tea.mathoverflow.net - Discussion Feed (MathOverflow 2.0 final sandbox testing)

Asaf Karagila comments on "MathOverflow 2.0 final sandbox testing" (22653)

2013-06-24T06:41:49-07:00

@Gerald: I don't see this as moderation. It was something initiated by the user, rather than something that someone saw and said to themselves "Well, that shouldn't be here..." and then acted on the issue.

In particular site moderators cannot delete accounts across the whole network, and it seems wasteful to ask each site to delete their own account. Moreover after the incident where a high reputation account were removed site moderators cannot delete accounts with "enough" reputation (I'm not sure how much is enough, though). So this sort of task might has to be carried out by a network moderator.

geraldedgar comments on "MathOverflow 2.0 final sandbox testing" (22652)

2013-06-24T06:37:45-07:00

network-wide managers do very little moderation

I seem to recall one case on MSE where an account was closed, and after a question in Meta is was found that it had been closed at the user's request (all his accounts in the entire SE system) by a network-wide manager.

Noah Snyder comments on "MathOverflow 2.0 final sandbox testing" (22645)

2013-06-23T12:26:23-07:00

While SE is sorting out the new better front page behavior, what's the harm in MO having our old behavior instead of the other old behavior? The current SE2.0 default is bad in general and particularly bad for MO. I don't at all understand the claim that the front page that registered users use doesn't consist of typical questions.

Otherwise the new site looks great. Feels familiar but also looks better.

Todd Trimble comments on "MathOverflow 2.0 final sandbox testing" (22644)

2013-06-23T11:35:46-07:00

Thanks for the reassurances, guys.

Asaf Karagila comments on "MathOverflow 2.0 final sandbox testing" (22643)

2013-06-23T11:33:29-07:00

Todd,

Certainly PM didn't mean Prime Minister. I suspected it may have meant programmer, or polynomial mesosaur.

I can attest that on MSE the network-wide managers do very little moderation, if at all. I do my best to keep an eye out for these things, and I can't recall a single incident. I believe that the situation on MO would be the same.

quid comments on "MathOverflow 2.0 final sandbox testing" (22642)

2013-06-23T11:31:20-07:00

@Todd Trimble: PM is product manager see the page I assume to be jtunnell's userpage on meta.SO (but this was also the user answering a question on meta.SO on the frontpage so this seems like a very save assumption).

Who the moderators will be for MO is defined in an agreement Anton Geraschenko mentioned some time ago in the transition thread; in brief and from memory, we keep the current ones and if the need arises there will be elections of new ones in consultation with the current ones. (In general for the SE sites I think moderators are elected by the respective communities, too.) There are some general moderators (SE employees in part or all of them) around, but judging from math.SE (while I am not active there I passively follow their meta since some time in some detail) they rather limit themselves to technical matters (how does this work, why doesn't this work, things like this).

Tobias comments on "MathOverflow 2.0 final sandbox testing" (22641)

2013-06-23T11:26:14-07:00

I am not sure what PM means, but note that he said SE rather than MSE (he is not an MSE moderator unless he uses a much different username here).

Todd Trimble comments on "MathOverflow 2.0 final sandbox testing" (22640)

2013-06-23T10:49:55-07:00

Earlier, jtunnell identified himself/herself as a mathstackexchange PM (not sure what this means; principal moderator?). Who will play the corresponding roles for MO 2.0? I'd sure like it if the current moderators will continue as such; they've been really great all this time -- consistently fair, firm, and at the same time sensitive and polite.

Whoever they are to be, I hope they are also mathematicians who are active on MO and who will continue to be sensitive to our concerns (the interchange with jt, or lack thereof, hasn't been exactly encouraging to me).

