanks for posting, Nick. >>> >>> >>> >>> On Tue, Jun 2, 2015 at 10:14 AM, Kevin Constantine >>> >>> wrote: >>> >>>> >>> >>>> Nick- >>> >>>> >>> >>>> Can you provide any more information on the 'modifications' to Qt? I >>> >>>> don't see anything talking through the changes on vfxplatform.com, >>> >>>> but >>> >>>> perhaps I missed it. >>> >>>> >>> >>>> Thanks! >>> >>>> >>> >>>> -kevin >>> >>>> >>> >>>> On Mon, Jun 1, 2015 at 12:49 PM, Nick Cannon >>> >>>> wrote: >>> >>>>> >>> >>>>> The Working Group behind the VFX Reference Platform, the target >>> >>>>> platform for building software for VFX, has just announced the draft >>> >>>>> of the Platform for next year, CY2016 at http://vfxplatform.com. >>> >>>>> This >>> >>>>> represents the set of components that all major software releases >>> >>>>> are >>> >>>>> expected to be built against from January 2016. >>> >>>>> >>> >>>>> This Draft version is still a work in progress and all feedback will >>> >>>>> be considered as we work towards the Final version of CY2016 which >>> >>>>> will be published around SIGGRAPH. Please send feedback privately to >>> >>>>> feedback@vfxplatform.com or publicly to the vfx-platform-discuss >>> >>>>> Google Group. >>> >>>>> >>> >>>>> The major change for CY2016 is a move to Qt 5 which requires a port >>> >>>>> of >>> >>>>> PySide (already in progress) and modifications to vanilla Qt to >>> >>>>> resolve issues that impact some vendors' tools. Our aim is for the >>> >>>>> final modified Qt to be available to anyone who wishes to target the >>> >>>>> CY2016 Platform, and for all major software vendors to build their >>> >>>>> CY2016 releases against the same modified Qt. More details will be >>> >>>>> shared in due course. For those attending SIGGRAPH, keep an eye out >>> >>>>> for our Birds Of A Feather session. >>> >>>>> >>> >>>>> Nick >>> >>>>> To unsubscribe from the list send a blank e-mail to >>> >>>>> >>> >>>>> >>> >>>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >>> >>>> >>> >>>> >>> >>>> >>> >>>> To unsubscribe from the list send a blank e-mail to >>> >>>> >>> >>>> >>> >>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >>> >>> >>> >>> >>> >>> >>> >>> To unsubscribe from the list send a blank e-mail to >>> >>> >>> >>> >>> >>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >>> >> >>> >> To unsubscribe from the list send a blank e-mail to >>> >> >>> >> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >>> > >>> > >>> > To unsubscribe from the list send a blank e-mail to >>> > >>> > mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >>> To unsubscribe from the list send a blank e-mail to >>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >> >> >> >> To unsubscribe from the list send a blank e-mail to >> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe > To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

