When do snippets get into Django?
martin f krafft
I am new to Django and could not find the answer to a very important
consideration I need to make before diving into Django more. I used
to work with Zope and Plone, and one of the major problems were
upgrades, not of Zope or Plone, but of all the products that were
needed for a particular site.
Django also uses pluggable apps. Between djangopluggables and
code.google.com (and possibly other sites), there's a lot of
external applications providing great functionality. However, I am
reluctant to rely on those, mainly due to my experience with Plone
and Zope.
What's Django's strategy to integrate external pluggables into the
main distribution and support it across upgrades? For instance,
django-tagging[0] seems like a good candidate for core
functionality. Does it stand a chance?
0. http://djangoplugables.com/projects/django-tagging/
What's the process for integrating such external functionality into
Django's core?
PS: Django seems a lot thinner and more elegant than Plone/Zope, so
upgrades are likely to be less of a problem, but still...
--
martin; (greetings from the heart of the sun.)
\____ echo mailto: !#^."<*>"|tr "<*> mailto:" net@madduck
"security here. yes, ma'am. yes. groucho glasses. yes, we're on it.
c'mon, guys. somebody gave an aardvark a nose-cut: somebody who
can't deal with deconstructionist humor. code blue."
-- http://azure.humbug.org.au/~aj/armadillos.txt
spamtraps: madduc...@madduck.net
Alex Gaynor
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
iEYEAREDAAYFAkoWZswACgkQIgvIgzMMSnVi5gCcC2S8dnR0VhKp2ZOiQZ7R04MZ
/3MAoJAacIK+HQ3kYteVJfVj214TjCIa
=jvHf
-----END PGP SIGNATURE-----
http://jacobian.org/writing/what-is-django-contrib/ here is a description of what one of the core developers see as the role of django contrib. It should be noted that 2 of the core devs said at euro django con that that they prefer the core of django to remain slim.
Alex
--
"I disapprove of what you say, but I will defend to the death your right to say it." --Voltaire
"The people's good is the highest law."--Cicero
Dougal Matthews
Or the seperate mailing list; http://groups.google.com/group/pinax-users
They are pushing towards improving the main release at the moment, with 0.7 coming soon. I think there will then be migration/upgrade support from then onwards. I believe its coming but not sure what stage it kicks in.
Pinax includes a huge number of apps, like django-tagging and more are being added.
Cheers,
Dougal
---
Dougal Matthews - @d0ugal
http://www.dougalmatthews.com/
martin f krafft
> http://jacobian.org/writing/what-is-django-contrib/ here is
> a description of what one of the core developers see as the role
> of django contrib. It should be noted that 2 of the core devs
> said at euro django con that that they prefer the core of django
> to remain slim.
Sounds good, and something like django.contrib is a great way
forward.
It does leave me wondering what the "de-facto standard" way to tag
contents is…
--
martin | http://madduck.net/ | http://two.sentenc.es/
"a kiss may ruin a human life."
-- oscar wilde
spamtraps: madduc...@madduck.net
Dougal Matthews
I've used it every time I've needed tagging and don't have a gripe with it yet.
also sprach Alex Gaynor <alex....@gmail.com> [2009.05.22.1545 +0200]:
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
iEYEAREDAAYFAkoXCs8ACgkQIgvIgzMMSnWZPQCgwpHeqinX1jwo8xhyvfLFz2Sf
JswAniqb2muIXEjw4p4LolqgDAJHPMPs
=95Qm
-----END PGP SIGNATURE-----
martin f krafft
> Pretty must everybody does seem to use django-tagging unless they have some
> special weird use-case.
>
> I've used it every time I've needed tagging and don't have a gripe with it
> yet.
So how would it find its way into contrib?
--
martin | http://madduck.net/ | http://two.sentenc.es/
"geld ist das brecheisen der macht."
- friedrich nietzsche
spamtraps: madduc...@madduck.net
Dougal Matthews
also sprach Dougal Matthews <doug...@gmail.com> [2009.05.22.2352 +0200]:
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
iEYEAREDAAYFAkoXJYUACgkQIgvIgzMMSnWN4gCgyHNiUCgnrEJPSpJrtUMhzaiO
V쀉털쭼鄃㥷쟋ॕ떳�핍�
=GAqC
-----END PGP SIGNATURE-----
Masklinn
> It doesn't. Why should it?
>
python modules are pulled into the stdlib?
martin f krafft
> It doesn't. Why should it?
Well, because django-tagging is, quote Jacob:
- optional
- de-facto standard
- the implementation of a common pattern
Tagging is a very common feature nowadays, isn't it? Having tag
support integrated with Django contrib would help create a tagging
standard across Django sites.
Or did I misunderstand?
--
martin | http://madduck.net/ | http://two.sentenc.es/
"prisons are built with stones of law,
brothels with bricks of religion."
-- william blake
spamtraps: madduc...@madduck.net
Kai Kuehne
On Fri, May 22, 2009 at 11:52 PM, Dougal Matthews <doug...@gmail.com> wrote:
> Pretty must everybody does seem to use django-tagging unless they have some
> special weird use-case.
Huh? Can you remove that "everybody", please? I've never used tagging
and I'm confident - many others too.
martin f krafft
> > Pretty must everybody does seem to use django-tagging unless
> > they have some special weird use-case.
