Re: Deprecation a little harsh?
Hi,
We've been able to do quite painless upgrades to our software as well.
Considering that we started with Django 1.1, stuck quite while at 1.3 and
recently we did upgrade to 1.5
Most notable change was {% url %} tag, but nothing really major.
Though we don't use many external libs - we've built our own.
Only real problem we have is that our deployment system sucks with the fact
that it had for a long time shared libraries (including Django) which did had
some impact on upgrades but since moving towards virtualenv usage we've been
able to get rid of that problem as well, but that has nothing to do with actual
Django.
Line count wise we hit quite similar as Florian:
main project consists:
~200 models,
.py 44805 lines
.html 143937 lines
.js 101317 lines (this is explained that we use dojotoolkit/extjs4 based single
page app)
And our library:
.py 25079 lines
.js 25500 lines
And my 2 cents here -
What I really like that Django getting new, better features while keeping very
good deprecation policy. And yes, that sometimes requires that I lose few hours
of sleep while upgrading but eventually it has to be done - it's just that I
can't sit forever on same version. Sooner it's done easier it is and less time
it takes.
On Tue, 13 Aug 2013 13:16:01 -0400
François Schiettecatte wrote:
> Florian
>
> Here are wc -l line counts:
>
> Project 1
> .py 28574
> .html 16967
> (this is a little misleading because we don't use model.py but a separate
> REST API to handle all the storage)
>
>
> Project 2
> .py 43199
> .html 91804
>
>
> Project 3
> .py 32441
> .html 86684
>
>
> Cheers
>
> François
>
>
> On Aug 13, 2013, at 12:31 PM, Florian Apolloner wrote:
>
> > Hi François,
> >
> > On Tuesday, August 13, 2013 5:46:10 PM UTC+2, François Schiettecatte wrote:
> > I have done 1.3.x -> 1.4.x -> 1.5.x and they have all been painless, each
> > migration taking less than 1/2 day. Point being that back-porting is not
> > something I would ever need.
> >
> > It's good to hear that some people are keeping up2date and it didn't cause
> > any pain! Do you mind sharing how big (lines of code wise) those apps are
> > (just a rough classification).
> >
> > Regards,
> > Florian
> >
--
Jani Tiainen
--
You received this message because you are subscribed to the Google Groups
"Django developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to django-developers+unsubscr...@googlegroups.com.
To post to this group, send email to django-developers@googlegroups.com.
Visit this group at http://groups.google.com/group/django-developers.
For more options, visit https://groups.google.com/groups/opt_out.
Re: Deprecation a little harsh?
Florian Here are wc -l line counts: Project 1 .py 28574 .html 16967 (this is a little misleading because we don't use model.py but a separate REST API to handle all the storage) Project 2 .py 43199 .html 91804 Project 3 .py 32441 .html 86684 Cheers François On Aug 13, 2013, at 12:31 PM, Florian Apolloner wrote: > Hi François, > > On Tuesday, August 13, 2013 5:46:10 PM UTC+2, François Schiettecatte wrote: > I have done 1.3.x -> 1.4.x -> 1.5.x and they have all been painless, each > migration taking less than 1/2 day. Point being that back-porting is not > something I would ever need. > > It's good to hear that some people are keeping up2date and it didn't cause > any pain! Do you mind sharing how big (lines of code wise) those apps are > (just a rough classification). > > Regards, > Florian > > -- > You received this message because you are subscribed to the Google Groups > "Django developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to django-developers+unsubscr...@googlegroups.com. > To post to this group, send email to django-developers@googlegroups.com. > Visit this group at http://groups.google.com/group/django-developers. > For more options, visit https://groups.google.com/groups/opt_out. > > -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To post to this group, send email to django-developers@googlegroups.com. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
Re: Deprecation a little harsh?
Hi François, On Tuesday, August 13, 2013 5:46:10 PM UTC+2, François Schiettecatte wrote: > > I have done 1.3.x -> 1.4.x -> 1.5.x and they have all been painless, each > migration taking less than 1/2 day. Point being that back-porting is not > something I would ever need. > It's good to hear that some people are keeping up2date and it didn't cause any pain! Do you mind sharing how big (lines of code wise) those apps are (just a rough classification). Regards, Florian -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To post to this group, send email to django-developers@googlegroups.com. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
Re: Deprecation a little harsh?