0 Plus One's     0 Comments  
   

Response from Nick Cannon @ June 2, 2015, 10:50 p.m.
Thanks for the feedback Nathan. The reason for it not being an entirely open process is that this involves coordination between software vendors with commercial interests, sharing confidential information and dealing with sometimes complex legal issues that would be challenging to discuss in the public domain. In answer to your questions, the modifications to Qt are to avoid similar regressions to those that result in software products shipping today with differently modified versions of Qt 4. While not ideal, it's a step forward to at least make sure the modified version of Qt 5 shipped with products is identical between them. The decision to port PySide to Qt 5 is the result of customer feedback. Many studios have a significant investment in PySide and indicated that a move to Qt 5 would be challenging if it also required a migration effort to PyQt. Nick On 2 June 2015 at 19:08, Nathan Dunsworth wrote: > The moment I hear the word "committee" I'm immediately turned off. > > I could never take something like the VFX Reference Platform seriously. A > reference spec directed towards the VFX industry decided by a "committee" of > oil baron software vendors and VES nerds in private emails. > > These decisions should be decided/discussed in the open by users/software > engineers/technical directors/software vendors. > > This "committee" already wreaks to high heaven with the Qt talk. Just what > sort of modifications need to be made to Qt? You are porting PySide, a dead > project for all practical purposes, for Qt 5? Who made that decision? > > If this were an open spec I wouldn't need to wait for "more details will be > shared in due course". I should be able to goto a mailing list and see all > the related discussions even chime in if I so feel inclined. Instead I'm > only provided a barren Google Group link with less than 10 discussions. > > On Tue, Jun 2, 2015 at 4:49 PM, Nick Cannon wrote: >> >> Thanks Dan but this is really a group voluntary effort from >> individuals representing software vendors and the VES Technology >> Committee. What we produce is only as good as the feedback we receive >> from the community so please keep it coming. While we can't guarantee >> a reply to everyone I can assure you it all gets read and actively >> discussed. >> >> Nick >> >> >> On 2 June 2015 at 17:07, Dan Young wrote: >> > You're doing God's work, sir. >> > >> > DY >> > >> > >> > On 06/02/2015 01:25 PM, Nick Cannon wrote: >> >> >> >> The PySide port is in progress with development resource assigned to >> >> the work. This is largely the result of the VFX Reference Platform >> >> effort giving confidence that there was demand for a port from the VFX >> >> industry and it would be adopted by multiple software vendors. It's >> >> also thanks to some investment being made which the Platform working >> >> group is very grateful for. >> >> >> >> I can't say much more about the Qt modifications at this point because >> >> there are still details to be worked out. Expect more specific details >> >> to be shared when we announce the Final version at SIGGRAPH. The >> >> current state of affairs with Qt 4 is that some vendors ship their own >> >> modified versions with their product and these differently modified >> >> versions of Qt contribute to the versionitis problem. We were hoping >> >> that Qt 5 could be used unmodified but that's not the case so >> >> modifications are needed. It's possible those modifications could make >> >> it into mainline Qt but there is no guarantee of that. What we are >> >> working towards is a commitment that the software vendors will use the >> >> same modified version of Qt 5. As for details of how this will work, >> >> watch this space... >> >> >> >> Nick >> >> >> >> >> >> On 2 June 2015 at 15:21, Will Rosecrans wrote: >> >>> >> >>> I am also curious about the Qt modifications. Is that something that >> >>> would >> >>> ultimately be pushed back to mainline Qt? Or is maintaining a >> >>> separate >> >>> forked version a long term plan? >> >>> >> >>> Also, who is doing the Qt5 port of PySide? Last I heard, the PySide >> >>> project >> >>> had mostly dried up and wasn't making a lot of forward progress with >> >>> Qt5. >> >>> Support theoretically being RealSoonNow for the past three years. Is >> >>> the >> >>> Working Group dedicated some resources to helping revive it, or are we >> >>> just >> >>> sort of hoping that a PySide rennaissance happens on its own in the >> >>> next >> >>> six >> >>> months? >> >>> >> >>> Aside from that, it looks solid. Keeping glibc and gcc stable should >> >>> keep >> >>> updating pretty straightforward. Thanks for posting, Nick. >> >>> >> >>> On Tue, Jun 2, 2015 at 10:14 AM, Kevin Constantine >> >>> wrote: >> >>>> >> >>>> Nick- >> >>>> >> >>>> Can you provide any more information on the 'modifications' to Qt? I >> >>>> don't see anything talking through the changes on vfxplatform.com, >> >>>> but >> >>>> perhaps I missed it. >> >>>> >> >>>> Thanks! >> >>>> >> >>>> -kevin >> >>>> >> >>>> On Mon, Jun 1, 2015 at 12:49 PM, Nick Cannon >> >>>> wrote: >> >>>>> >> >>>>> The Working Group behind the VFX Reference Platform, the target >> >>>>> platform for building software for VFX, has just announced the draft >> >>>>> of the Platform for next year, CY2016 at http://vfxplatform.com. >> >>>>> This >> >>>>> represents the set of components that all major software releases >> >>>>> are >> >>>>> expected to be built against from January 2016. >> >>>>> >> >>>>> This Draft version is still a work in progress and all feedback will >> >>>>> be considered as we work towards the Final version of CY2016 which >> >>>>> will be published around SIGGRAPH. Please send feedback privately to >> >>>>> feedback@vfxplatform.com or publicly to the vfx-platform-discuss >> >>>>> Google Group. >> >>>>> >> >>>>> The major change for CY2016 is a move to Qt 5 which requires a port >> >>>>> of >> >>>>> PySide (already in progress) and modifications to vanilla Qt to >> >>>>> resolve issues that impact some vendors' tools. Our aim is for the >> >>>>> final modified Qt to be available to anyone who wishes to target the >> >>>>> CY2016 Platform, and for all major software vendors to build their >> >>>>> CY2016 releases against the same modified Qt. More details will be >> >>>>> shared in due course. For those attending SIGGRAPH, keep an eye out >> >>>>> for our Birds Of A Feather session. >> >>>>> >> >>>>> Nick >> >>>>> To unsubscribe from the list send a blank e-mail to >> >>>>> >> >>>>> >> >>>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >> >>>> >> >>>> >> >>>> >> >>>> To unsubscribe from the list send a blank e-mail to >> >>>> >> >>>> >> >>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >> >>> >> >>> >> >>> >> >>> To unsubscribe from the list send a blank e-mail to >> >>> >> >>> >> >>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >> >> >> >> To unsubscribe from the list send a blank e-mail to >> >> >> >> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >> > >> > >> > To unsubscribe from the list send a blank e-mail to >> > >> > mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >> To unsubscribe from the list send a blank e-mail to >> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe > > > > To unsubscribe from the list send a blank e-mail to > mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

0 Plus One's     0 Comments  
   

Response from Anonymous @ June 2, 2015, 9:40 p.m.
1: "well, it's a pretty good thing that your opinion is fairly irrelevant, then"
LOL, thx for the good laugh!
2: "why do you feel entitled to be a part of such a discussion?"
And please tell me who you and anyone else are to decide who is and who isn't? Thats right nobody.
Everyone should have an opinion on a reference spec, even arrogant douchebags such as yourself.
3: "probably the manufacturers/vendors that utilize it to create and support their GUIs and viewports?"
Ah so what you are saying is you don't know the current state of PySide/PyQt and wanted to toss out some form of response, got it.
I keep up on it thanks.
4: "I'd suggest to you that you're lucky enough to have the ability to hear about this, rather than the alternative, which is nothing - and I would question again why you feel that you are entitled to participation in such discussions."

Are you serious? I mean like no seriously are you serious?
You're thankfulness for something that should happen as a default action, software developers following common standard practices, is vomit inducing to say the least.
If you find people/companies communicating about standards "incredible" you obviously live by low standards accepting whatever is thrown at you as some form of miracle and a favor by the giving parties. Plenty of other people however live/work to higher standards and expect/demand better.
The only "butthurt" here seems to come from your end.
On Tue, Jun 2, 2015 at 7:58 PM, Dan Young <dyoung@tippett.com> wrote:
I'm sorry - but, what?!