>
> Huh? Can you remove that "everybody", please? I've never used
> tagging and I'm confident - many others too.
I think he implicitly meant "everybody who uses tagging does so with
django-tagging". But of course, it's still optional, as
all the apps in django.contrib should be. You won't be forced to use
it, but those who want tagging support need not look any further.
--
martin | http://madduck.net/ | http://two.sentenc.es/
"a man who does not realise
that he is half an animal
is only half a man."
-- thornton wilder
spamtraps: madduc...@madduck.net
Dougal Matthews
Good contrib packages tackle issues that that are not trivial, are bikeshed prone, and are difficult to get right for the common case.
martin f krafft
> I don't think tagging is a particularly hard or non-trivial.
No. All the more reason to push for a standard and prevent dozens
of different implementations, no?
> Anyway, this conversation is a bit irrelevant because; Core
> developers want contrib to be slim, so its going to have a really
> strong usecase to get in.
I probably won't be hard to make a use-case for this, given how tags
are omnipresent in the Web 2.0 and part of the "Semantic Web" idea.
They are also incredibly useful at structuring information. Sure,
it's easy to do, but why should't the feature be in Django?
I think I'll collect some more information and then take this to the
-devel list.
> If you start adding 'tagging' a huge number of other things start
> to feel like they need to be added too. It because a case of well,
> you added tagging, so why not this?
If there are strong use-cases, why not?
> Also, what benefit do we really get from having it in contrib? Ok,
> so it means everybody has it but by the sounds of things everybody
> uses it anyway...
>
> The difference between it being in contrib and being external is
> very minimal to the developer that uses it.
I disagree. For me, it's the difference of having to track it
myself, or letting my distro do that.
--
martin | http://madduck.net/ | http://two.sentenc.es/
"love is a grave mental disease."
-- platon
spamtraps: madduc...@madduck.net
Dougal Matthews
I can just imagine them saying no :) so thought I would question it as
a move/idea.
Also, might be a good idea to wait until 1.1 is out the door so they
have time to consider it properly...
Dougal
--
Sent from my mobile device
martin f krafft
> Fair enough. I don't really have an issue with the idea, could be
> quite handy.
>
> I can just imagine them saying no :) so thought I would question
> it as a move/idea.
Look what I found! Check out the fifth point, labeled #1, top-most
annoyance with Python, in
http://jacobian.org/writing/hate-python/
I think this is strong leverage. ;)
> Also, might be a good idea to wait until 1.1 is out the door so
> they have time to consider it properly...
I couldn't resist the blog comment, but this is good advice.
Thanks,
--
martin | http://madduck.net/ | http://two.sentenc.es/
"my experience is that as soon as people are old enough to know better,
they don't know anything at all."
James Bennett
> Look what I found! Check out the fifth point, labeled #1, top-most
> annoyance with Python, in
And?
While I like it as much as the next guy, django-tagging, to take your
example, is currently sitting at version 0.2.1, released in January
*2008*. It doesn't work with 1.0.x Django or with the forthcoming
1.1.x Django (per the note on its home page telling people to use an
SVN checkout). Prior to 0.2.1, there are two releases available; one
(0.2) just prior to the current release and the other (0.1) in May
2007. And there are currently over a hundred unresolved tickets open
in its issue tracker (by way of comparison, that's about three times
as many tickets as are open for the 1.1 milestone for the entire
Django framework).
Again: I like the app, and I like the folks behind it, but that just
won't cut it as a candidate for contrib. There's got to be strong
evidence that somebody's committed to fixing bugs and keeping the
application up-to-date and, right now, that just doesn't exist.
And, really, this is the sort of thing that leads me to say it
shouldn't be in contrib right now. We do need to have some sort of
document outlining criteria, but I'd like to think that commitment to
a stable API (e.g., the author is willing to call it "1.0"), and to
support and updates are absolute requirements.
And lest anyone complain, I'll point out that I'm living by that
standard as well; I've got several applications in the wild which seem
to be popular (one's even being packaged and distributed by Debian,
much as that frightens me), but I know that so far none of them live
up to the expectations I'd have of something in contrib, and so I
haven't really made any noise about why they're not in there.
--
"Bureaucrat Conrad, you are technically correct -- the best kind of correct."
martin f krafft
> And, really, this is the sort of thing that leads me to say it
> shouldn't be in contrib right now. We do need to have some sort of
> document outlining criteria, but I'd like to think that commitment to
> a stable API (e.g., the author is willing to call it "1.0"), and to
> support and updates are absolute requirements.
I completely agree, and it was a poor choice of example. After all,
I started "using" Django three days ago, have never used
django-tagging, but it /seemed/ a good example, since it /seemed/ to
fit Jacob's criteria perfectly well.
In the end, at the core of my question was the quest for a document
outlining the criteria, which I think is an important asset for
anyone evaluating Django.
Thank you for getting me/us back on the ground wrt django-tagging.
I'll take a look and since this is a core feature I am seeking,
maybe I can put time into it.
But how do we go about a document to specify what goes into contrib,
beyond Jacob's three criteria?
--
martin | http://madduck.net/ | http://two.sentenc.es/
"da haben wir es also: eine kirchliche ordnung mit priesterschaft,
theologie, kultus, sakrament;
kurz, alles das, was jesus von nazareth bekämpft hatte..."