blog.Resource

News-Feeds:


RSS 2.0
RSS 0.91
RDF
ATOM 0.3
August 21, 2012

T3CON12DE Paper Voting results clarification

Category: Christian Mueller

The why and how of paper voting and conference session planning.

This text was originally posted to Google+ and linked via a Twitter message to those that raised the questions hopefully answered below, as others seem to have the same questions and missed this Twitter/Google+ part I share it again here.

After Roberts post (https://plus.google.com/110277469483406332394/posts/QX5gpXMNqnm) about the paper voting and decisions for T3CON12DE there still seems need for some clarification.

Little disclaimer: The views in this post are my personal opinion and not necessarily shared by the TYPO3 Association, the organisation team of T3CON or anyone else. That said some more insight on the whole topic:

I totally agree that the voting process should be transparent and I personally have no problems to have the votes for my talks publicised (and if you read on you can actually see them) but I really believe that because we didn't announce we would publish voting results beforehand it wouldn't be fair to publish them now because maybe someone wouldn't have entered their talk if they had known before. So I fully support the decision to not publish the results for this year.

And that is exactly the point as this was the first year with public voting there are things to learn and things to improve. For next year I would announce beforehand that all voting results would be disclosed and also add the possibility to (public) comment on talks and (hopefully) get feedback from the speakers during the voting process.

Now I will share some stats about this years voting without actually publishing the results in detail:

First there is a little flaw in the voting that I would change for next year which is that people could actually not give any vote to a paper so that the amount of votes on each paper differs. Because of that we only took the average (voting divided by amount of votes for the paper) into account. The average could range from 1.0 with all given votes +1 votes to -1.0 where all given votes would have been -1 votes.

Paper which most people voted on (not with most points) had 96 votes.
Paper with highest average +0.78
Paper with lowest average -0.32
Accepted paper with highest average +0.78
Accepted paper with lowest average 0.0 (an invited external speaker) next accepted had 0.14 average.
Actually from the highest ranking paper to about 0.24 almost all papers got accepted.
Those in the high rankings that didn't get accepted were not because we had several talks with similar topics and picked one or two of them and rejected the rest to have a broader range of topics.

Now my personal results:
Developing a TYPO3 Phoenix Website = 0.53 (accepted)
Phoenix Website Integration (Tutorial) = 0.37 (accepted)
FLOW3 Search integration with ElasticSearch = 0.19 (rejected)
This one got rejected because there were more interesting topics with a bit lower score and it would have been the third talk I would have been involved with which we found generally a rather bad idea.

To close this long explanation anyone who wants to know her scores can contact me from the email address she used for submitting the talk to get her votes (please include your email address) (males can ask too, just wanted to do something for gender equality ;) ). But please just ask if you really need to know because I do that in my spare time and looking up 50 talks takes time, same goes btw. for looking up all those stats and writing this post.

Some more numbers that I added in comments:

 

Amount of submitted talks is 92.
Amount of accepted talks is 40.
Amount of talks with an average voting result of 0 or below is 23.

For the higher ranking talks that got rejected I will compile a list with a short explanation why they got rejected and add it here as this was a requested information too.

TYPO3, Faster, Easier - Performance Updating by Michael Cannon was rejected because we already accepted another talk with a similar topic.

Responsive TYPO3 by Sven Wolfermann was rejected because he submitted two talks around responsive and we accepted the other one and because this topic was overrun in general.

How automatism can help you and your customers by  Georg Ringer was rejected because we already accepted another talk with a similar topic.

Key concepts in quality assurance for TYPO3 sites by Ingo Gächter was also rejected because the key points were already topic of other accepted talks.

Same again for Responsive Design: Delivering for Desktop & Mobile by Peter Pröll, we had other talks with the same topic that were already accepted.

That is just an excerpt, I don't want to comment all the talks... In the lower rankings that followed we looked out for external speakers which were invited to the event and a few "must have" talks, like Keynote, Talk about TYPO3 Association, so that others with similar ranking were sorted out.


comments

comment #1
Gravatar: Jigal van Hemert Jigal van Hemert August 21, 2012 20:48
There was no prior consent of (all) the people who submitted a paper for the voting either. With the organisers deciding "spontaneously (...) to do a community voting" all the submitted subjects and descriptions were published.
If someone has copied that list and compares it with the papers that are accepted it's easy to see which papers were rejected. I can imagine that someone has objections against publishing the submitted, but not yet accepted list as much as they would have objections against publishing the vote results. So, why publish one and not the other?

comment #2
Gravatar: Christian Christian August 22, 2012 18:20
Hi Jigal,

well yes but you still don't know the individual scores and that is what matters I guess. But I just wrote my post to add to transparency. I understand and I am in favor of the decision, but I might not have taken it myself, so I won't start to argument to defend it. I just accepted that this is the way to go for this year and wanted to bring you as much information anyway.

Sorry, comments are closed for this post.