The plone-based replacement for www.gnome.org struggles forward. But it won’t get much further unless we find some new Plone developers who can help make it happen.
I have a whole other blog post that I’ve been waiting to publish for about a year as things have crawled forwards, variously delayed by an svn account request (Olav’s mango system fixed that problem), an unstable test server, and our single Plone developer’s extreme lack of time. It’s called “New www.gnome.org: Test it now” and it has lots of great links to a test site, our design plans, and where to report bugs. Instead you get this, based on my email to gnome-web-list:
Status
The Plone stuff is now in GNOME’s svn, in the gnomeweb-plone module. We also have a backup of the database with the content (Zope format).
However, the svn is not quite in the desired structure and the “buildout” system probably needs some work. See the README for hints from Ramon.
I had hoped that Ramon could get those done so we would be more ready for other people to contribute but that’s obviously not going to happen. He is now busy with his PhD and can’t help us much more. I’m posting this after talking to him.
So we need some Plone people who can do this for us. I personally have no experience of Plone so I wouldn’t trust me to do it even if I taught myself about Plone.
What to do
The first aim should be to make it possible for anyone to install it locally to test it. That includes instructions (hopefully simple) so that even I could deploy it and test it. I would like some reassurance that this “buildout” system is the best way to do that. Experienced Plone sysadmins should know the answer to that.
Then I would like the Plone experts to generally check it over and point out anything that we could do better.
Then we can deploy it as a beta on gnome.org, fix the last few bugs, and see if it’s good enough.
NoooooooooooooooooooooooooooooooooooooooO!!
Please don’t make gnome.org use plone!
The current design, url structure and everything else are great!
Hi Murray, just a reminder: feel free to add the w.g.o roadmap in http://live.gnome.org/RoadMap :-)
IMHO Plone was a bad idea for the site for this very reason.
I’m not sure why Plone is such a bad idea; it’s not different than anything else new to others. I’d be willing to invest sometime but I don’t have much to give. Maybe I could set aside a couple of hours every weekend or so. The README is pretty much the standard way to build a Plone site but i’m not sure if http://www.gnome.org is being proxied via apache or what. My day job involves Plone pretty extensively as I’m replacing pieces of nymag.com with it.
Lucas, I didn’t put anything on the RoadMap yet because the whole schedule has become so completely unpredictable. I think I’ll still wait a bit to see how it turns out.
For the plone haters: I hate all CMS systems that I’ve ever seen. They all seem to force ridiculous structures and processes on us, while rarely offering the features that most people actually need. However, several GNOME people did a quite thorough review of several candidate systems and Plone was chosen (I believe) precisely because we seemed to have enough experts people to help, and because those experts were ready to adapt Plone to meet our needs. For instance, to add a real po/gettext-based translation system for content. We could be in the current situation with any other system, and we’d maybe have hit problems sooner or never even have got this close to deploying.
Christopher, your help would be very welcome even if it’s just a few hours per week. Please do subscribe to gnome-web-list and introduce yourself:
http://mail.gnome.org/mailman/listinfo/gnome-web-list
Cheers.
Yes, buildout is the way to go nowadays.
Adding the url for the buildout checkout would be a great addition to the readme, btw :-)
Reinout
> Adding the url for the buildout checkout would be a great addition to the readme, btw :-)
Reinout, I don’t know what you mean. Could you be more specific, please? (It’s OK to assume that I am an idiot with this stuff).
I just meant that it would be great to add the following to the readme:
To checkout the buildout, use “svn co http://svn.gnome.org/svn/gnomeweb-plone/trunk/wgo.buildout“.
That saves people the hassle from translating the url of the viewvc to the real subversion checkout url.
Reinout
If you want someone to install a zope/plone site you only need to send the products folder and the data.fs file
Good luck!
Garito: not completely. Plone 3.0 starts to move towards regular python, so it increasingly uses the lib/python/ directory in addition to the Products/ directory. So that’s one .tgz extra.
And sending over a .tgz for developers (who are the ones that are needed now) is a bad idea. You want the officially distributed tgz files (of plone and addons) for repeatability. And the product you’re developing on ought to be in svn.
Buildout takes care of that in a tidy way.
Reinout
I don’t know what are you talking about (I want to say if my english want: no estoy muy al caso de lo que estas diciendo)
I don’t know what kind of code needs to put outside the products folder but if you develop archetypes correctly you only need to send the products folder
data.fs is the database itself
This is very interesting, I made the same choice recently to swap to being a Plone developer for my latest client. I’m not sure if I’ll be any help to the gnome web team but I’ll be interested to see the results. I guess I’ll join the mailing list and see if there’s anything I can do.
Garito: I know perfectly well what I’m talking about :-) The latest and greatest plone 3.0 *does* have a lot of code inside lib/python. Even archetypes development can be done inside lib/python.
lib/python (for regular python libraries) is also possible, mostly, in plone 2.5 but in plone 3.0 it has really taken off. The good thing is that it makes plone more python-like. The bad thing is that there’s a bit more install effort needed (which buildout takes care of for you, btw).
Reinout
Since both GNOME and Plone are projects I really really like (obviously ;), I’m happy to take part in an advisory role on the lists.
Ramon — who is heading up the current wgo initiative — has recently gone through a lot of personal changes, moving etc, so I’m not surprised that things have slowed down a bit. He’s on the right track, however — and has earned the respect of the Plone community as well as approaching things in the correct way.
There are many shared goals between the Plone and GNOME communities here, especially when it comes to the multilingual parts, so I believe there is lots of potential for cooperation here.
See you on the lists. :)
Thanks, Alexander, that should be a big help.
Ditto. I really like GNOME (since back in the day ;-) and I’m a Plone Consultant, so I’d definitely be willing to devote some time to this (possibly in early 2008-ish?).
Alexander (and Alex), please do introduce yourself on gnome-web-list. I’m waiting for people to arrive there before attempting to organize them a bit.