[PySide] PySide Digest, Vol 14, Issue 19

Sergio Pulgarín serpulga at gmail.com
Thu Mar 21 03:31:45 CET 2013 

> Message: 1
> Date: Wed, 20 Mar 2013 15:48:19 +0000
> From: Henry Gomersall <heng at cantab.net>
> Subject: Re: [PySide] Honkin Idea for PySide's thriving
> To: Christian Tismer <tismer at stackless.com>
> Cc: pyside at qt-project.org, "Samual M. Rushing" <rushing at ironport.com>
> Message-ID: <1363794499.13681.16.camel at farnsworth>
> Content-Type: text/plain; charset="UTF-8"
>
> On Wed, 2013-03-20 at 15:04 +0000, Henry Gomersall wrote:
> > Given Ubuntu's recent noises about Qt and QML, in particular with
> > respect to Ubuntu Touch (http://en.wikipedia.org/wiki/Ubuntu_Touch),
> > would perhaps a better target be to derive some support from that
> > direction?
> >
> > To quote from the Ubuntu developers resources
> > (http://developer.ubuntu.com/resources/programming-languages/python/):
> >
> > "Many important parts of Ubuntu are already written directly in
> > Python,
> > and we work to make every important API and framework within Ubuntu
> > available from Python."
>
> I emailed the Ubuntu Touch list to ask if there is a plan.
>
> cheers,
>
> Henry
>
>
>
>
>
> ------------------------------
>
> Message: 2
> Date: Wed, 20 Mar 2013 16:52:12 +0100
> From: Aaron Richiger <a.richi at bluewin.ch>
> Subject: Re: [PySide] Honkin Idea for PySide's thriving
> To: pyside at qt-project.org
> Message-ID: <5149DB2C.3090403 at bluewin.ch>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> Am 20.03.2013 15:54, schrieb Stephan Deibel:
> > Christian Tismer wrote:
> >> Make PySide the new standard Gui toolkit for Python!
> > I'm also a bit concerned it would slow things down and might be hard
> > given the copyright ownership.  It seems like an idea to keep in mind
> > for the future, but there is much to do before I would attempt this.
> >
> > - Stephan
> Great idea in theory, but I have to agree with the others, that I prefer
> PySide available via site-packages for the named reasons. But I'm very
> glad, that we are moving and that there seems to be already a little bit
> of effort towards PySide for Qt5 and bug fixing / code review
> improvements. We are still working a lot on PySide for Android and will
> provide you with more information as soon as we have taken the most
> important first steps!
>
> Aaron
> >
> > _______________________________________________
> > PySide mailing list
> > PySide at qt-project.org
> > http://lists.qt-project.org/mailman/listinfo/pyside
>
>
>
> ------------------------------
>
> Message: 3
> Date: Wed, 20 Mar 2013 11:11:04 -0500
> From: Sergio Pulgar?n <serpulga at gmail.com>
> Subject: Re: [PySide] PySide Digest, Vol 14, Issue 18
> To: pyside at qt-project.org
> Message-ID:
>         <
> CAKiRVASUoV2fJXku3tnvARQbUfwnCoQuz31xU7oXe3RYk3APkw at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hi, I might be late to the discussion here, but I would like to
> share my thoughts anyways.
> It is no secret that PySide and PyQt4 are highly compatible
> with each other; with a few modifications you can port back
> and forth easly, and even create cross-compatible sources.
> So basically, at a end user level, there are no (or a few) arguments
> to chose one or the other. The real advantage the PySide has over
> PyQt4 are the licensing options. PySide has the LGPL option, while
> PyQt4 only GPL, last time I checked. I think this alone, makes people
> want to use/port to PySide.
> Another potential advantage is PySide's visibility through the Qt-Project,
> which I don't think it's exploited at its potential.
>
> About being included in the Python standard library, I agree that it's not
> a viable option, however, I don't think that PySide needs to be in the
> standard,
> in order for it to be the defacto GUI toolkit for Python, it needs to
> install easily across OSs and distros, be well documented, be active
> as a project, have a newbie friendly community, etc.
>
> Finally, I would like to mention GitHub. GitHub currently has a huge
> momentum, and PySide can take advantage of that, to gain activeness and
> user contribution.