Poppy comments on "MathOverflow 2.0 final sandbox testing" (22639)

2013-06-23T10:34:50-07:00

Will the new site use the same cookies as the current site? That is, will unregistered users still be "logged in" when they return on Tuesday?

quid comments on "MathOverflow 2.0 final sandbox testing" (22638)

2013-06-23T09:05:21-07:00

@joro: I do not think your description that the number of upvotes for a tag badge is lower is quite to the point. Rather it seems that on 2.0 there are also bronze tag badges (in addition to silver and gold we have here). For the existing ones (silver and gold) they are actually somewhat harder to get as here it is 400 and 1000 votes, resp., while then it will be also 400 and 1000 but in addition one needs 80 and 200 answers, resp. For bronze it is 100 votes and 20 answers.

Poppy comments on "MathOverflow 2.0 final sandbox testing" (22637)

2013-06-22T19:55:21-07:00

Having looked up KConrad's reference to "son's sum of some" all I can say is: wow, I won't ever be looking at math.stackexchange.com again.

KConrad comments on "MathOverflow 2.0 final sandbox testing" (22634)

2013-06-22T12:59:17-07:00

@Michael Greinecker: the MSE meta link in quid's reply just above is a question that I posted a couple of weeks ago, and it never seemed to get much traction. I wonder why a site can't have a vote (by users above a particular reputation, say) on what format of questions appears as its front page for unregistered users, e.g., active questions vs. newest question vs. some other tab option, esp. if moderators feel the default list shown to those users does not reflect well the types of questions that should be encouraged.

Michael Greinecker comments on "MathOverflow 2.0 final sandbox testing" (22631)

2013-06-22T10:05:25-07:00

@quid Thank you very much! MSE users have no real influence on such issues. I am a MSE moderator and I think this is horrible.

quid comments on "MathOverflow 2.0 final sandbox testing" (22624)

2013-06-22T05:11:17-07:00

@Michael Greinecker: The link "(again)" in my first post in this thread leads to a thread containing various information. In the middle it was about something else, but scroll down to contributions by Willie Wong and me towards the end for several links. In particular, on meta.math.SE there is http://meta.math.stackexchange.com/questions/9820/what-is-on-front-page-for-new-unregistered-users and in a comment to the answer it contains a so far unanswered/unnoticed(?) question addressed at the math.SE moderators.

Michael Greinecker comments on "MathOverflow 2.0 final sandbox testing" (22623)

2013-06-22T04:26:22-07:00

@KConrad: Do you have a link where this horriple first-page-policy is described?

Asaf Karagila comments on "MathOverflow 2.0 final sandbox testing" (22622)

2013-06-21T23:51:08-07:00

@Joel: You can't see my reputation change from downvoting, the software hides it.

It seems that the serial voting script might not run, we'll have to wait for Geoff or someone else to confirm this, or point out that we didn't do enough to trigger it.

Andres Caicedo comments on "MathOverflow 2.0 final sandbox testing" (22621)

2013-06-21T23:09:40-07:00

+1 Yemon.

HJRW comments on "MathOverflow 2.0 final sandbox testing" (22620)

2013-06-21T23:00:26-07:00

+1 Yemon.

quid comments on "MathOverflow 2.0 final sandbox testing" (22619)

2013-06-21T20:07:56-07:00

@jtunnell: you explain us things we know. It makes little sense to just repeat this. Could you perhaps at least answer my question why this frontpage is not used for SO?

And by the way the old MO frontpage was not "newest" but "active" (and was this even different on SE2.0?)., in any case both would be better then this selection.

DavidRoberts comments on "MathOverflow 2.0 final sandbox testing" (22618)

2013-06-21T20:03:42-07:00

With the sort of voting patterns we have on MO, the proposed front page sorting would promote the sort of questions that get locked on StackOverflow. These are not typical questions for SO, and certainly not for MO.

Joel Reyes Noche comments on "MathOverflow 2.0 final sandbox testing" (22617)

2013-06-21T17:06:30-07:00

@Asaf Karagila, I suspected it was you, but how come this doesn't seem to be reflected in your profile page? (Unless it wasn't "you". :P)

Your reply about serial voting answers my question, thanks. (Although you might want to edit the sentence that starts "Serial voting is [...] a trade secret but" if it is a trade secret.)

Also, FYI, it's been almost 10 hours but it seems the serial voting mechanism hasn't kicked in yet. Maybe in a while...