Hi Not sure if this is useful, but I thought it might be helpful to throw my perspective as someone who has built four sites with Django (two of which are public, and one of those has to be HIPAA compliant). I will update to dot releases after the main release to give the main release time to shake out, so I will do 1.4.x to 1.5.1. And I will update to security releases as soon as they come out. I have done 1.3.x -> 1.4.x -> 1.5.x and they have all been painless, each migration taking less than 1/2 day. Point being that back-porting is not something I would ever need. Cheers François On Aug 13, 2013, at 11:34 AM, Michael Manfre wrote: > If there is interest in the community to backport security fixes to no longer > supported versions of Django, what is the likelihood that a core dev would > merge them in to the appropriate stable branch? This would not include > packaging an official release, but would provide a way for those stuck on > older versions a better way to help others who are similarly stuck. I realize > that it does require time to verify pull requests and "too much effort" is a > completely valid answer. > > I'm not stuck and currently have no vested interest in backporting anything > to unsupported versions. I'm only trying to further the discussion. > > Regards, > Michael Manfre > > > > On Tue, Aug 13, 2013 at 11:21 AM, Jacob Kaplan-Moss > wrote: > I'm sorry; I was snarkier and nastier than I should have been (and than I > intended to be). Thanks for calling me on it; I'll try to do better next time. > > Jacob > > > On Tue, Aug 13, 2013 at 10:03 AM, Andre Terra wrote: > On Tue, Aug 13, 2013 at 9:22 AM, Jacob Kaplan-Moss wrote: > > We have apps in production running Django 1.3. There won't be any security > fixes. If there's a critical vulnerability, we may have to do a lot of unpaid > work to either backport the fix, > > I have to say I find this kinda hilarious: you *know* it's a lot of work to > backport stuff, and you'd like *us* to do that work instead of you. > > I'm not asking anyone to do my job for me (I hope) but it would be really > nice to have something like 3 years of support for core infrastructure like > Django, that's really painful to upgrade, and even more painful to replace. > It would certainly help me to sleep better at night. > > But you are, actually, asking us to work for you. And we're happy to do it! > This is what open source is all about; volunteering to do work (often rather > thankless work) to help other people sleep at night. But there's a limit to > the free time we have, and there's a limit to the amount of scut work you can > expect a volunteer community to do for you. > > > Not to hijack the thread purposely, but it's hard not to point out what a > great example of "poor attitude" this is, as was called out in a thread here > on django-developers a while back[0]. Since you'd rather not have discussions > post facto, I guess it doesn't hurt to be timely. > > Cheers, > AT > > [0] > https://groups.google.com/forum/#!topic/django-developers/DUQtBrM2iTs/discussion > > -- > You received this message because you are subscribed to the Google Groups > "Django developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to django-developers+unsubscr...@googlegroups.com. > To post to this group, send email to django-developers@googlegroups.com. > Visit this group at http://groups.google.com/group/django-developers. > For more options, visit https://groups.google.com/groups/opt_out. > > > > > -- > You received this message because you are subscribed to the Google Groups > "Django developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to django-developers+unsubscr...@googlegroups.com. > To post to this group, send email to django-developers@googlegroups.com. > Visit this group at http://groups.google.com/group/django-developers. > For more options, visit https://groups.google.com/groups/opt_out. > > > -- > You received this message because you are subscribed to the Google Groups > "Django developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to django-developers+unsubscr...@googlegroups.com. > To post to this group, send email to django-developers@googlegroups.com. > Visit this group at http://groups.google.com/group/django-developers. > For more options, visit https://groups.google.com/groups/opt_out. > > -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To post to this group, send email to django-developers@googlegroups.com. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
Re: Deprecation a little harsh?