- "I could never take something like the VFX Reference Platform seriously" - well, it's a pretty good thing that your opinion is fairly irrelevant, then.
- "These decisions should be decided/discussed in the open by users/software engineers/technical directors/software vendors." - why do you feel entitled to be a part of such a discussion?
- "You are porting PySide, a dead project for all practical purposes, for Qt 5? Who made that decision?" - probably the manufacturers/vendors that utilize it to create and support their GUIs and viewports?
- "I should be able to goto a mailing list and see all the related discussions even chime in if I so feel inclined" - I'd suggest to you that you're lucky enough to have the ability to hear about this, rather than the alternative, which is nothing - and I would question again why you feel that you are entitled to participation in such discussions.

The point is, despite vendors being "oil barons" and the direction from the "industry at large" coming from the VESTC - the fact that they're even willing to work together on a voluntary basis to get us to a wonderfully symbiotic platform where we don't screw up softwares that we _have_ to use with dependencies that a particular software _has_ to use - is incredible. The fact that it _might_ come together to the point where things can be certified against things like QT versions, python versions in a unified way between multimillion (and billion) dollar companies - is incredible.

In short? Sit down and watch this play out before you start bitching about not being a part of it - you "wreak to high heaven" of butthurt animosity towards software vendors, which is OK - but it's not OK when you go on some funny powertrip where you think you're an authority on software and platforming and petulantly demand to be a part of it.

I look forward to seeing what this VFX reference platform can achieve/accomplish and I thank those who are volunteering to make this happen.

DY


On 06/02/2015 04:08 PM, Nathan Dunsworth wrote:
The moment I hear the word "committee" I'm immediately turned off.

I could never take something like the VFX Reference Platform seriously. A reference spec directed towards the VFX industry decided by a "committee" of oil baron software vendors and VES nerds in private emails.
These decisions should be decided/discussed in the open by users/software engineers/technical directors/software vendors.
This "committee" already wreaks to high heaven with the Qt talk. Just what sort of modifications need to be made to Qt? You are porting PySide, a dead project for all practical purposes, for Qt 5? Who made that decision?
If this were an open spec I wouldn't need to wait for "more details will be shared in due course". I should be able to goto a mailing list and see all the related discussions even chime in if I so feel inclined. Instead I'm only provided a barren Google Group link with less than 10 discussions.
On Tue, Jun 2, 2015 at 4:49 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
Thanks Dan but this is really a group voluntary effort from
individuals representing software vendors and the VES Technology
Committee. What we produce is only as good as the feedback we receive
from the community so please keep it coming. While we can't guarantee
a reply to everyone I can assure you it all gets read and actively
discussed.

Nick


On 2 June 2015 at 17:07, Dan Young <dyoung@tippett.com> wrote:
> You're doing God's work, sir.
>
> DY
>
>
> On 06/02/2015 01:25 PM, Nick Cannon wrote:
>>
>> The PySide port is in progress with development resource assigned to
>> the work. This is largely the result of the VFX Reference Platform
>> effort giving confidence that there was demand for a port from the VFX
>> industry and it would be adopted by multiple software vendors. It's
>> also thanks to some investment being made which the Platform working
>> group is very grateful for.
>>
>> I can't say much more about the Qt modifications at this point because
>> there are still details to be worked out. Expect more specific details
>> to be shared when we announce the Final version at SIGGRAPH. The
>> current state of affairs with Qt 4 is that some vendors ship their own
>> modified versions with their product and these differently modified
>> versions of Qt contribute to the versionitis problem. We were hoping
>> that Qt 5 could be used unmodified but that's not the case so
>> modifications are needed. It's possible those modifications could make
>> it into mainline Qt but there is no guarantee of that. What we are
>> working towards is a commitment that the software vendors will use the
>> same modified version of Qt 5. As for details of how this will work,
>> watch this space...
>>
>> Nick
>>
>>
>> On 2 June 2015 at 15:21, Will Rosecrans <wrosecrans@gmail.com> wrote:
>>>
>>> I am also curious about the Qt modifications. Is that something that
>>> would
>>> ultimately be pushed back to mainline Qt? Or is maintaining a separate
>>> forked version a long term plan?
>>>
>>> Also, who is doing the Qt5 port of PySide? Last I heard, the PySide
>>> project
>>> had mostly dried up and wasn't making a lot of forward progress with Qt5.
>>> Support theoretically being RealSoonNow for the past three years. Is the
>>> Working Group dedicated some resources to helping revive it, or are we
>>> just
>>> sort of hoping that a PySide rennaissance happens on its own in the next
>>> six
>>> months?
>>>
>>> Aside from that, it looks solid. Keeping glibc and gcc stable should
>>> keep
>>> updating pretty straightforward. Thanks for posting, Nick.
>>>
>>> On Tue, Jun 2, 2015 at 10:14 AM, Kevin Constantine
>>> <kevin.constantine@disneyanimation.com> wrote:
>>>>
>>>> Nick-
>>>>
>>>> Can you provide any more information on the 'modifications' to Qt? I
>>>> don't see anything talking through the changes on vfxplatform.com, but
>>>> perhaps I missed it.
>>>>
>>>> Thanks!
>>>>
>>>> -kevin
>>>>
>>>> On Mon, Jun 1, 2015 at 12:49 PM, Nick Cannon <nick.cannon@gmail.com>
>>>> wrote:
>>>>>
>>>>> The Working Group behind the VFX Reference Platform, the target
>>>>> platform for building software for VFX, has just announced the draft
>>>>> of the Platform for next year, CY2016 at http://vfxplatform.com. This
>>>>> represents the set of components that all major software releases are
>>>>> expected to be built against from January 2016.
>>>>>
>>>>> This Draft version is still a work in progress and all feedback will
>>>>> be considered as we work towards the Final version of CY2016 which
>>>>> will be published around SIGGRAPH. Please send feedback privately to
>>>>> feedback@vfxplatform.com or publicly to the vfx-platform-discuss
>>>>> Google Group.
>>>>>
>>>>> The major change for CY2016 is a move to Qt 5 which requires a port of
>>>>> PySide (already in progress) and modifications to vanilla Qt to
>>>>> resolve issues that impact some vendors' tools. Our aim is for the
>>>>> final modified Qt to be available to anyone who wishes to target the
>>>>> CY2016 Platform, and for all major software vendors to build their
>>>>> CY2016 releases against the same modified Qt. More details will be
>>>>> shared in due course. For those attending SIGGRAPH, keep an eye out
>>>>> for our Birds Of A Feather session.
>>>>>
>>>>> Nick
>>>>> To unsubscribe from the list send a blank e-mail to
>>>>>
>>>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>>>>
>>>>
>>>>
>>>> To unsubscribe from the list send a blank e-mail to
>>>>
>>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>>>
>>>
>>>
>>> To unsubscribe from the list send a blank e-mail to
>>>
>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>>
>> To unsubscribe from the list send a blank e-mail to
>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>
>
> To unsubscribe from the list send a blank e-mail to
> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe



To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


0 Plus One's     0 Comments  
   

Response from Dan Young @ June 2, 2015, 9 p.m.
I'm sorry - but, what?!

- "I could never take something like the VFX Reference Platform seriously" - well, it's a pretty good thing that your opinion is fairly irrelevant, then.
- "These decisions should be decided/discussed in the open by users/software engineers/technical directors/software vendors." - why do you feel entitled to be a part of such a discussion?
- "You are porting PySide, a dead project for all practical purposes, for Qt 5? Who made that decision?" - probably the manufacturers/vendors that utilize it to create and support their GUIs and viewports?
- "I should be able to goto a mailing list and see all the related discussions even chime in if I so feel inclined" - I'd suggest to you that you're lucky enough to have the ability to hear about this, rather than the alternative, which is nothing - and I would question again why you feel that you are entitled to participation in such discussions.

The point is, despite vendors being "oil barons" and the direction from the "industry at large" coming from the VESTC - the fact that they're even willing to work together on a voluntary basis to get us to a wonderfully symbiotic platform where we don't screw up softwares that we _have_ to use with dependencies that a particular software _has_ to use - is incredible. The fact that it _might_ come together to the point where things can be certified against things like QT versions, python versions in a unified way between multimillion (and billion) dollar companies - is incredible.

In short? Sit down and watch this play out before you start bitching about not being a part of it - you "wreak to high heaven" of butthurt animosity towards software vendors, which is OK - but it's not OK when you go on some funny powertrip where you think you're an authority on software and platforming and petulantly demand to be a part of it.

I look forward to seeing what this VFX reference platform can achieve/accomplish and I thank those who are volunteering to make this happen.

DY

On 06/02/2015 04:08 PM, Nathan Dunsworth wrote:
The moment I hear the word "committee" I'm immediately turned off.

I could never take something like the VFX Reference Platform seriously. A reference spec directed towards the VFX industry decided by a "committee" of oil baron software vendors and VES nerds in private emails.
These decisions should be decided/discussed in the open by users/software engineers/technical directors/software vendors.
This "committee" already wreaks to high heaven with the Qt talk. Just what sort of modifications need to be made to Qt? You are porting PySide, a dead project for all practical purposes, for Qt 5? Who made that decision?
If this were an open spec I wouldn't need to wait for "more details will be shared in due course". I should be able to goto a mailing list and see all the related discussions even chime in if I so feel inclined. Instead I'm only provided a barren Google Group link with less than 10 discussions.
On Tue, Jun 2, 2015 at 4:49 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
Thanks Dan but this is really a group voluntary effort from
individuals representing software vendors and the VES Technology
Committee. What we produce is only as good as the feedback we receive
from the community so please keep it coming. While we can't guarantee
a reply to everyone I can assure you it all gets read and actively
discussed.