KConrad comments on "MathOverflow 2.0 final sandbox testing" (22616)

2013-06-21T15:43:16-07:00

jtunnell: Thanks for your comments. You wrote "The old sort was "newest", which I would argue does not reflect a representative sample of questions from the site."

I don't understand this rationale. When I have come to MO (and see the front page as a new user because my browser always clears cookies) I've always thought that the list of questions I see is varied and interesting in large part because of the regular change of topics and themes. I thought the same about math.stackexchange's front page until a couple of months ago when the front page for new users there began to show widely viewed but unappealing mathematical topics that remain at the top for an interminable amount of time (case in point: the "Son's Sum of Some" question on math.stackexchange, which surely is far from being a representative question for the goals of that site). I wondered for a while what had changed, and then someone pointed me to a page explaining the new algorithm for determining what questions are shown to new users on the stackexchange network.

Has anyone from the MO side who discussed the transition to MO 2.0 with stackexchange admins raised the issue of what the front page will look like for new (unregistered) users?

SteveLandsburg comments on "MathOverflow 2.0 final sandbox testing" (22615)

2013-06-21T15:39:19-07:00

Yemon wrote:

My interest is piqued precisely by specific questions which don't have answers, rather than the massively upvoted musings of very clever mathematicians.

I second both this sentiment and its implications for the design of the front page.

Yemon Choi comments on "MathOverflow 2.0 final sandbox testing" (22614)

2013-06-21T14:34:44-07:00

(In my previous comment, imagine that word as spoken by Bill S. Preston, Esq. and then amplify and lengthen it by a factor of 10.)

Yemon Choi comments on "MathOverflow 2.0 final sandbox testing" (22613)

2013-06-21T14:33:30-07:00

jtunnell writes:

The old sort was "newest", which I would argue does not reflect a representative sample of questions from the site. Further, most of these newest questions are unanswered and thus not as interesting as some with answers.

This seems to me to be waaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaay off base for MO. My interest is piqued precisely by specific questions which don't have answers, rather than the massively upvoted musings of very clever mathematicians. Who are Stackexchange to judge what the MO community deem to be "a representative sample of questions from the site"? (Outliers can be very noticeable, it doesn't mean they should be catered to.)

Asaf Karagila comments on "MathOverflow 2.0 final sandbox testing" (22612)

2013-06-21T14:30:01-07:00

jtunnel,

I think that the problem is that when I tell someone "Yeah, you should check out MathOverflow", I am hoping that the first thing they will see on the site is a list of questions which are relevant and accepted and actually represent the site, rather than a list of questions which are just popular but are very unrepresentative to this website.

Also many people do not keep the log in from one browser session to another. If you need to log in to see what happened since your last visit, that's an extra effort which is not always worth the energy. If I were someone who logs out after every session (e.g. working in private mode, or having a browser that deletes cookies) I might have reconsidered the amount of participation on the site.

It is therefore essential that the front page is made more "up to date" and current, rather than "popular".

HJRW comments on "MathOverflow 2.0 final sandbox testing" (22611)

2013-06-21T14:07:29-07:00

jtunnell,

I'm afraid all I can say at the present time is that the homepage sort is very unlikely to remain as it is, but also very unlikely to return as it was.

Could you confirm that the views of the Mathoverflow community (as expressed on meta, say) will at least be taken into account? As you are probably aware, many users of meta were very worried about MO handing over its autonomy to Stackexchange. Your comment, which explains your motivations but doesn't acknowledge any of the concerns raised in this thread, isn't very reassuring.

jtunnell comments on "MathOverflow 2.0 final sandbox testing" (22610)

2013-06-21T11:34:44-07:00

Stack exchange PM here. A few words about the homepage.

First, the only changes we have made are to the anonymous homepage. If you are an actual registered user who is logged in, you will never see the changes that we've made.

The changes, including the sort, were made to give drive-by traffic a better idea of what the site was about. We added the banner explaining the Q&A concept, moved the top tags for the site front and center for those who scan for familiar terms, and we did change the sort on the question list.