On Aug 13, 2013, at 11:34 AM, Michael Manfre wrote: > If there is interest in the community to backport security fixes to no longer > supported versions of Django, what is the likelihood that a core dev would > merge them in to the appropriate stable branch? This would not include > packaging an official release, but would provide a way for those stuck on > older versions a better way to help others who are similarly stuck. I realize > that it does require time to verify pull requests and "too much effort" is a > completely valid answer. Lack of CI is probably going to be one of the biggest blockers. Without CI it's up to the merging developers to run the tests on all the combinations of stuff we support which isn't the easiest or the quickest thing to setup and maintain. - Donald Stufft PGP: 0x6E3CBCE93372DCFA // 7C6B 7C5D 5E2B 6356 A926 F04F 6E3C BCE9 3372 DCFA signature.asc Description: Message signed with OpenPGP using GPGMail
Re: Deprecation a little harsh?
If there is interest in the community to backport security fixes to no longer supported versions of Django, what is the likelihood that a core dev would merge them in to the appropriate stable branch? This would not include packaging an official release, but would provide a way for those stuck on older versions a better way to help others who are similarly stuck. I realize that it does require time to verify pull requests and "too much effort" is a completely valid answer. I'm not stuck and currently have no vested interest in backporting anything to unsupported versions. I'm only trying to further the discussion. Regards, Michael Manfre On Tue, Aug 13, 2013 at 11:21 AM, Jacob Kaplan-Moss wrote: > I'm sorry; I was snarkier and nastier than I should have been (and than I > intended to be). Thanks for calling me on it; I'll try to do better next > time. > > Jacob > > > On Tue, Aug 13, 2013 at 10:03 AM, Andre Terra wrote: > >> On Tue, Aug 13, 2013 at 9:22 AM, Jacob Kaplan-Moss wrote: >> >>> >>> We have apps in production running Django 1.3. There won't be any security fixes. If there's a critical vulnerability, we may have to do a lot of unpaid work to either backport the fix, >>> >>> >>> I have to say I find this kinda hilarious: you *know* it's a lot of work >>> to backport stuff, and you'd like *us* to do that work instead of you. >>> >>> I'm not asking anyone to do my job for me (I hope) but it would be really nice to have something like 3 years of support for core infrastructure like Django, that's really painful to upgrade, and even more painful to replace. It would certainly help me to sleep better at night. >>> >>> But you are, actually, asking us to work for you. And we're happy to do >>> it! This is what open source is all about; volunteering to do work (often >>> rather thankless work) to help other people sleep at night. But there's a >>> limit to the free time we have, and there's a limit to the amount of scut >>> work you can expect a volunteer community to do for you. >>> >> >> >> Not to hijack the thread purposely, but it's hard not to point out what a >> great example of "poor attitude" this is, as was called out in a thread >> here on django-developers a while back[0]. Since you'd rather not have >> discussions *post facto*, I guess it doesn't hurt to be timely. >> >> Cheers, >> AT >> >> [0] >> https://groups.google.com/forum/#!topic/django-developers/DUQtBrM2iTs/discussion >> >> -- >> You received this message because you are subscribed to the Google Groups >> "Django developers" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to django-developers+unsubscr...@googlegroups.com. >> To post to this group, send email to django-developers@googlegroups.com. >> Visit this group at http://groups.google.com/group/django-developers. >> For more options, visit https://groups.google.com/groups/opt_out. >> >> >> > > -- > You received this message because you are subscribed to the Google Groups > "Django developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to django-developers+unsubscr...@googlegroups.com. > To post to this group, send email to django-developers@googlegroups.com. > Visit this group at http://groups.google.com/group/django-developers. > For more options, visit https://groups.google.com/groups/opt_out. > -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To post to this group, send email to django-developers@googlegroups.com. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
Re: Deprecation a little harsh?