Nick


On 2 June 2015 at 17:07, Dan Young <dyoung@tippett.com> wrote:
> You're doing God's work, sir.
>
> DY
>
>
> On 06/02/2015 01:25 PM, Nick Cannon wrote:
>>
>> The PySide port is in progress with development resource assigned to
>> the work. This is largely the result of the VFX Reference Platform
>> effort giving confidence that there was demand for a port from the VFX
>> industry and it would be adopted by multiple software vendors. It's
>> also thanks to some investment being made which the Platform working
>> group is very grateful for.
>>
>> I can't say much more about the Qt modifications at this point because
>> there are still details to be worked out. Expect more specific details
>> to be shared when we announce the Final version at SIGGRAPH. The
>> current state of affairs with Qt 4 is that some vendors ship their own
>> modified versions with their product and these differently modified
>> versions of Qt contribute to the versionitis problem. We were hoping
>> that Qt 5 could be used unmodified but that's not the case so
>> modifications are needed. It's possible those modifications could make
>> it into mainline Qt but there is no guarantee of that. What we are
>> working towards is a commitment that the software vendors will use the
>> same modified version of Qt 5. As for details of how this will work,
>> watch this space...
>>
>> Nick
>>
>>
>> On 2 June 2015 at 15:21, Will Rosecrans <wrosecrans@gmail.com> wrote:
>>>
>>> I am also curious about the Qt modifications. Is that something that
>>> would
>>> ultimately be pushed back to mainline Qt? Or is maintaining a separate
>>> forked version a long term plan?
>>>
>>> Also, who is doing the Qt5 port of PySide? Last I heard, the PySide
>>> project
>>> had mostly dried up and wasn't making a lot of forward progress with Qt5.
>>> Support theoretically being RealSoonNow for the past three years. Is the
>>> Working Group dedicated some resources to helping revive it, or are we
>>> just
>>> sort of hoping that a PySide rennaissance happens on its own in the next
>>> six
>>> months?
>>>
>>> Aside from that, it looks solid. Keeping glibc and gcc stable should
>>> keep
>>> updating pretty straightforward. Thanks for posting, Nick.
>>>
>>> On Tue, Jun 2, 2015 at 10:14 AM, Kevin Constantine
>>> <kevin.constantine@disneyanimation.com> wrote:
>>>>
>>>> Nick-
>>>>
>>>> Can you provide any more information on the 'modifications' to Qt? I
>>>> don't see anything talking through the changes on vfxplatform.com, but
>>>> perhaps I missed it.
>>>>
>>>> Thanks!
>>>>
>>>> -kevin
>>>>
>>>> On Mon, Jun 1, 2015 at 12:49 PM, Nick Cannon <nick.cannon@gmail.com>
>>>> wrote:
>>>>>
>>>>> The Working Group behind the VFX Reference Platform, the target
>>>>> platform for building software for VFX, has just announced the draft
>>>>> of the Platform for next year, CY2016 at http://vfxplatform.com. This
>>>>> represents the set of components that all major software releases are
>>>>> expected to be built against from January 2016.
>>>>>
>>>>> This Draft version is still a work in progress and all feedback will
>>>>> be considered as we work towards the Final version of CY2016 which
>>>>> will be published around SIGGRAPH. Please send feedback privately to
>>>>> feedback@vfxplatform.com or publicly to the vfx-platform-discuss
>>>>> Google Group.
>>>>>
>>>>> The major change for CY2016 is a move to Qt 5 which requires a port of
>>>>> PySide (already in progress) and modifications to vanilla Qt to
>>>>> resolve issues that impact some vendors' tools. Our aim is for the
>>>>> final modified Qt to be available to anyone who wishes to target the
>>>>> CY2016 Platform, and for all major software vendors to build their
>>>>> CY2016 releases against the same modified Qt. More details will be
>>>>> shared in due course. For those attending SIGGRAPH, keep an eye out
>>>>> for our Birds Of A Feather session.
>>>>>
>>>>> Nick
>>>>> To unsubscribe from the list send a blank e-mail to
>>>>>
>>>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>>>>
>>>>
>>>>
>>>> To unsubscribe from the list send a blank e-mail to
>>>>
>>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>>>
>>>
>>>
>>> To unsubscribe from the list send a blank e-mail to
>>>
>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>>
>> To unsubscribe from the list send a blank e-mail to
>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>
>
> To unsubscribe from the list send a blank e-mail to
> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe



To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


0 Plus One's     0 Comments  
   

Response from Anonymous @ June 2, 2015, 7:10 p.m.
The moment I hear the word "committee" I'm immediately turned off.

I could never take something like the VFX Reference Platform seriously. A reference spec directed towards the VFX industry decided by a "committee" of oil baron software vendors and VES nerds in private emails.
These decisions should be decided/discussed in the open by users/software engineers/technical directors/software vendors.
This "committee" already wreaks to high heaven with the Qt talk. Just what sort of modifications need to be made to Qt? You are porting PySide, a dead project for all practical purposes, for Qt 5? Who made that decision?
If this were an open spec I wouldn't need to wait for "more details will be shared in due course". I should be able to goto a mailing list and see all the related discussions even chime in if I so feel inclined. Instead I'm only provided a barren Google Group link with less than 10 discussions.
On Tue, Jun 2, 2015 at 4:49 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
Thanks Dan but this is really a group voluntary effort from
individuals representing software vendors and the VES Technology
Committee. What we produce is only as good as the feedback we receive
from the community so please keep it coming. While we can't guarantee
a reply to everyone I can assure you it all gets read and actively
discussed.