The old sort was "newest", which I would argue does not reflect a representative sample of questions from the site. Further, most of these newest questions are unanswered and thus not as interesting as some with answers.

We wanted to convince visitors that there was valuable content on the site and it was worth joining and contributing.

Currently, the sort is something very similar to "Best by month", which pulls in slightly more popular questions with answers.

This is not at all our final algorithm, and we plan to continue testing various other combinations of highly rated content, popular content, and unanswered content.

I'm afraid all I can say at the present time is that the homepage sort is very unlikely to remain as it is, but also very unlikely to return as it was.

Asaf Karagila comments on "MathOverflow 2.0 final sandbox testing" (22609)

2013-06-21T08:53:10-07:00

@Joel: That was me, I also flagged one of the posts to notify the moderators to check that the script has run.

Serial voting is, to my better understanding, a trade secret but roughly five votes within a very short time span should trigger the script. It should run in roughly 10 hours from now. The script removes the votes and resets the reputation. If you have capped and got back the downvoted by going over the cap then no reputation is restored (for obvious reasons), and similarly when upvotes are reversed if you have gone high enough above the cap then nothing is taken from you.

I should also add that as far as I know the moderators should have the tools to find out "vengeance voting" when those occur (i.e. someone constantly downvotes your posts or whatever) but I doubt that in four days of testing we can get this much done.

quid comments on "MathOverflow 2.0 final sandbox testing" (22608)

2013-06-21T08:19:30-07:00

On MO the default for mathjax-rendering for titles (when looking at lists of questions) is off (but users can turn it on if they want). In the sandbox it is on (I did not yet test if I could turn it off, but this is not my point anyway).

Already relatively long ago it was discussed and then decided that it is better the default is off (also, if "we" would not consider it as better it is off it would be on, obviously).

Thus, could the default for redenering titles please be kept as off (and thus changed in the sandbox) after the migration.

(This suggestion assumes there is still the possibilty to have this option, if not, I would consider it as somewhat unfortunate, but will live with it. But if there is still a default to set, I am very strongly in favor of keeping it off; I provide provide detailed arguments on request.)

[To avoid any possibility of a misunderstanding: I am not talking about mathjax-rendering in general (this should be on, IMO), but specifically about the rendering in the titles of questions (when looking at a list of questions). And, to repeat, this is as we have it here all along, while we could easily have differently if we wanted.]

Side note on the front-page, if anybody reading this now wonders what all the talk about wikipedia is or should get happy that something was changed: the top-question used to be Why don't more mathematcians improve Wikipedia but it is now gone yet it seems only since "a month" (since it was asked) has just passed.

(By the way, the current MO-tab month-tab has as "top"-question Mathematicians with unusual education, closed, as duplicate, yet also so annoying somebody even asked a question just to make fun of and to complain about it.)

Bill Johnson comments on "MathOverflow 2.0 final sandbox testing" (22607)

2013-06-21T07:49:34-07:00

So delete votes on closed questions are recorded before two days pass from the closing time?

Joel Reyes Noche comments on "MathOverflow 2.0 final sandbox testing" (22606)

2013-06-21T07:36:51-07:00

I just got a string of downvotes at the dev site. I'm assuming someone is testing the serial downvoting mechanism. I have a question, though.

Sometimes, "serial upvoting reversed" doesn't lead to changes in reputation (e.g., Anton Geraschenko), but sometimes it does (e.g., -30 for Scott Carnahan). What does "serial upvoting reversed" actually mean? Someone serially upvoted, say, Scott, Scott got the increase in rep sometime ago and now the system is taking it back?

mt comments on "MathOverflow 2.0 final sandbox testing" (22605)

2013-06-21T06:44:12-07:00

When I try to use Account Recovery, I get "oops, something bad happened". Is it maybe linked to the fact that two character usernames aren't allowed on SE2.0?

Henry Cohn comments on "MathOverflow 2.0 final sandbox testing" (22604)

2013-06-21T06:01:17-07:00

I strongly agree with Keith Conrad's concern about the front page. The rest of the migration looks generally great (aside from tiny things that will rapidly get smoothed out), but it's a big problem if the front page specifically highlights questions that are unrepresentative of MO and perhaps even inappropriate for the site.