I'm sorry; I was snarkier and nastier than I should have been (and than I intended to be). Thanks for calling me on it; I'll try to do better next time. Jacob On Tue, Aug 13, 2013 at 10:03 AM, Andre Terra wrote: > On Tue, Aug 13, 2013 at 9:22 AM, Jacob Kaplan-Moss wrote: > >> >> >>> We have apps in production running Django 1.3. There won't be any >>> security fixes. If there's a critical vulnerability, we may have to do a >>> lot of unpaid work to either backport the fix, >> >> >> I have to say I find this kinda hilarious: you *know* it's a lot of work >> to backport stuff, and you'd like *us* to do that work instead of you. >> >> >>> I'm not asking anyone to do my job for me (I hope) but it would be >>> really nice to have something like 3 years of support for core >>> infrastructure like Django, that's really painful to upgrade, and even more >>> painful to replace. It would certainly help me to sleep better at night. >>> >> >> But you are, actually, asking us to work for you. And we're happy to do >> it! This is what open source is all about; volunteering to do work (often >> rather thankless work) to help other people sleep at night. But there's a >> limit to the free time we have, and there's a limit to the amount of scut >> work you can expect a volunteer community to do for you. >> > > > Not to hijack the thread purposely, but it's hard not to point out what a > great example of "poor attitude" this is, as was called out in a thread > here on django-developers a while back[0]. Since you'd rather not have > discussions *post facto*, I guess it doesn't hurt to be timely. > > Cheers, > AT > > [0] > https://groups.google.com/forum/#!topic/django-developers/DUQtBrM2iTs/discussion > > -- > You received this message because you are subscribed to the Google Groups > "Django developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to django-developers+unsubscr...@googlegroups.com. > To post to this group, send email to django-developers@googlegroups.com. > Visit this group at http://groups.google.com/group/django-developers. > For more options, visit https://groups.google.com/groups/opt_out. > > > -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To post to this group, send email to django-developers@googlegroups.com. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
Re: Deprecation a little harsh?
On Tue, Aug 13, 2013 at 9:22 AM, Jacob Kaplan-Moss wrote: > > >> We have apps in production running Django 1.3. There won't be any >> security fixes. If there's a critical vulnerability, we may have to do a >> lot of unpaid work to either backport the fix, > > > I have to say I find this kinda hilarious: you *know* it's a lot of work > to backport stuff, and you'd like *us* to do that work instead of you. > > >> I'm not asking anyone to do my job for me (I hope) but it would be really >> nice to have something like 3 years of support for core infrastructure like >> Django, that's really painful to upgrade, and even more painful to replace. >> It would certainly help me to sleep better at night. >> > > But you are, actually, asking us to work for you. And we're happy to do > it! This is what open source is all about; volunteering to do work (often > rather thankless work) to help other people sleep at night. But there's a > limit to the free time we have, and there's a limit to the amount of scut > work you can expect a volunteer community to do for you. > Not to hijack the thread purposely, but it's hard not to point out what a great example of "poor attitude" this is, as was called out in a thread here on django-developers a while back[0]. Since you'd rather not have discussions *post facto*, I guess it doesn't hurt to be timely. Cheers, AT [0] https://groups.google.com/forum/#!topic/django-developers/DUQtBrM2iTs/discussion -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To post to this group, send email to django-developers@googlegroups.com. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
Re: Deprecation a little harsh?
On Tue, Aug 13, 2013 at 1:37 AM, Chris Wilson wrote: > I would love to see support extended for a bit longer after deprecation. This is a matter of resources; we struggle to maintain security releases against 3 simultaneous releases (e.g. right now 1.4.x, 1.5.x, and the up-coming 1.6). Adding a fourth probably isn't possible without a ton of help. > We have apps in production running Django 1.3. There won't be any security > fixes. If there's a critical vulnerability, we may have to do a lot of > unpaid work to either backport the fix, I have to say I find this kinda hilarious: you *know* it's a lot of work to backport stuff, and you'd like *us* to do that work instead of you. > I'm not asking anyone to do my job for me (I hope) but it would be really > nice to have something like 3 years of support for core infrastructure like > Django, that's really painful to upgrade, and even more painful to replace. > It would certainly help me to sleep better at night. > But you are, actually, asking us to work for you. And we're happy to do it! This is what open source is all about; volunteering to do work (often rather thankless work) to help other people sleep at night. But there's a limit to the free time we have, and there's a limit to the amount of scut work you can expect a volunteer community to do for you. Look at it this way: you run 1.3 still, so you have a personal vested interest in backporting work. If *you* don't have the time to do it, what makes you think that *we* do? Jacob -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To post to this group, send email to django-developers@googlegroups.com. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
Re: Deprecation a little harsh?
