Re: What about WordPress or something similar?

[Thomas Wood <thos gnome org>]

Stefan Rotsch wrote:
> Hi all,
>
> following the call for volunteers on a.g.o., I took a look at the ToDo 
> list and thought by myself: Why won't they use an already existing 
> platform instead of "reinventing the wheel". Why not starting with 
> WordPress or something similar?
>
> Some of my thoughts/reasons for this suggestion:
>
> - Stability: WordPress & Co. are running in thousands of installations. 
> There are communities contributing and fixing bugs (maybe this could 
> prevent a.g.o. from being hacked again)
>
> - Features: WordPress & Co. e.g. generate RSS (and Atom) feeds for 
> nearly all and everything, and they usually don't break the feeds
>
> - Accounts:  CMS offer facilities for managing user accounts, containing 
> options for dealing with lost passwords (-> ToDo) or re-entering 
> passwords when changing them (also on ToDo)
>
> - PlugIns: New features could be easily implemented, and there are 
> already lots of plugins available (for example, a "rotate banner images" 
> plugin for WP comes to my mind (on ToDo))
>
> Switching to a different platform will cause some work, but I don't 
> think it would get too hard. Converting the a.g.o. look to a WP theme 
> could be easily done in a couple of hours, for example. User accounts 
> will be supported "out of the box", as comments and RSS feeds would be. 
> Voting, Uploads, etc. could be realized using plugins, maybe by 
> transforming the existing php code into plugins for the used platform.
>
> What do you think?

There are several issues I would raise with this idea:

- Migration: How easy would it be to migrate the existing users, data, 
files, screenshots and the rest to a CMS? I think this would be 
especially hard for users, since I can't see anyway that passwords could 
be transferred (unless the new CMS supports exactly the same method for 
storing passwords as art-web currently does).

- Features: We would still have to do a lot of work to implement the 
plugins needed to handle art-web specific data and pages. There is a lot 
more metadata associated with backgrounds and themes than is obvious. 
Also, themes and backgrounds are not handled the same way as you might 
expect (backgrounds have many more requirements than themes). A third 
type of item may also appear soon, in the form of collections of other 
items.

- Control: If you move to a CMS you loose control of your source code to 
a certain extent. I don't particularly have any motivation to read and 
learn another CMS's source code just to make some minor (but possibly 
vital) changes. There are several people already familiar with the bulk 
of the art-web code, and it is probably much easier to get to grips with 
due to it's (relative) small size.

There are also factors out of our control, such as restrictions placed 
on us by the gnome.org system policies and procedures (this is the 
reason it has not been possible to implement a file upload system until 
recently).

Not to mention the time and effort people have put in to coding art-web. 
It would be a shame to throw this all away, so I personally would not 
like to see art-web replaced with an external CMS.

-Thomas