On another topic, is anyone else having trouble logging into the dev site with OpenID? When I try, it want to create a new account, despite the fact that I log into MO this way normally. Then it declares that "Oops! Something Bad Happened!" if I go ahead and try that. (Maybe I screwed it up by doing that once and then deleting the newly created account?)

joro comments on "MathOverflow 2.0 final sandbox testing" (22603)

2013-06-21T04:10:22-07:00

Not a bug, but the number of votes for a "tag badge" is lower, 100 upvotes.

KConrad comments on "MathOverflow 2.0 final sandbox testing" (22602)

2013-06-21T02:50:42-07:00

Keeping community wiki or closed questions out of the front page for new (unregistered) users might be somewhat helpful for making the front page seen by new users align closer with the goals of the MO site, and maybe there could be some kind of button available to moderators to take selected questions off the front page that is seen by unregistered users, but generally this feels like it's going to become a game of whack-a-mole. Instead of all this, why can't MO 2.0 simply use the same algorithm it does now for presenting its front page of questions? When I look at the current MO site as an unregistered user, I see a list of reasonable math questions, and from experience the front page on MO for unregistered users has always reflected the site well (unlike the "Sum of Some" question that has been at the top of the front page for unregistered users on math.stackexchange for the last few weeks). In contrast, the default unregistered-user front page on the test site for MO 2.0 has stuff like mathematicians editing Wikipedia, which overall I think is a poor representation of what people want MO to be. If it ain't broke, don't fix it, so is there some way to keep MO 2.0 from breaking what doesn't need fixing in this regard?

Evan Jenkins comments on "MathOverflow 2.0 final sandbox testing" (22601)

2013-06-21T01:08:55-07:00

Having the entire site in Latin would help keep out the homework questions.

quid comments on "MathOverflow 2.0 final sandbox testing" (22600)

2013-06-21T00:20:45-07:00

To repeat a technical point mentioned in another thread if ever it got lost:

there are many links on main to this meta, at the moment they work in the sandbox-version, but at the moment this meta is also still at its current location (regarding url). It was discussed that it would be important to keep these links operational and this should not be difficult to achieve (various ideas got mentioned). So, the question:

In which way is it planned to achieve that the existing links to meta will remain operational after the transsition?

Asaf Karagila comments on "MathOverflow 2.0 final sandbox testing" (22599)

2013-06-21T00:12:35-07:00

@Joel: Well then, we better get to work. We need to write a lot of things until Monday!

Joel Reyes Noche comments on "MathOverflow 2.0 final sandbox testing" (22598)

2013-06-20T23:51:04-07:00

@Asaf, it's filler text (https://en.wikipedia.org/wiki/Lorem_ipsum). We'll just need to replace it.

Joel Reyes Noche comments on "MathOverflow 2.0 final sandbox testing" (22597)

2013-06-20T23:49:12-07:00

Right now there are no users with the mortarboard badge. http://dev.mathoverflow.stackexchange.com/help/badges/25/mortarboard

Asaf Karagila comments on "MathOverflow 2.0 final sandbox testing" (22596)

2013-06-20T23:48:39-07:00

Am I the only one seeing the 10k tools help page in a strange langauge or something?

http://dev.mathoverflow.stackexchange.com/help/10k-tools

Also http://dev.mathoverflow.stackexchange.com/help/good-question is Latin, as well as http://dev.mathoverflow.stackexchange.com/help/accepted-answers and http://dev.mathoverflow.stackexchange.com/help/deleted-answers and http://dev.mathoverflow.stackexchange.com/help/create-account

Julian-Kuelshammer comments on "MathOverflow 2.0 final sandbox testing" (22595)

2013-06-20T23:37:23-07:00

Looking at the badges, it seems strange to me that Greg Kuperberg has been awarded the Epic badge, but not the Mortarboard badge. Am I missing something or isn't 1<50?