2013/8/13 Chris Wilson > It certainly seems that people are eager to rip things out the very second > that the deprecation waiting period expires and they are "allowed" to. If someone introduced a deprecation with a PendingDeprecationWarning, and someone else later advanced all PendingDeprecationWarnings to DeprecationWarnings, that deprecation would start immediately with a DeprecationWarning. To avoid this problem, every time we fork a stable branch from master, we must: 1) remove all code that's under DeprecationWarning, 2) bump PendingDeprecationWarning to DeprecationWarning. And then we're in a position where it's safe to introduce new deprecations. You may be under the impression that we're rushing to remove deprecated code, but that isn't the whole story. It's necessary to avoid accidentally accelerated deprecations. -- Aymeric. -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To post to this group, send email to django-developers@googlegroups.com. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
Re: Deprecation a little harsh?
Hi all, On Mon, 12 Aug 2013, Jacob Kaplan-Moss wrote: If you'd like to help push us closer to where *you* think the right place balance is, the best thing to do is to watch our development process and speak up in the moment. These sorts of general, post-facto observations don't give us a ton that we can really act on, and ultimately are going to come down to "feeling" arguments. I'd like to avoid those because they're kinda counter productive. If you want to do something about it, help us on future decisions; second-guessing the past doesn't really get us anywhere. I would love to see support extended for a bit longer after deprecation. Maybe not in every case, but where it doesn't hurt too much. It certainly seems that people are eager to rip things out the very second that the deprecation waiting period expires and they are "allowed" to. We have apps in production running Django 1.3. There won't be any security fixes. If there's a critical vulnerability, we may have to do a lot of unpaid work to either backport the fix, or upgrade the site to Django 1.6, which will undoubtedly break a lot of stuff. I'm not asking anyone to do my job for me (I hope) but it would be really nice to have something like 3 years of support for core infrastructure like Django, that's really painful to upgrade, and even more painful to replace. It would certainly help me to sleep better at night. Cheers, Chris. -- Aptivate | http://www.aptivate.org | Phone: +44 1223 967 838 Citylife House, Sturton Street, Cambridge, CB1 2QF, UK Aptivate is a not-for-profit company registered in England and Wales with company number 04980791. -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To post to this group, send email to django-developers@googlegroups.com. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
Re: Deprecation a little harsh?
Hi Simon - Here's the thing: I'm sensitive to the fact that you think we're moving too fast, but you have to understand that we also hear that we're moving too *slow*. We have to strike a balance, and I'm happy where we've struck that balance. If you'd like to help push us closer to where *you* think the right place balance is, the best thing to do is to watch our development process and speak up in the moment. These sorts of general, post-facto observations don't give us a ton that we can really act on, and ultimately are going to come down to "feeling" arguments. I'd like to avoid those because they're kinda counter productive. If you want to do something about it, help us on future decisions; second-guessing the past doesn't really get us anywhere. So again, thanks for your suggestion, I hear your concerns. Jacob On Mon, Aug 12, 2013 at 7:07 PM, Simon Litchfield wrote: > > On Monday, 12 August 2013 17:23:40 UTC+10, Aymeric Augustin wrote: >> >> It would be interesting to describe what actual problems these libraries >> encountered. Were they caused by actual deprecations or by changes in >> private APIs? >> > > One example would be simplejson, why not leave a stub there that imports > json? IMHO that was a little draconian and needlessly broke several third > party apps. > > -- > You received this message because you are subscribed to the Google Groups > "Django developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to django-developers+unsubscr...@googlegroups.com. > To post to this group, send email to django-developers@googlegroups.com. > Visit this group at http://groups.google.com/group/django-developers. > For more options, visit https://groups.google.com/groups/opt_out. > -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To post to this group, send email to django-developers@googlegroups.com. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
Re: Deprecation a little harsh?
On Monday, 12 August 2013 17:23:40 UTC+10, Aymeric Augustin wrote: > > It would be interesting to describe what actual problems these libraries > encountered. Were they caused by actual deprecations or by changes in > private APIs? > One example would be simplejson, why not leave a stub there that imports json? IMHO that was a little draconian and needlessly broke several third party apps. -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To post to this group, send email to django-developers@googlegroups.com. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
Re: Deprecation a little harsh?
On 12 août 2013, at 06:21, Simon Litchfield wrote:
> Celery, sorl-thumbnail, mptt, registration, shorturls, compressor, tinymce (I
> could go on) have been having trouble keeping up.
Hi Simon,
It would be interesting to describe what actual problems these libraries
encountered. Were they caused by actual deprecations or by changes in private
APIs?
