Andre- (regarding the commons and similar ideas)

I want this, I have wanted it for many months. I have made proposals to implement it, and all I need is to get a wiki set up.

As I've explained on the proposal talk page, I think this project needs proper planning so that we avoid shooting ourselves in the foot in one way or another. But let me just assure you that I completely understand your frustration. I have here 4 CDs full of interesting material, historical images of different sorts, which I can't wait to upload, but I have no immediate use for them on Wikipedia, for some of them I don't have precise source information (although the images are all public domain), etc., so this would be perfect for a media repository like the one we both want. And part of me says "Why not set this up now?".

My main concern is doing things right from the start. Evolutionary growth is a strange thing. You may end up with useless appendices, or with a chaotic structure that somehow works but which nobody understands. And evolutionary changes must always build upon what's there. A third arm is not going to pop up out of nowhere just because you need it.

In our context, I'm worried about the following:

- This is our best opportunity to get single sign-on working, a feature which we desperately need. Because the Commons effectively requires access to a shared database, this is a natural extension. But starting with a separate login system makes single sign-on a *separate* goal. This is not just a technical issue of yet another redundant user table. It's also a social issue of not being able to use the synergy from the launch of the commons to promote single sign-on (which will likely have some initial hurdles to overcome) and vice versa. It's difficult to generate excitement about small, evolutionary steps.

- The initial edits on a wiki lay the foundation of what that wiki will become. If just a few people get involved in this project, because it offers no really cool, exciting possibilities, then the project foundation may well not be as solid as it could be. For example, people may decide to create image categories and upload requirements in the first two weeks. This structure will then become harder and harder to change as it seeps in, and when we add all the new cool features which attract more people -- a better upload form, transparent use of commons media from all wikis, single sign-on -- it may already be too late to quickly and effectively fix certain problems. Too much may have grown into the structure already.

- More generally, any and all things which we do on a quickly set up wiki will have to be upgraded when the wiki itself is. If we decide at some point that we need certain information on all image description pages, we have to add that information to all existing ones. If we add some fancy new feature for building image galleries, we have to convert our existing ones. We do this all the time, of course, but that is born out of necessity. Here no necessity exists. We can afford to wait a couple of months until we have everything we need.

These concerns outweigh my desire to have something usable as soon as possible. As soon as MediaWiki 1.3 is final and all the wikis are upgraded, I and hopefully others will start working on this project. This *is* going to happen, but I want to get it right from day 1.

Regards,

Erik