DavidRoberts comments on "MathOverflow 2.0 final sandbox testing" (22594)

2013-06-20T21:07:07-07:00

Community Bulletin

metaMathOverflow has no Site.ChildMeta defined

What are we going to put here?

DavidRoberts comments on "MathOverflow 2.0 final sandbox testing" (22593)

2013-06-20T20:53:15-07:00

On top of Andy's suggestion, we could also forbid closed questions from the front page.

Joel Reyes Noche comments on "MathOverflow 2.0 final sandbox testing" (22592)

2013-06-20T20:17:35-07:00

The question http://dev.mathoverflow.stackexchange.com/questions/134004/lets-try-and-edit-this-over-and-over-and-over-again has been closed and the reason shown is

closed as off topic by Asaf Karagila, David Roberts, Anton Geraschenko♦ 3 mins ago

Questions on Empty are expected to generally relate to , within the scope defined in the faq.

So some text needs to be edited here.

Tom Church comments on "MathOverflow 2.0 final sandbox testing" (22591)

2013-06-20T20:12:57-07:00

However, Andy Putman's suggestion would raise the importance level of the "shouldn't this be community wiki" discussions.

Poppy comments on "MathOverflow 2.0 final sandbox testing" (22589)

2013-06-20T18:02:22-07:00

+1 Andy Putman.

Andy Putman comments on "MathOverflow 2.0 final sandbox testing" (22588)

2013-06-20T17:22:10-07:00

One possible solution (if SE would allow it) would be to simply exclude community wiki questions from the front page. That would (I think) prevent most of the really terrible ones from cropping up.

quid comments on "MathOverflow 2.0 final sandbox testing" (22586)

2013-06-20T17:07:28-07:00

@DavidRoberts: the description of math.SE for Do and Do not ask is a lot more site specific. I strongly assume the moderators or "we" could decide what to write there precisely also on math.SE the link to meta goes to meta.math.SE. So, I assume these are purely temporal things as meta.MO does not yet exist on SE. Or, to put it differently, extrapolating from math.SE yes it seems MO can have their own things.

But specifically for what to write for Do and Do not ask this is an interesting question we might want to discuss (though perhaps not in this thread, see my first post in this thread).

quid comments on "MathOverflow 2.0 final sandbox testing" (22585)

2013-06-20T17:00:11-07:00

Some remarks on the front-page: I do not know precisely the criterion for questions to show up on this front page list but here is an educted guess and some commentary:

According to a thread on meta.SO, which I beleive from other threads to be still current, the "hotness" of a question is computed as

4 (log Qviews) + (Qscore x Qanswers)/5 + Sum (Ascores)

divided by something sort of 'continous' depending on the time asked and last active. And I infer from other threads that the months and week tabs are computed about like this with the time-dependency not being continous but a hard cut-off.

Now, if this is like this the main problem to me are not the views, but that the number of given answers has a considerable influence, once directly via Qanswers then via Sum (Ascores) and also via views.

This will all the time give us the almost exclusisevly opinion based questions that allow too many possible answers and both is discouraged even in the standard (not specific to MO) description question on top, creating ever more answers to them.

Also, if ever for the list as it is now, and basically it will be like this most of the time, this should be a guide for new users what the site is about and what questions can be asked here (which is as I infer from a thread over at the meta.SEs a motivation for this), this is in my opinion a user-guide for potential desaster, the ones will actually ask things like this and in the majority of cases face an unfriendly welcome, and the others leave because they find it not interesting to discuss why mathematicians do not improve wikipedia or the pros and cons of using a smartboard or what research topics are limited to students at top universities, and not to be overly pessimistic, yes, some will still find there way over this self-inflicted obstacle.

In my opinion, for a page like MO, that has very specific audience and really essentially nothing positive to expect form users that just happen to find the site somehow (inside the network or via a general search) and find it somehow fun and interesting this frontpage is completely unsuitable.

A question that would interest me to get an answer to from SE is why StackOverflow does in fact not have this type of frontpage. (Of course, the volume on SO is a lot higher, but it can not only be the volume, in fact it is not so clear how high volume should play a role in a descision here.) I could imagine some reasons they have for this might apply for MO.

