Monday, November 06, 2006

EclipseCon2007 is (Unnecessarily) Controversial

I really liked the EclipseCon2006 process. I wish they did not change it this year.

A controversial aspect of the EclipseCon2007 program is that, for the most part, a single person (i.e. chair) selects the content of each track. Given that the program committee is pretty distinguished, this seems like a good idea. So why is it controversial?

It's controversial because a chair is allowed to make a submission in their own track. Conflict of Interest, anyone?

How can the chair ever convince the community that the selection process will be fair and balanced? It's virtually impossible. Case in point: submission 3676. As the plot thickens, said submission was made past the Nov 1 deadline. Do deadlines matter? These are all fair questions.

I am going to tivo this one to see how it plays out :)


Gunnar said...

Oh that's sad. There is clearly a conflict here. But there are also other issues. For example, the Rich Client track currently doesn't have a slot for a Long Tutorial. How do we want to teach beginners how to learn Eclipse RCP?

pookzilla said...

In the words of Wanda from Corner Gas:

Scorch Pow!

Neil Bartlett said...

I'm sure the conflict of interest will be resolved amicably.. while I'd like to do my tutorial, I'm sure Peter and Thomas' would be great as well

The problem appears to be that the allocation of tracks takes no account of the supply of material. So, there are now 5 tutorials competing for one OSGi slot, and nothing at all for (say) Java development! Perhaps this should have been done as last year: all of the submissions go into a single pot, and the right number of tutorials come out the other end.

If that results in a heavy OSGi weighting, surely it just reflects a high degree of interest in OSGi this year.

Also, as Gunnar said, why no tutorial for RCP this year?? Doesn't make sense.

Wassim Melhem said...

There are several problems with EclipseCon planning this year that I plan to comment on. The RCP tutorial issue is one of them. This post is only the first of a multi-part series.

Also it is worth noting that submission 3676 was only used for illustration purposes. There are other examples of conflict of interest already in the lineup.

Bjorn Freeman-Benson said...

Wassim, I look forward to your series - I will avidly read them for ideas on how to improve EclipseCon. In the meantime, I wrote a direct response to your post over on my blog.

Anonymous said...

Actually, submission 3676 was at least in part a response to some pushing from me. Like the lottery, you have to play to win (Lotto 649 is $36M now!). In this case there has be to a submission if the community's voice is to be heard. So that submission was in support of democracy. In the end Peter, as with the other PC members, is a reasonable and fair person and will do what is right for the attendees.

Which brings me the the next point. The job of the PC is to put together a program that attracts and informs the attendees. Lots of submissions in an area indicates alot of interest in presenting on a particular topic, not necessarily alot of interest in hearing about a topic. Of course, WRT OSGi, who wouldn't be interested?! OK, I'm biased. But seriously, it about the attendees not the presenters.

Case in point, the RCP tutorials. I'm sure there are lots of people wanting to hear about RCP but in talking with the usual suspects from the Eclipse team most were doing other things (like being on the PC) and wanted to give others in the community a chance. Note that there may well be some late submissions that would appear to be conflicts but are in fact the PC, in an effort to create a balanced and exciting program, arranging for speakers to present on particular topics. That is their job.

Peter Kriens said...

I am a bit puzzled about the "complaint". The submissions have a voting button and as Bjorn stated, everything is done in the open; with this transparency abusing given powers will be very hard.

I see my role more as moderating the selection process than as making the final selections. Despite that, I was hesitant to register the tutorial because of this reason, but someone convinced me it should at least have a chance to be chosen by the community.

About the number of slots, the process this year seems very flexible this year as long as it helps towards making a great conference. I could ask for more tutorial slot but I am not sure that would be such a good idea. I think it is better to get lots of attendants for a few tutorials than a few attendants at many tutorials?

Kind regards,

Peter Kriens

Neil Bartlett said...

This comment is really to Bjorn, but he doesn't like comments on his blog... also this is in response to Jeff and Peter.

You're certainly right that it's all about the attendees rather than the presenters. But I think it's possible that pre-advertising the allocations for each track was a bad idea, versus just putting all the submissions in a big pot and allowing the weighting of each track to fall out of the voting patterns.

