Terça-feira, 12 de Outubro de 2010

An inside look at candidate selection

While organizing Codebits can be a very rewarding experience, selecting candidates is most definitely not.

A small group of moderators has been going through all the applications, in many cases more than once; we asked them to tell us more about this process.

The first thing we were told was, as expected, that it takes a lot of work. Reading through almost 1,000 bios is not an easy task, and certainly not something you can do during the course of a few afternoons and expect to still be digesting the information properly.

The second thing was that it is not, in any way, a rewarding task. Mostly because of the complains. And we're not talking about candidates who weren't approved.

It seems that many people have been complaining on how some of their acquaintances have not been accepted.

A typical email of these would say:

"I can't believe you haven't approved candidate X! This is an outrage and clearly you have no idea of what you are doing. This person has been the main supporter of Y in our country, has managed to single-handedly organize community Z, which now has like a bazillion users and, on top of it all, is a core Q hacker! Just go to their profile and click on approve, OK?"

Usually the team goes to the mentioned profile only to find a bio such as:

"Founder and CTO at Something Technologies.

I attended SAPO Codebits 2009 and simply loved it.

I am really funny, here's a joke.

If you want to know more: http://www.linkedin.com/in/myprofile"

We usually reply to those complaints explaining that all the information they used to argue that the person should be accepted is NOT present on the bio.

Taking this "ficticious" bio - which truly is a mix of several bios we've seen - as an example, one can note how it doesn't describe any technical achievement or knowledge. Founding a company doesn't really say anything; we hope it does, but we really have no way of knowing; assuming that the moderators know the company is just a huge assumption. Out of curiosity, there are currently 72 "founders" on the system.

Saying one attended a previous edition of Codebits and that they liked it is not really something that will make one go "OMG we just have to approve this person!" It's something nice to say, as is adding some humor, but it won't get you there if there isn't much else.

Also, we currently have 76 profiles directing us to LinkedIn (not necessarily a bad thing, just not good enough on its own). 620 contain URLs; usually a bunch of them. Following and reading all of them is just not feasible.

But there's a lot more to be said about candidate selection.

There's probably a lot of information we have access to that the candidate is not aware of, the major one being if the candidate presented a project last year; that's a big plus (it should be noted that first time participants are never rejected because of this).

An example of this would be participation: the amount of thumbs up/down the candidate has given, or at least whether they have taken some time to peruse the talk proposals.

Another example would be the last time the candidate visited the website. Just this morning there was a complaint (coming from someone at SAPO) about someone who hadn't but apparently should have been approved. The bio didn't mention half the things contained in that complaint, plus the person had received a short bio notice about a month ago and hadn't got back to the website ever since.


There are no miracles here. No such thing as mind reading.

Moderators can only go so far with the information they're provided with, and they don't claim to be all-knowing.

There have also been a couple of complaints comparing bios that haven't been approved with others that have.

Some of you may have guessed it. We do deprehend a lot of information such as the degree of interest and commitment the person seems to have. We do this from the accesses to the website, actions such as rating talks, and more.

In some cases, as mentioned in earlier posts, the fact that some people registered and applied earlier has also counted on their favor. As would be expected, the process of candidate approval was much more lenient when there were only 400 candidates and no clear idea of how many would show up later.

Let's get to a brighter note.

We do understand that it's difficult not just for us, but also for everyone still rewriting their bios.

We have managed to take the limit of attendees up to 700, which means we currently have 50 more seats (650 have already been approved).

These final 50 attendees are likely to be selected between today and tomorrow.

We need your help.


Please help us minimize the chance for errors. Make sure your bio contains relevant information. Make sure it is descriptive, explicit, creative, technical, hardcore.

publicado por jac às 11:26
link do post | comentar
4 comentários:
De Marco Sousa a 12 de Outubro de 2010 às 13:33
Thanks for the insight, I just updated my bio the best I can, fingers crossed :)

De Sérgio Laranjeira a 12 de Outubro de 2010 às 14:03
First of all, I wan to congratulate you for the selection process. Can't even imagine how hard it has been.
I understand that some people complain, but maybe they should first read the bio of the person that hasn't been accepted yet. I read some that's just like you say, they write something but don't say a thing. Keep up the good work.

De Waldir a 12 de Outubro de 2010 às 21:08
Kudos for your hard work. You have a big point about the unjustified complaints.

Just make sure you don't bend over backwards to accept borderline cases and then let people who've had good bios all along slip past the limit.

I know at least one case of a terrific bio (by all your standards) which has been under review for almost 3 weeks now, while others less impressive (imho) got accepted meanwhile... so be careful, please :)

De fLaSh a 13 de Outubro de 2010 às 17:40
I'll wait a little to see if I get approved!
Would like to go this year.
I am with nothing to do, no occupation...

Please review my application!

Keep up the good work :)

Comentar post