Nick


On 2 June 2015 at 17:07, Dan Young <dyoung@tippett.com> wrote:
> You're doing God's work, sir.
>
> DY
>
>
> On 06/02/2015 01:25 PM, Nick Cannon wrote:
>>
>> The PySide port is in progress with development resource assigned to
>> the work. This is largely the result of the VFX Reference Platform
>> effort giving confidence that there was demand for a port from the VFX
>> industry and it would be adopted by multiple software vendors. It's
>> also thanks to some investment being made which the Platform working
>> group is very grateful for.
>>
>> I can't say much more about the Qt modifications at this point because
>> there are still details to be worked out. Expect more specific details
>> to be shared when we announce the Final version at SIGGRAPH. The
>> current state of affairs with Qt 4 is that some vendors ship their own
>> modified versions with their product and these differently modified
>> versions of Qt contribute to the versionitis problem. We were hoping
>> that Qt 5 could be used unmodified but that's not the case so
>> modifications are needed. It's possible those modifications could make
>> it into mainline Qt but there is no guarantee of that. What we are
>> working towards is a commitment that the software vendors will use the
>> same modified version of Qt 5. As for details of how this will work,
>> watch this space...
>>
>> Nick
>>
>>
>> On 2 June 2015 at 15:21, Will Rosecrans <wrosecrans@gmail.com> wrote:
>>>
>>> I am also curious about the Qt modifications. Is that something that
>>> would
>>> ultimately be pushed back to mainline Qt? Or is maintaining a separate
>>> forked version a long term plan?
>>>
>>> Also, who is doing the Qt5 port of PySide? Last I heard, the PySide
>>> project
>>> had mostly dried up and wasn't making a lot of forward progress with Qt5.
>>> Support theoretically being RealSoonNow for the past three years. Is the
>>> Working Group dedicated some resources to helping revive it, or are we
>>> just
>>> sort of hoping that a PySide rennaissance happens on its own in the next
>>> six
>>> months?
>>>
>>> Aside from that, it looks solid. Keeping glibc and gcc stable should
>>> keep
>>> updating pretty straightforward. Thanks for posting, Nick.
>>>
>>> On Tue, Jun 2, 2015 at 10:14 AM, Kevin Constantine
>>> <kevin.constantine@disneyanimation.com> wrote:
>>>>
>>>> Nick-
>>>>
>>>> Can you provide any more information on the 'modifications' to Qt? I
>>>> don't see anything talking through the changes on vfxplatform.com, but
>>>> perhaps I missed it.
>>>>
>>>> Thanks!
>>>>
>>>> -kevin
>>>>
>>>> On Mon, Jun 1, 2015 at 12:49 PM, Nick Cannon <nick.cannon@gmail.com>
>>>> wrote:
>>>>>
>>>>> The Working Group behind the VFX Reference Platform, the target
>>>>> platform for building software for VFX, has just announced the draft
>>>>> of the Platform for next year, CY2016 at http://vfxplatform.com. This
>>>>> represents the set of components that all major software releases are
>>>>> expected to be built against from January 2016.
>>>>>
>>>>> This Draft version is still a work in progress and all feedback will
>>>>> be considered as we work towards the Final version of CY2016 which
>>>>> will be published around SIGGRAPH. Please send feedback privately to
>>>>> feedback@vfxplatform.com or publicly to the vfx-platform-discuss
>>>>> Google Group.
>>>>>
>>>>> The major change for CY2016 is a move to Qt 5 which requires a port of
>>>>> PySide (already in progress) and modifications to vanilla Qt to
>>>>> resolve issues that impact some vendors' tools. Our aim is for the
>>>>> final modified Qt to be available to anyone who wishes to target the
>>>>> CY2016 Platform, and for all major software vendors to build their
>>>>> CY2016 releases against the same modified Qt. More details will be
>>>>> shared in due course. For those attending SIGGRAPH, keep an eye out
>>>>> for our Birds Of A Feather session.
>>>>>
>>>>> Nick
>>>>> To unsubscribe from the list send a blank e-mail to
>>>>>
>>>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>>>>
>>>>
>>>>
>>>> To unsubscribe from the list send a blank e-mail to
>>>>
>>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>>>
>>>
>>>
>>> To unsubscribe from the list send a blank e-mail to
>>>
>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>>
>> To unsubscribe from the list send a blank e-mail to
>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
>
>
> To unsubscribe from the list send a blank e-mail to
> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


0 Plus One's     0 Comments  
   

Response from Nick Cannon @ June 2, 2015, 5:50 p.m.
Thanks Dan but this is really a group voluntary effort from individuals representing software vendors and the VES Technology Committee. What we produce is only as good as the feedback we receive from the community so please keep it coming. While we can't guarantee a reply to everyone I can assure you it all gets read and actively discussed. Nick On 2 June 2015 at 17:07, Dan Young wrote: > You're doing God's work, sir. > > DY > > > On 06/02/2015 01:25 PM, Nick Cannon wrote: >> >> The PySide port is in progress with development resource assigned to >> the work. This is largely the result of the VFX Reference Platform >> effort giving confidence that there was demand for a port from the VFX >> industry and it would be adopted by multiple software vendors. It's >> also thanks to some investment being made which the Platform working >> group is very grateful for. >> >> I can't say much more about the Qt modifications at this point because >> there are still details to be worked out. Expect more specific details >> to be shared when we announce the Final version at SIGGRAPH. The >> current state of affairs with Qt 4 is that some vendors ship their own >> modified versions with their product and these differently modified >> versions of Qt contribute to the versionitis problem. We were hoping >> that Qt 5 could be used unmodified but that's not the case so >> modifications are needed. It's possible those modifications could make >> it into mainline Qt but there is no guarantee of that. What we are >> working towards is a commitment that the software vendors will use the >> same modified version of Qt 5. As for details of how this will work, >> watch this space... >> >> Nick >> >> >> On 2 June 2015 at 15:21, Will Rosecrans wrote: >>> >>> I am also curious about the Qt modifications. Is that something that >>> would >>> ultimately be pushed back to mainline Qt? Or is maintaining a separate >>> forked version a long term plan? >>> >>> Also, who is doing the Qt5 port of PySide? Last I heard, the PySide >>> project >>> had mostly dried up and wasn't making a lot of forward progress with Qt5. >>> Support theoretically being RealSoonNow for the past three years. Is the >>> Working Group dedicated some resources to helping revive it, or are we >>> just >>> sort of hoping that a PySide rennaissance happens on its own in the next >>> six >>> months? >>> >>> Aside from that, it looks solid. Keeping glibc and gcc stable should >>> keep >>> updating pretty straightforward. Thanks for posting, Nick. >>> >>> On Tue, Jun 2, 2015 at 10:14 AM, Kevin Constantine >>> wrote: >>>> >>>> Nick- >>>> >>>> Can you provide any more information on the 'modifications' to Qt? I >>>> don't see anything talking through the changes on vfxplatform.com, but >>>> perhaps I missed it. >>>> >>>> Thanks! >>>> >>>> -kevin >>>> >>>> On Mon, Jun 1, 2015 at 12:49 PM, Nick Cannon >>>> wrote: >>>>> >>>>> The Working Group behind the VFX Reference Platform, the target >>>>> platform for building software for VFX, has just announced the draft >>>>> of the Platform for next year, CY2016 at http://vfxplatform.com. This >>>>> represents the set of components that all major software releases are >>>>> expected to be built against from January 2016. >>>>> >>>>> This Draft version is still a work in progress and all feedback will >>>>> be considered as we work towards the Final version of CY2016 which >>>>> will be published around SIGGRAPH. Please send feedback privately to >>>>> feedback@vfxplatform.com or publicly to the vfx-platform-discuss >>>>> Google Group. >>>>> >>>>> The major change for CY2016 is a move to Qt 5 which requires a port of >>>>> PySide (already in progress) and modifications to vanilla Qt to >>>>> resolve issues that impact some vendors' tools. Our aim is for the >>>>> final modified Qt to be available to anyone who wishes to target the >>>>> CY2016 Platform, and for all major software vendors to build their >>>>> CY2016 releases against the same modified Qt. More details will be >>>>> shared in due course. For those attending SIGGRAPH, keep an eye out >>>>> for our Birds Of A Feather session. >>>>> >>>>> Nick >>>>> To unsubscribe from the list send a blank e-mail to >>>>> >>>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >>>> >>>> >>>> >>>> To unsubscribe from the list send a blank e-mail to >>>> >>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >>> >>> >>> >>> To unsubscribe from the list send a blank e-mail to >>> >>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >> >> To unsubscribe from the list send a blank e-mail to >> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe > > > To unsubscribe from the list send a blank e-mail to > mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

