You already know my reStInPeace soft, maybe you can give PyK! a try as it is an evolution from the first one and can help you writting good articles for PyCon 2008 in reSt format for example.

You’ll find it here, and of course it’s free !

http://kib2.free.fr/PyK/

Cheers,

I should ignore them as well, but find myself unable to do so ;-).

I know about the requirements page and have withheld from commenting. I am also on the europython-improve mailing list. The requirements look almost identical to the ones that AMK wrote up in 2004 for PyCon, and were the basis of PyCon-Tech (origionally his project). They are missing way to much to be considered real requirements. Registration and Peer reviewed proposals (missing in the requirements!) will be the hard parts for PyCon-Tech.

At the end of the day registration is the #1 thing to get working well. Everything else can be done with e-mail and a wiki. This does not scale well, and I am very glad to have a software solution in place for PyCon 2008. I do not believe we could handle the load otherwise.

Thanks for the history behind the EuroPython efforts. I completly agree, you get out of the system what you put into it. I have not used Indico myself, but I have heard comments on both ends of the spectrum about it. I have also heard the exact same comments about PyCon-Tech. I even had Jim Fullton yelling at me durring the sprints about the software :-). (I was not upset at all, he was right to be mad, and all I could think was ‘Wow! Jim fullton is yelling at me!’)

I also agree that the reviewers need to use something that they are comfortable with. I fear that parts of the proposal submission software have become a barrier to submitting talks (even with the 140+ submissions). That is the worst possible outcome for a submission system, and one that I hope to get some good feedback on.

I have also been talking with someone from a rather large international conference about forming a more centralized place for discussing conference management and technology. Hopefully something will come of that.

Oh, there is no question that my management skills for PyCon-Tech thus far suck eggs :-). I know what I need to do to fix things, but life keeps getting in the way and I just don’t have the time. I am hopping that as more clients come on board (there are 3 other conferences looking at it), that I will get my butt in gear. I was even considering giving a talk ‘How not to run an open source project’, but there is still a question of wether I can make it to PyCon. If it happens it will be an open space talk.

And from a reviewer’s perspective, I have just ignored the tags. =)

First of all, there’s the complexity in getting started, and I think the EuroPython infrastructure has some inertia in this respect. We used Indico for 2006 and 2007, mostly because CERN developed Indico and preferred to use it when they were running things in 2006, and we only continued to use it with CERN’s support in 2007 because no-one was inclined to deploy anything else. Indeed, we didn’t even get as far as to deploy Indico anywhere else, and that really caused inconvenience for users later on. But it really takes people from the target audience to get their hands dirty, which doesn’t always seem to happen.

Then, there are the perceptions of the target audience for whatever does get used. Indico makes itself a convenient target for criticism because it’s meant to be able to do almost everything, so if the reviewers don’t want to use it for talk selection and scheduling, preferring someone’s kitchen table instead (which is up to them, of course), then Indico gets a reputation for being inflexible or whatever. If users find the registration too confusing, it’s because Indico’s registration form isn’t shiny enough, and “why didn’t you do a Django application with AJAX?” (As if I had loads of time to burn and decided to spend it playing with Indico.)

In my experience, as the scope of a tool or set of tools increases, the number of user types who have to use it also increases and, inevitably, there are then fewer users who have good things to say about the experience. That’s not necessarily the fault of something like Indico, however: as David says, there are lots of social aspects involved in organising a conference, and it seems too easy to me to just blame the tools when humans were responsible for choosing them in the first place.

So, I’ve come to believe that each special interest group is best served by choosing their own favourite tool, even if the infrastructure might be more complicated as a result. I don’t care how the reviewers choose the talks (and was actually more concerned about there being enough talk submissions in the first place for EuroPython, which is another story), and if they don’t manage to arrange a nice Web-based solution and end up doing it by having proposals arrive by e-mail which they then print out and pin up on a noticeboard, or whatever, then as long as it doesn’t impact the registration system and the general publicity, then so be it.

Especially in the light of “democratizing” conferences, the “high volume” participant-facing technology is the most important part before the event itself: information about the conference, where/when it is, who it’s for, how to register, where to stay, who else is going. I’d go as far to say that all other “virtual” processes are “luxury processes”, but I’ve probably left something out of that list. Sadly, when I’ve read anything about conference technology development, the people most excited about it seem to be more interested in the exotic stuff: geotagging, Web 2.0, social networking, tag clouds and so on. My embittered personal feeling about this is that it’s like designing the ultimate spacestation before inventing the rocket (ie. getting people to register in the first place).

Perhaps there should be more experience sharing between the conferences, though. It needn’t be all horror stories and complaints.

It’s now too late to change course for PyCon 2008, but it’s probably still worth throwing the Wiki page into the mix to help focus discussion of software for later conferences. Ideally, there would be a central mailing list for this discussion - I don’t think it helps to have separate discussions on EuroPython and PyCon-specific mailing lists.

I agree that the tag cloud would be useful if I actually gave one fig for what the hot topic was in anything. So far I have been severely disappointed in the blogosphere (or whatever) and every news source. I just don’t care about what all these other people care about. Not once has a hot topic interested me. I might be getting old and out of touch, but I think it has always been this way. Even in the middle of the Harry Potter mania, I was completely uninterested in 99% of the Harry Potter anything out there.

So to date, the clouds have been the worst interface for me personally, and as such I want nothing to do with them. If I find one aggrigator which I find useful, I might change my mind, but after 4+ years of the cloud, I am not holding my breath.