Here's my theory: suppose somebody has a great idea for a tutorial on OSGi - they might have looked at the existing submissions on 1st Nov and seen four competing for a single slot, and been intimidated or discouraged from making their proposal. However if they were not aware that there was only one slot, then they might well have made their proposal anyway. The community would then have had a greater choice.

I think this is especially the case because now, after the submission deadline, Bjorn has hinted that the allocations are in fact quite open to change (which was certainly NOT clear previously).

The two solutions I can see for this are (1) less openness, ie don't publish the number of slots. But that goes against the Eclipse grain. Alternatively (2) don't fix the number of slots until the proposals and some votes are in. Hmm isn't that the way Eclipsecon 2006 worked? :)

I guess it was my comment to Peter's submission that kicked this all off. I didn't mean for it to snowball like this, I just wanted him to clarify his role! All in the interest of openness, you understand (... but admittedly I still do want to do my tutorial!)

Anyway, bring on the public vote.

Anonymous said...

This is also a comment for mainly Bjorn, but as he doesn't like comments on his blog, answering it here makes sense.

The submissions page clearly states:

"Important Dates

November 1st - Tutorial proposals must be in."

If the process was open, then this is what should be followed. It doesn't say that "unless we decide that other tutorials should be included afterwards". We're all supposed to play by the same rules.

Let me just say: I'm not against a program chair submitting their own proposals per se. However, I don't think it was particularly necessary (given the number of tutorials proposed already) and the fact that it's a recycled proposal from last year (compare with this year). From the submissions page:

"We strongly suggest that this be original material because, to quote a source, "I'd hate to attend a tutorial that uses the same content as I can get online.". Submission deadline November 1st."

So, is it one rule for everyone else and one rule for EclipseCon organisers?

Gunnar said...

Just two short comments.

Al, you cannot have only new tutorials every year. You also need to attract beginners (new attendees) and this requires repeating some content year for year.

Bjorn, too bad that you have comments disabled on your blog. But you made a wrong statement about the 2006 PC. As a matter of fact, we had a public voting in place through EclipseZilla (Bugzilla flags) and no single PC member was able to vote a submission in or out.

RefuX said...

>This comment is really to Bjorn,
>but he doesn't like comments on his blog

I also find this flabbergasting, I know Bjorn has stated there are alternative places to post responses... but lets face it, if I want to comment on one of his blog posts and I can't, am I going to go to the extra effort to dig out his prefferred method of communication... uh... NO!

p.s. I only accept responses via pegion.

Bjorn Freeman-Benson said...

Alex, I'm surprised that you are suggesting that we not allow a submission X from a PC member just because there is a submission Y from a non-PC member on the same topic and _you_ don't see X as necessary. Wouldn't that be the same as "one rule for everyone else and one rule for EclipseCon organisers"? Shouldn't we just allow everyone to submit that then, in an open and public way, choose amongst the submissions?

Alex, it's true that we want original tutorials, but do remember that it's the speaker and the presentation that people are coming to see - they can get the slides and the articles from a zillion places on the internet. So just because the slides have been given before doesn't mean that the tutorial is the same - the speaker is a big part of the equation. (Just pointing this out.)

Neil, Alex, Gunnar, Refux, can you explain how I can make Blogger moderate all comments?

Anonymous said...

Bjorn: I'm not objecting to a PC proposing a tutorial; I've made that point clear in my longer post at EclipseCon that it's not about whether the proposer is a PC or not.

What I do object to is not everyone playing by the same rules. Either a deadline is a deadline, or it isn't. If anyone else proposes a tutorial, talk, or panel late later on, they can point back to this and say "But you said deadlines don't matter". If Peter had submitted his proposal a few days earlier, this conversation wouldn't have come up. But, by letting this through, you've effectively invalidated the deadlines for every other talk because it's clear they have no part in the EclipseCon process.

Frankly, I hope that the tutorial that I proposed doesn't get selected. It's not about the community any more, or what's interesting.