# Should Math.SE self-evaluation continue?

Now that the recent round of Math.SE self-evaluation is over, several related questions arise:

• will Math.SE users see the results of the survey? (answered by Willie Wong)
• will any action be taken based on the results?
• should the self-evaluation be repeated every 90 days? According to Anna Lear,

If these evaluations prove too distracting or not particularly helpful on graduated sites, we'll turn them off. In the meantime, the queue should pop up for a week every 90 days.

And indeed, the screenshot of results says

Next eval begins May 3 at 3:00.

I think we should either figure out how to use this information, or stop collecting it.

-
The moderators have access to the results, they can post them whenever they like. They're at math.stackexchange.com/admin/review/community-eval-stats –  Mad Scientist Feb 10 '13 at 21:33
My opinions on your question: I vote for making the results public. I didn't find the self-evaluation particularly useful. (Maybe it could be changed a bit, to make it more useful, e.g. to improve randomly chosen questions or to generate more votes if this functionality was available, but this would be another tool) –  Julian Kuelshammer Feb 10 '13 at 22:10

## 3 Answers

I'm sorry, I didn't see this discussion as well as the one before it until just today.

As others have mentioned, these evaluations were experimental on graduated sites. The philosophy behind them is potentially sound - just because a site leaves beta doesn't mean it's keeping up the standards and activity that makes for a successful site.

However, I have reviewed a bunch of the results and responses from this round of evaluations and in practice it turned out that graduated sites are doing as well as we would like them to and as I'm sure they would like to as well.

So we have disabled this review queue on non-beta sites. Thank you everyone who has participated in this round. You've helped us make and validate this decision. :)

-
This reminds me of an episode of the first season of The Big Bang Theory, where Sheldon gets fired and starts experimenting with cooking eggs. He later finds out that we already got the best eggs possible, and the field is a dead end. :-) –  Asaf Karagila Feb 13 '13 at 2:34

No, the self-evaluation of Math.SE should not continue.

Most users here, and that certainly includes me, are quite specialized in their knowledge of mathematics. The voting system, with all its many drawbacks, leads by and large to people voting on questions in areas that interest them and that they are competent in. As a result, these judgements are less noisy then the one used in the self-evaluation and render the latter superflous.

-

I don't see why we shouldn't post the results. So here they are (sorry about it being in picture form, I cannot see an easy way to copy-and-paste).

If you are curious about the questions themselves, the question numbers are, in order appearing above:

• 262508 Limit computation of $(e^x+x)^{1/x}$ as $x$ approaches zero
• 262778 Compact linear operator from $L^p(\mathbb{R})$ to $L^p(\mathbb{R})$
• 260096 Find the coordinates of a point on a circle
• 256658 A question about complex numbers
• 252513 Planar and non-planar graphs
• 260674 Complex analysis (periodic function)
• 251201 Uniformization Theorem for compact surface
• 264211 Congruence relationship used for primitive residue classes modulo n result
• 251990 Empirical distribution vs. the true
• 258919 Logic about systems?
-
I am not sure how this table of numbers would be useful for the community. Ideas are welcome! –  Willie Wong Feb 11 '13 at 12:26
Well, if anyone wanted to know if my answer to 251201 needs improvement, they could have asked me. It's actually pretty lousy. In particular, I later found that Hubbard does not cover the non-orientable case, and I don't see how to deduce the required form of uniformization from looking at the orientation cover. // Overall, I agree with Asaf when he compared this process to taking a cup of water from the ocean to judge its quality. –  user53153 Feb 11 '13 at 13:38
I agree with 5PM about agreeing with me. "It's a good thing." –  Asaf Karagila Feb 11 '13 at 21:38
Some of the judgements are very strange; it’s hard to see how the answer to 264211 could be improved, for instance. –  Brian M. Scott Feb 12 '13 at 3:40
Not having participated in the evaluation, it is not clear to me whether it was the answers that were being evaluated, or the questions. Would someone fill me in on this? Thanks. –  Gerry Myerson Feb 12 '13 at 23:48
@GerryMyerson Both, but mostly the answers. This evaluation used to be done by Community Managers "by selecting 10 random open questions ... to see how their answers compare to other sites on the Internet." Recently it was converted into a review queue, but with the same purpose: "Are the questions of high quality? Are the answers better than what someone can easily find through Google?". Overall, trying to gauge to what extent the site "makes the Internet a better place"... –  user53153 Feb 13 '13 at 2:51
@GerryMyerson: Note also that the definition of 'excellent' and 'satisfactory' wasn't at all how I would use the terms. The standard required to call something 'excellent' as defined here was so crushing that I found myself voting mostly everything 'satisfactory', although simply calling it that felt wrong. –  gnometorule Feb 14 '13 at 0:49