>
>
>
> Sergio
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL:
> http://lists.qt-project.org/pipermail/pyside/attachments/20130320/032a8f69/attachment-0001.html
>
> ------------------------------
>
> Message: 4
> Date: Wed, 20 Mar 2013 17:35:04 +0100
> From: todd rme <toddrme2178 at gmail.com>
> Subject: Re: [PySide] PySide Digest, Vol 14, Issue 18
> To: pyside at qt-project.org
> Message-ID:
>         <CADb7s=uG7hywRN5LyWoTqC6nHGD-eLgnidJO=
> KFWhdYd2yH9TQ at mail.gmail.com>
> Content-Type: text/plain; charset=ISO-8859-1
>
> On Wed, Mar 20, 2013 at 5:11 PM, Sergio Pulgar?n <serpulga at gmail.com>
> wrote:
> >
> > Hi, I might be late to the discussion here, but I would like to
> > share my thoughts anyways.
> > It is no secret that PySide and PyQt4 are highly compatible
> > with each other; with a few modifications you can port back
> > and forth easly, and even create cross-compatible sources.
> > So basically, at a end user level, there are no (or a few) arguments
> > to chose one or the other. The real advantage the PySide has over
> > PyQt4 are the licensing options. PySide has the LGPL option, while
> > PyQt4 only GPL, last time I checked. I think this alone, makes people
> > want to use/port to PySide.
>
> That only helps if you are writing a closed-source application.
> Anyone who want to create an open-source end-user application is not
> going to care.  They are going to base it on things like features,
> support, how active the community is, what is being used by other
> projects, and what middle-level python modules they want to use
> support.  For these people, PyQt4 is currently the better choice on
> all fronts.  And these people are exactly the ones who are most likely
> to want to get involved in and contribute back to pyside if they use
> it.
>
> If pyside is going to succeed, it will have to do so based on
> something other than license alone.  Community, support, and what is
> being used by other projects is a chicken-and-egg problem, pyside will
> not have these until it gains some momentum.  And it will be hard to
> get middle-level toolkits to work with pyside exclusively precisely
> because it is not much harder than supporting both pyqt4 and pyside at
> the same time, and without much user interest there is no reason to
> support pyside at all.
>
> So if pyside is going to get ahead it I think it needs to have some
> sort of compelling features that are lacked by PyQt4.  I think that is
> the only way open-source projects are going to use it.
>

> ------------------------------
>
> Message: 5
> Date: Wed, 20 Mar 2013 13:17:01 -0400
> From: Stephan Deibel <sdeibel at wingware.com>
> Subject: [PySide] Summary of PySide BoF at PyCon
> To: "pyside at qt-project.org" <pyside at qt-project.org>
> Message-ID: <5149EF0D.9060107 at wingware.com>
> Content-Type: text/plain; charset=UTF-8; format=flowed
>
> Hi,
>
> Here is a summary of the PySide BoF that was held on Saturday March
> 16that PyCon 2013 in Santa Clara, CA.  We found there is a lot of
> interest at companies that chose PySide for its licensing.  Several
> companies that were present started a collaboration in order to move
> towards a work plan and coordinated funding and further development of
> PySide.  If you are interested in joining this effort, please contact me.
>
> Introductions
> -------------
>
> The purpose of this BoF was to gather together users invested in PySide,
> review the status of PySide, and move towards revitalizing the project.
> Twelve individuals attended, including representatives of Wingware,
> Enthought, Valve, Disney, and several other companies.
>
> The meeting started with each individual introducing themselves and
> describing their use of PySide and any concerns that they have.
> Licensing issues were the reason most of the users present (representing
> at least 5 companies) chose PySide over PyQt.  All of those citing
> licensing as a problem have a scripting API for their users, which
> requires royalties to be paid on revenues if PyQt is used.
>
> Experiences with stability of PySide varied.  Though most had run into
> some issues, some were able to work around them while most others felt
> that fixes are needed for significant bugs.  Several present indicated a
> concern for the overall future of PySide, particularly in the context of
> supporting Qt5, making PySide more maintainable, and reestablishing a
> team of reviewers and committers.
>
> Several attendees expressed gratitude to the previous developers of
> PySide, especially Hugo for remaining involved as approver/committer.
>
> Discussion of Issues
> --------------------
>
> A discussion of the major issues to be solved followed.  Issues brought
> up included stability, Qt5 support, lack of reviewers and committers,
> problems with the current review process, need for doing a release from
> latest revision, and need to process the accumulated list of bug reports
> that have not received any action.
>
> It was noted that Digia is not returning calls or emails concerning
> PySide.  The possibility of forking PySide and moving it out of the Qt
> Project development toolset was brought up but consensus was established
> that this was not desirable if it could be avoided.  If a fork is
> considered later, legal consul would be asked about trademark, licensing
> considerations, and the potential for re-merging changes into the Qt
> Project at a later date.