0 Plus One's     0 Comments  
   

Response from Dan Young @ June 2, 2015, 5:10 p.m.
You're doing God's work, sir. DY On 06/02/2015 01:25 PM, Nick Cannon wrote: > The PySide port is in progress with development resource assigned to > the work. This is largely the result of the VFX Reference Platform > effort giving confidence that there was demand for a port from the VFX > industry and it would be adopted by multiple software vendors. It's > also thanks to some investment being made which the Platform working > group is very grateful for. > > I can't say much more about the Qt modifications at this point because > there are still details to be worked out. Expect more specific details > to be shared when we announce the Final version at SIGGRAPH. The > current state of affairs with Qt 4 is that some vendors ship their own > modified versions with their product and these differently modified > versions of Qt contribute to the versionitis problem. We were hoping > that Qt 5 could be used unmodified but that's not the case so > modifications are needed. It's possible those modifications could make > it into mainline Qt but there is no guarantee of that. What we are > working towards is a commitment that the software vendors will use the > same modified version of Qt 5. As for details of how this will work, > watch this space... > > Nick > > > On 2 June 2015 at 15:21, Will Rosecrans wrote: >> I am also curious about the Qt modifications. Is that something that would >> ultimately be pushed back to mainline Qt? Or is maintaining a separate >> forked version a long term plan? >> >> Also, who is doing the Qt5 port of PySide? Last I heard, the PySide project >> had mostly dried up and wasn't making a lot of forward progress with Qt5. >> Support theoretically being RealSoonNow for the past three years. Is the >> Working Group dedicated some resources to helping revive it, or are we just >> sort of hoping that a PySide rennaissance happens on its own in the next six >> months? >> >> Aside from that, it looks solid. Keeping glibc and gcc stable should keep >> updating pretty straightforward. Thanks for posting, Nick. >> >> On Tue, Jun 2, 2015 at 10:14 AM, Kevin Constantine >> wrote: >>> Nick- >>> >>> Can you provide any more information on the 'modifications' to Qt? I >>> don't see anything talking through the changes on vfxplatform.com, but >>> perhaps I missed it. >>> >>> Thanks! >>> >>> -kevin >>> >>> On Mon, Jun 1, 2015 at 12:49 PM, Nick Cannon >>> wrote: >>>> The Working Group behind the VFX Reference Platform, the target >>>> platform for building software for VFX, has just announced the draft >>>> of the Platform for next year, CY2016 at http://vfxplatform.com. This >>>> represents the set of components that all major software releases are >>>> expected to be built against from January 2016. >>>> >>>> This Draft version is still a work in progress and all feedback will >>>> be considered as we work towards the Final version of CY2016 which >>>> will be published around SIGGRAPH. Please send feedback privately to >>>> feedback@vfxplatform.com or publicly to the vfx-platform-discuss >>>> Google Group. >>>> >>>> The major change for CY2016 is a move to Qt 5 which requires a port of >>>> PySide (already in progress) and modifications to vanilla Qt to >>>> resolve issues that impact some vendors' tools. Our aim is for the >>>> final modified Qt to be available to anyone who wishes to target the >>>> CY2016 Platform, and for all major software vendors to build their >>>> CY2016 releases against the same modified Qt. More details will be >>>> shared in due course. For those attending SIGGRAPH, keep an eye out >>>> for our Birds Of A Feather session. >>>> >>>> Nick >>>> To unsubscribe from the list send a blank e-mail to >>>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >>> >>> >>> To unsubscribe from the list send a blank e-mail to >>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >> >> >> To unsubscribe from the list send a blank e-mail to >> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe > To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