DavidRoberts comments on "MathOverflow 2.0 final sandbox testing" (22584)

2013-06-20T16:32:31-07:00

In the tour one finds the instruction

"Don't ask about... - Questions with too many possible answers or that would require an extremely long answer"

(although at http://dev.mathoverflow.stackexchange.com/help/avoid-asking one has "If you can imagine an entire book that answers your question, you’re asking too much.")

I don't know if that latter clause should be there. There are many excellent questions which generate long answers which are just brilliant. Also, 'too many possible answers'? What does that even mean?

Also at http://dev.mathoverflow.stackexchange.com/help/avoid-asking there is

"If your question is about the site itself, please don't ask it here. Visit our meta-discussion site, where you can talk about things like what questions are appropriate, what tags should be used, suggest a feature, point out a bug, or generally discuss how this site works."

where the meta link is to meta.stackoverflow - I can see an argument for some site-related discussion going there, but not all. (Ditto the subjective question link - can we come up with our own version of this?)

DavidRoberts comments on "MathOverflow 2.0 final sandbox testing" (22583)

2013-06-20T16:25:11-07:00

+1 KConrad

KConrad comments on "MathOverflow 2.0 final sandbox testing" (22579)

2013-06-20T14:16:10-07:00

Since quid pointed to my previous meta question about the front page, I'll raise it again now that I see it at http://dev.mathoverflow.stackexchange.com/: what is the basis for making questions rise to the top there? At the moment the number 1 question (from May 30) is asking why more mathematicians don't edit Wikipedia pages. Can we please consider having the latest asked questions be what shows up at the top by default, rather than the most viewed questions for the past month (or whatever else the current obscure rule is)?

Asaf Karagila comments on "MathOverflow 2.0 final sandbox testing" (22578)

2013-06-20T12:47:16-07:00

quid,

In SE2.0 recalcs are done automatically whenever posts are removed/unremoved, and also one can activate it by going to the /reputation page (xxx.stackexchange.com/reputation), then scrolling to the bottom and clicking the obvious button.

markvs comments on "MathOverflow 2.0 final sandbox testing" (22577)

2013-06-20T11:32:51-07:00

@quid: thanks, that explains it! Also does anybody know whether 10K+ users will have the tools button?

quid comments on "MathOverflow 2.0 final sandbox testing" (22576)

2013-06-20T11:25:10-07:00

@janjitse: Yes, an upvote on a question on SE2.0 awards only 5 points (not 10). This is also applied to existing question (and this was also handled like this for other sites when this was changed for them). So, this is expected. I think there are also other sources for loss of points (I lost some too while never having asked a single question), namely due to the fact that points awarded for things that later were delted typically did not get subtracted here after the deletion (only if one asked for a points-recalculation). On the one hand, such a recalc now happened (I assume), and on the other hand I think on SE2.0 such a recalc happens more regularly and/or can be triggered by the user (as opposed to having to ask a moderator). I do not know if there are still other sources. But for questions a considerable loss (about half as you say) is expected, and there can be additional losses (though typically on a smaller scale).

janjitse comments on "MathOverflow 2.0 final sandbox testing" (22575)

2013-06-20T11:15:55-07:00

Is the amount of reputation points awarded for questions halved? In the dev, I lost a lot of points compared to the normal mathoverflow.

Is there any way to change the color of the favorite tags? On my current monitor, the light blue is hardly distinguishable from the normal white background.

quid comments on "MathOverflow 2.0 final sandbox testing" (22574)

2013-06-20T11:06:09-07:00

@markvs: something like this came up on the last test. Here, if somebody gave more then one answer two the same question they count as one answer in this count (but the presence of them is signaled by a number in parenthesis at the entry of the individual question). There, this is not done, but each answer counts as one. I consider it as very likely this is the cause for the discrepancy.

Added: looking through your answers, it seems you have three such "double" answers. So using the count as there, here you would have 322, further taking into account that two were given after june 18th, so are not yet there, this gives 320.