I can talk a little bit about django-registration, having encountered an issue
when upgrading www.djangoproject.com to Django 1.5.
A trivial change was required to maintain equivalent functionality on Django
1.5: replace direct_to_template by TemplateView. ("Equivalent functionality"
doesn't require supporting custom user models.) The change was proposed in at
least two pull requests and received several +1 on BitBucket. Go look at the
reasons why the maintainer didn't merge it; it has nothing to do with Django
changing too fast.
--
Aymeric.
--
You received this message because you are subscribed to the Google Groups
"Django developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to django-developers+unsubscr...@googlegroups.com.
To post to this group, send email to django-developers@googlegroups.com.
Visit this group at http://groups.google.com/group/django-developers.
For more options, visit https://groups.google.com/groups/opt_out.
Re: Deprecation a little harsh?
Hi, On Monday, August 12, 2013 6:21:46 AM UTC+2, Simon Litchfield wrote: > > One of Django's key strengths is the large collection of apps. Some aren't > as regularly maintained as we'd like but we still love them. Is it a little > unreasonable to expect them all to move so fast? > Fast? Imo the web world is one of the fastest moving places in the IT; and at the speed we release Django releases you still have a year or more and often you are not affected by those changes… > Any one else found themselves spending too much time lately patching other > peoples' stuff? > Not really; generally people fix stuff quickly if I ask them, also the projects I maintain have compatibility for 1.7 already -- I agree that the database backend changes had been a bit much for me, but they are so much better now that it's totally worth it (and they are private api ;)) Cheers, Florian -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To post to this group, send email to django-developers@googlegroups.com. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
Re: Deprecation a little harsh?
Hi Simon, On Mon, Aug 12, 2013 at 12:21 PM, Simon Litchfield wrote: > It's great all the housekeeping we've been doing lately, and I'm sure we > all agree nice to have clean, tidy code; but I wonder if we've been a > little too unforgiving at the expense of easy compatibility with important > third party apps? > > Celery, sorl-thumbnail, mptt, registration, shorturls, compressor, tinymce > (I could go on) have been having trouble keeping up. > > Can you elaborate on exactly what deprecations are causing the problem here? Yes, we've been deprecating things, but AFAICT they've mostly been contrib apps, and the deprecation has been accompanied by moving the third party app into a public repository so anyone dependent on it can continue to use (and maintain) the code if they want. Also - if you pick one of those apps -- registration -- the issue wasn't so much deprecation it was integration with a new feature (custom user models). One of Django's key strengths is the large collection of apps. Some aren't > as regularly maintained as we'd like but we still love them. Is it a little > unreasonable to expect them all to move so fast? > What constitutes "too fast"? Class based views, for example, were put on deprecation path *2 years ago*. Yes, some apps are probably affected. But they've also had *lots* of time to be updated, and for at least 12 months of that time, any affected project should have been getting *lots* of warnings that they were using code that was due to be deprecated. So…. I suppose my problem here is that I don't see the problem -- at least, not as you've described it. Deprecation is already done on a long timeframe, and with plenty of very vocal warning messages. Is there a *specific* deprecation (or deprecations) that you see causing problems? Or is the problem that there are lots of apps out there using unofficial API internals, and are breaking as a result when Django is updated? Or is it that there are packages in our community that aren't being actively maintained? Yours Russ Magee %-) -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To post to this group, send email to django-developers@googlegroups.com. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
Deprecation a little harsh?
It's great all the housekeeping we've been doing lately, and I'm sure we all agree nice to have clean, tidy code; but I wonder if we've been a little too unforgiving at the expense of easy compatibility with important third party apps? Celery, sorl-thumbnail, mptt, registration, shorturls, compressor, tinymce (I could go on) have been having trouble keeping up. One of Django's key strengths is the large collection of apps. Some aren't as regularly maintained as we'd like but we still love them. Is it a little unreasonable to expect them all to move so fast? Any one else found themselves spending too much time lately patching other peoples' stuff? -- You received this message because you are subscribed to the Google Groups "Django developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To post to this group, send email to django-developers@googlegroups.com. Visit this group at http://groups.google.com/group/django-developers. For more options, visit https://groups.google.com/groups/opt_out.