>
> Some specific bugs were discussed, including causes of life cycle
> issues, and whether bugs are fixable without rewriting parts of PySide.
>
> Plan of Action
> --------------
>
> Several attending stated that it may be possible for their companies to
> providing funds or staff, given an acceptable development plan and
> process.  This will be investigated further by each potential
> contributor, and some additional companies were identified and will be
> approached.
>
> Representatives of Enthought stated that they could host a week-long
> meeting aimed at creating a work plan and kick-starting work on PySide.
>
> [The one-hour time available in the open space slot ended here and a
> subset of the attendees adjourned to a nearby bar :-]
>
> Several attendees offered to act as future reviewers and committers for
> PySide:  John Ehresman (contributor to PyGTK) from Wingware, Christian
> Tismer (author of Stackless Python), along with Enthought staff:  Robin
> Dunn (author of wxPython), Jason McCampbell, and Robert Kern.
>
> Discussion followed concerning development approaches.  Those present
> agreed that rewriting PySide from scratch is a large task and would lose
> the special cases already worked through and dealt with by PySide.  The
> possibility of a partial rewrite in Python was raised.  Another
> possibility raised was to convert the XML encoding of special cases into
> some more readable and maintainable form, similar to the one used in
> PyGTK and PyScintilla2.  Expanded developer documentation is also needed.
>
> A discussion of the amount and cost of work involved followed.
> Estimates ranged from about 1 to 4 FTE years to cover bug fixes, Qt5
> support, and sheparding the project towards a sustainable future.
>
> Action Items
> ------------
>
> The meeting attendees will work together to arrange a week in the next
> few months, to be held at Enthought in Austin TX, in order to develop a
> work plan to serve as the foundation for collaboration among interested
> companies.  Possible attendees include John Ehresman, Robin Dunn,
> Christian Tismer and any others interested and able to attend.
>
> Several individuals will further investigate the possibility of their
> companies contributing development funds and/or effort.
>
> Enthought will look into any legal issues that need to be addressed.
>
> Several attendees will try to contact other companies identified at the
> meeting, that might be able to participate in drafting and
> implementation of a work plan.  This could include using the recent
> PySide survey on this mailing list to find potential contributors.
>
> A PySide sprint will be held at SciPy 2013 in June.
>
>
>
This is very good news, it sounds very promising to me.
I still see branching off as an option, if it would help move
things more quickly. If Digia is indeed looking the other way regarding
PySide, maybe they will everything easy, legal aspects, keeping
the brand, etc.







>
> ------------------------------
>
> Message: 6
> Date: Wed, 20 Mar 2013 14:30:30 -0500
> From: Tim Doty <thoromyr at mac.com>
> Subject: Re: [PySide] PySide Digest, Vol 14, Issue 18
> To: todd rme <toddrme2178 at gmail.com>
> Cc: pyside at qt-project.org
> Message-ID: <305F0703-E4A0-49BE-B18D-D3C566CDABED at mac.com>
> Content-Type: text/plain; charset=iso-8859-1
>
> For what its worth, the reason I switched from PyQt to PySide was not the
> license. At the time it looked like it was being supported (Nokia) and was
> finally stable. The push for me was that PyQt isn't really very friendly --
> if someone wants to use your app they have to get PyQt installed -- and the
> site is geared toward developers, not end users.
>
> I had no end of arguments from people insisting they couldn't use the app
> because they couldn't get SIP installed and besides which they didn't have
> a compiler. And as if that wasn't enough PyQt isn't particularly easy to
> get setup on OS X. It only works well on linux, or Windows if you create a
> self-contained installer.
>
> Although in principle PyQt is cross platform, my experience is that it is
> cross platform only if your platform is linux or you create an installer
> for Windows. PySide has been easier in that regard, though it has had some
> snafus. I still prefer it because I want to remain platform agnostic. I
> used to be solely linux myself, but I've been using OS X more and more. And
> most people who would use anything I write are on some flavor of windows.
>
> >From that perspective having an easy way to bundle needed components is a
> must. Admittedly, if the applications are in-house or focused on linux this
> is less of an issue but for broad use it really is a must. End users need
> simple.
>
> Tim Doty
>
> On Mar 20, 2013, at 11:35 AM, todd rme <toddrme2178 at gmail.com> wrote:
>
> > On Wed, Mar 20, 2013 at 5:11 PM, Sergio Pulgar?n <serpulga at gmail.com>
> wrote:
> >>
> >> Hi, I might be late to the discussion here, but I would like to
> >> share my thoughts anyways.