0 Plus One's     0 Comments  
   

Response from Nick Cannon @ June 2, 2015, 4:30 p.m.
The PySide port is in progress with development resource assigned to the work. This is largely the result of the VFX Reference Platform effort giving confidence that there was demand for a port from the VFX industry and it would be adopted by multiple software vendors. It's also thanks to some investment being made which the Platform working group is very grateful for. I can't say much more about the Qt modifications at this point because there are still details to be worked out. Expect more specific details to be shared when we announce the Final version at SIGGRAPH. The current state of affairs with Qt 4 is that some vendors ship their own modified versions with their product and these differently modified versions of Qt contribute to the versionitis problem. We were hoping that Qt 5 could be used unmodified but that's not the case so modifications are needed. It's possible those modifications could make it into mainline Qt but there is no guarantee of that. What we are working towards is a commitment that the software vendors will use the same modified version of Qt 5. As for details of how this will work, watch this space... Nick On 2 June 2015 at 15:21, Will Rosecrans wrote: > I am also curious about the Qt modifications. Is that something that would > ultimately be pushed back to mainline Qt? Or is maintaining a separate > forked version a long term plan? > > Also, who is doing the Qt5 port of PySide? Last I heard, the PySide project > had mostly dried up and wasn't making a lot of forward progress with Qt5. > Support theoretically being RealSoonNow for the past three years. Is the > Working Group dedicated some resources to helping revive it, or are we just > sort of hoping that a PySide rennaissance happens on its own in the next six > months? > > Aside from that, it looks solid. Keeping glibc and gcc stable should keep > updating pretty straightforward. Thanks for posting, Nick. > > On Tue, Jun 2, 2015 at 10:14 AM, Kevin Constantine > wrote: >> >> Nick- >> >> Can you provide any more information on the 'modifications' to Qt? I >> don't see anything talking through the changes on vfxplatform.com, but >> perhaps I missed it. >> >> Thanks! >> >> -kevin >> >> On Mon, Jun 1, 2015 at 12:49 PM, Nick Cannon >> wrote: >>> >>> The Working Group behind the VFX Reference Platform, the target >>> platform for building software for VFX, has just announced the draft >>> of the Platform for next year, CY2016 at http://vfxplatform.com. This >>> represents the set of components that all major software releases are >>> expected to be built against from January 2016. >>> >>> This Draft version is still a work in progress and all feedback will >>> be considered as we work towards the Final version of CY2016 which >>> will be published around SIGGRAPH. Please send feedback privately to >>> feedback@vfxplatform.com or publicly to the vfx-platform-discuss >>> Google Group. >>> >>> The major change for CY2016 is a move to Qt 5 which requires a port of >>> PySide (already in progress) and modifications to vanilla Qt to >>> resolve issues that impact some vendors' tools. Our aim is for the >>> final modified Qt to be available to anyone who wishes to target the >>> CY2016 Platform, and for all major software vendors to build their >>> CY2016 releases against the same modified Qt. More details will be >>> shared in due course. For those attending SIGGRAPH, keep an eye out >>> for our Birds Of A Feather session. >>> >>> Nick >>> To unsubscribe from the list send a blank e-mail to >>> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe >> >> >> >> To unsubscribe from the list send a blank e-mail to >> mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe > > > > To unsubscribe from the list send a blank e-mail to > mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe

0 Plus One's     0 Comments  
   

Response from Will Rosecrans @ June 2, 2015, 3:25 p.m.
I am also curious about the Qt modifications. Is that something that would ultimately be pushed back to mainline Qt? Or is maintaining a separate forked version a long term plan?
Also, who is doing the Qt5 port of PySide? Last I heard, the PySide project had mostly dried up and wasn't making a lot of forward progress with Qt5. Support theoretically being RealSoonNow for the past three years. Is the Working Group dedicated some resources to helping revive it, or are we just sort of hoping that a PySide rennaissance happens on its own in the next six months?
Aside from that, it looks solid. Keeping glibc and gcc stable should keep updating pretty straightforward. Thanks for posting, Nick.
On Tue, Jun 2, 2015 at 10:14 AM, Kevin Constantine <kevin.constantine@disneyanimation.com> wrote:
Nick-

Can you provide any more information on the 'modifications' to Qt? I don't see anything talking through the changes on vfxplatform.com, but perhaps I missed it.

Thanks!

-kevin

On Mon, Jun 1, 2015 at 12:49 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
The Working Group behind the VFX Reference Platform, the target
platform for building software for VFX, has just announced the draft
of the Platform for next year, CY2016 at http://vfxplatform.com. This
represents the set of components that all major software releases are
expected to be built against from January 2016.

This Draft version is still a work in progress and all feedback will
be considered as we work towards the Final version of CY2016 which
will be published around SIGGRAPH. Please send feedback privately to
feedback@vfxplatform.com or publicly to the vfx-platform-discuss
Google Group.

The major change for CY2016 is a move to Qt 5 which requires a port of
PySide (already in progress) and modifications to vanilla Qt to
resolve issues that impact some vendors' tools. Our aim is for the
final modified Qt to be available to anyone who wishes to target the
CY2016 Platform, and for all major software vendors to build their
CY2016 releases against the same modified Qt. More details will be
shared in due course. For those attending SIGGRAPH, keep an eye out
for our Birds Of A Feather session.

Nick
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


0 Plus One's     0 Comments  
   

Response from Kevin Constantine @ June 2, 2015, 1:20 p.m.
Nick-

Can you provide any more information on the 'modifications' to Qt? I don't see anything talking through the changes on vfxplatform.com, but perhaps I missed it.

Thanks!

-kevin

On Mon, Jun 1, 2015 at 12:49 PM, Nick Cannon <nick.cannon@gmail.com> wrote:
The Working Group behind the VFX Reference Platform, the target
platform for building software for VFX, has just announced the draft
of the Platform for next year, CY2016 at http://vfxplatform.com. This
represents the set of components that all major software releases are
expected to be built against from January 2016.

This Draft version is still a work in progress and all feedback will
be considered as we work towards the Final version of CY2016 which
will be published around SIGGRAPH. Please send feedback privately to
feedback@vfxplatform.com or publicly to the vfx-platform-discuss
Google Group.

The major change for CY2016 is a move to Qt 5 which requires a port of
PySide (already in progress) and modifications to vanilla Qt to
resolve issues that impact some vendors' tools. Our aim is for the
final modified Qt to be available to anyone who wishes to target the
CY2016 Platform, and for all major software vendors to build their
CY2016 releases against the same modified Qt. More details will be
shared in due course. For those attending SIGGRAPH, keep an eye out
for our Birds Of A Feather session.

Nick
To unsubscribe from the list send a blank e-mail to mailto:studiosysadmins-discuss-request@studiosysadmins.com?subject=unsubscribe


0 Plus One's     0 Comments