Asaf Karagila comments on "MathOverflow 2.0 final sandbox testing" (22573)

2013-06-20T11:03:22-07:00

Here is a screenshot.

I'm using Firefox 21.0/Arch Linux 64bit

(Here's the thread)

markvs comments on "MathOverflow 2.0 final sandbox testing" (22572)

2013-06-20T11:00:14-07:00

Small remark: the number of "my" answers on dev (320) does not coincide with the number of my answers on mathoverflow (319).

Geoff Dalgas comments on "MathOverflow 2.0 final sandbox testing" (22571)

2013-06-20T10:53:41-07:00

@Neil I believe that the help link should be working now. Let me know if it is not.

Geoff Dalgas comments on "MathOverflow 2.0 final sandbox testing" (22570)

2013-06-20T10:49:20-07:00

@Asaf I'm not sure I understand where the issue is on the custom message for bounty. Here is what I see:

http://i.stack.imgur.com/uLzzW.png

Geoff Dalgas comments on "MathOverflow 2.0 final sandbox testing" (22569)

2013-06-20T10:45:31-07:00

I have added moderators to the sandbox from mathoverflow - if you find you need moderator access and haven't been promoted provide a link to your user profile and I'll be happy to mod you. Also, if you find that you have an existing account but the system has created a new one for you, please provide a link to your newly created user account along with a link to your existing account and we'll merge the two. This will likely be the case for people who have logged in to mathoverflow.com using an open Id credential that we currently don't have in our system. Merging the accounts will effectively merge the credentials so you will have access to your existing account. SE 1.0 did a poor job of keeping up with the many credentials you may have while logging in so there may be a bit of effort involved getting accounts correctly mapped.

Neil Strickland comments on "MathOverflow 2.0 final sandbox testing" (22568)

2013-06-20T10:37:39-07:00

The 'help' link at the top of the page (http://dev.mathoverflow.stackexchange.com/) is broken.

Asaf Karagila comments on "MathOverflow 2.0 final sandbox testing" (22567)

2013-06-20T10:36:18-07:00

There's some problem with the bounty dialog. Its width is "on the pixel" and when writing the custom message it screw up the frame, and the text moves and gets cut off.

quid comments on "MathOverflow 2.0 final sandbox testing" (22566)

2013-06-20T10:27:05-07:00

@François G. Dorais: the front page is visible. Follow the link, do not login, instead click the MO logo. I just was in the process of deciding if complaining about it (again) was on-topic here. But, for something positive, I like the "Tour"! (It will have to be decided what to write in the ask about and do not ask about, as from math.SE it seems this is costumizable, but this is not for this thread I think.)

François G. Dorais comments on "MathOverflow 2.0 final sandbox testing" (22565)

2013-06-20T10:11:30-07:00

A lot of our user concerns were about things that are not visible on dev: meta, help, front page, etc.

François G. Dorais comments on "MathOverflow 2.0 final sandbox testing" (22564)

2013-06-20T10:10:36-07:00

Can current moderators be awarded diamonds on dev so we can check out the new moderation tools?

Asaf Karagila comments on "MathOverflow 2.0 final sandbox testing" (22563)

2013-06-20T09:30:08-07:00

Awesome. Awesome to the max.

Geoff Dalgas comments on "MathOverflow 2.0 final sandbox testing" (22562)

2013-06-20T09:08:02-07:00

We need your help! We have refreshed the dev tier for the migration of MathOverflow from version 1.0 to 2.0. It is current has of 6/18/2013 - this will be the final opportunity to find any issues and report them before we go live. Check it out here:

http://dev.mathoverflow.stackexchange.com/grant-beta-access?key=4885d961c6cfa02e839d7da263cacb38

We will monitor this thread for any feedback or bug reports over the next few days. There are a significant number of new features that we have introduced as part of this release (review system, new badges, etc) and we hope they are helpful to the MathOverflow community. These new features have been in production for all other Stack Exchange 2.0 communities for quite some time and we will continue to improve them as we make tweaks based on user feedback.