> >> It is no secret that PySide and PyQt4 are highly compatible
> >> with each other; with a few modifications you can port back
> >> and forth easly, and even create cross-compatible sources.
> >> So basically, at a end user level, there are no (or a few) arguments
> >> to chose one or the other. The real advantage the PySide has over
> >> PyQt4 are the licensing options. PySide has the LGPL option, while
> >> PyQt4 only GPL, last time I checked. I think this alone, makes people
> >> want to use/port to PySide.
> >
> > That only helps if you are writing a closed-source application.
> > Anyone who want to create an open-source end-user application is not
> > going to care.  They are going to base it on things like features,
> > support, how active the community is, what is being used by other
> > projects, and what middle-level python modules they want to use
> > support.  For these people, PyQt4 is currently the better choice on
> > all fronts.  And these people are exactly the ones who are most likely
> > to want to get involved in and contribute back to pyside if they use
> > it.
> >
> > If pyside is going to succeed, it will have to do so based on
> > something other than license alone.  Community, support, and what is
> > being used by other projects is a chicken-and-egg problem, pyside will
> > not have these until it gains some momentum.  And it will be hard to
> > get middle-level toolkits to work with pyside exclusively precisely
> > because it is not much harder than supporting both pyqt4 and pyside at
> > the same time, and without much user interest there is no reason to
> > support pyside at all.
> >
> > So if pyside is going to get ahead it I think it needs to have some
> > sort of compelling features that are lacked by PyQt4.  I think that is
> > the only way open-source projects are going to use it.
> > _______________________________________________
> > PySide mailing list
> > PySide at qt-project.org
> > http://lists.qt-project.org/mailman/listinfo/pyside
>
>
>
> ------------------------------
>
> Message: 7
> Date: Wed, 20 Mar 2013 19:09:14 -0700
> From: Christian Tismer <tismer at stackless.com>
> Subject: Re: [PySide] Honkin Idea for PySide's thriving
> To: Aaron Richiger <a.richi at bluewin.ch>
> Cc: pyside at qt-project.org
> Message-ID: <514A6BCA.2010308 at stackless.com>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> On 3/20/13 8:52 AM, Aaron Richiger wrote:
> > Am 20.03.2013 15:54, schrieb Stephan Deibel:
> >> Christian Tismer wrote:
> >>> Make PySide the new standard Gui toolkit for Python!
> >> I'm also a bit concerned it would slow things down and might be hard
> >> given the copyright ownership.  It seems like an idea to keep in mind
> >> for the future, but there is much to do before I would attempt this.
> >>
> >> - Stephan
> > Great idea in theory, but I have to agree with the others, that I prefer
> > PySide available via site-packages for the named reasons. But I'm very
> > glad, that we are moving and that there seems to be already a little bit
> > of effort towards PySide for Qt5 and bug fixing / code review
> > improvements. We are still working a lot on PySide for Android and will
> > provide you with more information as soon as we have taken the most
> > important first steps!
>
> Well, thanks for all the input!
>
> I'm essentially also in favor to have PySide in site-packages.
> My enthusiastic mail was more expressing what I like to see
> PySide perceived as, without necessarily having it as a super-large
> battery. (As a side, the Python Batteries have actually grown too large
> already and we need way to strip that down...).
>
> So what I want is to make PySide a very supported, very well acknowledged
> package that gets its #1 recommendation from Python, something like that.
> I don't want to slow down the development and avoid licensing issues.
>
> But getting many people interested in working on PySide could be a
> benefit if it would get its place in Python.org.
>
> Maybe we can move the mailing list there, and or have some announcement
> on python-announce that we want to improve PySide and spread the word?
>
> cheers - chris
>
> --
> Christian Tismer             :^)   <mailto:tismer at stackless.com>
> Software Consulting          :     Have a break! Take a ride on Python's
> Karl-Liebknecht-Str. 121     :    *Starship* http://starship.python.net/
> 14482 Potsdam                :     PGP key -> http://pgp.uni-mainz.de
> phone +49 173 24 18 776  fax +49 (30) 700143-0023
> PGP 0x57F3BF04       9064 F4E1 D754 C2FF 1619  305B C09C 5A3B 57F3 BF04
>        whom do you want to sponsor today?   http://www.stackless.com/
>
>
>
> ------------------------------
>
> _______________________________________________
> PySide mailing list
> PySide at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/pyside
>
>
> End of PySide Digest, Vol 14, Issue 19
> **************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20130320/30120a48/attachment.html>

More information about the PySide mailing list