MATLAB Central Answers website problem?

Did anyone have any problem?

When I click "My Answers", it shows both my questions and answers with questions listed first. Same thing happend if I click "My Questions". It happened once before. I cleared borwsing history and the problem disappeared. Today it happened again. No effect after I cleared browsing history. It happened both in Chrome and IE.

Answer 1

[Randy Souza]

Hi everyone, sorry for the problem.

Walter is close; this is actually a style sheet issue as we roll out a new UI for the contributor question/answer page. It should be resolved by 6:00pm EST.

And, once the fix is out, it should be possible to view Walter's profile page without waiting 20+ seconds ;-)

[Fangjun Jiang]

Right after Wednesday evening's downtime, I had that problem. It disappeared after I cleared my browsing history. Is that a coincidence, or the style sheet has been updated multiple times?

[Randy Souza]

No coincidence -- it was accidentally released on Wednesday. Today it was on purpose.

[Fangjun Jiang]

Seems working now. But the order is not right. For answers, oldest first. For questions, newest first.

[Walter Roberson]

Dang, I hoped it might change the order if I hacked the URL to include the sorting options used by the rest of the search engine, but that doesn't have any effect. :(

[Fangjun Jiang]

What do you mean, Walter?

[Walter Roberson]

For example,

http://www.mathworks.com/matlabcentral/answers/?sort=updated+desc

'desc' sorts in descending order, 'asc' for ascending order; 'updated' vs 'asked' vs 'votes' vs 'answers'

However, if you take the base url

http://www.mathworks.com/matlabcentral/answers/contributors/434782-walter-roberson/answers

and add /?sort= and any of those combinations, the order of the answers does not change.

[Oleg Komarov]

I appreciate the improvement in speed. Hope the sorting issue will be solved soon too.

[Randy Souza]

We're aware of the sorting issue and are sorting it out...

Answer 2

[Sean de Wolski]

Arghh.

< rant>

Why do we have to click though our answers in groups of 50.  It was excellent before, you could see all of your answers at once, you could search for keywords, you didn't have to click things and or guesstimate where answers from a few months ago were.  Don't fix what ain't broken!!  *The minimal number of clicks the better!!*

< /rant>

[Walter Roberson]

Everything at one time was breaking for my profile, as I am past the 3000 Answers point. The situation was unsustainable.

[Sean de Wolski]

Okay, so broken up into 1000s (or a selectable number). 50 makes finding a post from two months ago a real pain. Personally, I wouldn't mind having 3000 show, I could search through every answer I've ever written quickly.

Answer 3

[Fangjun Jiang]

Talking about search, I found the search capability of the website is under par. For example, right now there are more than 9000 answers. If you search for "controllable", there are some 700 answers. I have no idea why the first few answers show up. When I click those answers and try to find "controllable" in the text, none was found!

Also, should it provide the capability to search in "My Answers" or "My Questions"? I think that will be useful.

[Walter Roberson]

Please read through the MATLAB Answers Wish-list (tag:wish-list) as several of these issues have already come up. You will, for example, see there the best available documentation as to what the search engine is doing.

Answer 4

[Paulo Silva]

OMG now when I choose my own answers I get the oldest first and no way to reverse the order, that makes the feature completly useless for me, I want to be able to choose the sort order and have it default to new ones first, I really don't care much about answers from 5 months ago

end rant

[Arnaud Miege]

I second that.

[Kaustubha Govind]

I brought this up with the MATLAB Answers team yesterday, and they mentioned that they were going to fix it soon.

[Paulo Silva]

Great to know that I'm not the only one with the problem and that the fix is coming, hope they don't put the disable MATLAB Answers for the fix to be done

Answer 5

[Walter Roberson]

Yes, now that you point it out, I see that too, with Firefox 4. The Questions are shaded differently.

Both are on the same "page" with Javascript toggling which part is visible, so it sounds as if that bit of Javascript broke, perhaps during last Wednesday's software downtime.

Answer 6

[Image Analyst]

The counts on the Contributor's page http://www.mathworks.com/matlabcentral/answers/contributors has stopped working. The number of questions answered has remained stuck since middle to end of August 2012. I've notified the Mathworks already.

[Lindsay Coutinho]

We're aware that the number of accepted answers is not updating. We have triaged it for an upcoming release. Thanks for reporting it!

Answer 7

[Image Analyst]

Another weird thing I've been noticing over the past few weeks. I can leave an answer or comment, click Save, and see it posted in a refreshed page. So apparently it's done, right? Then when I click Answers (on the top menu line) to go back to the main page, it will ask me if I want to leave the page or stay on the page. Apparently, for some reason, it thinks I'm still typing stuff in. Has anyone experienced this? Sometimes I'll start something off as a comment, and then click cancel because I decided to make it an answer, but I don't think this is always the case. Anyway, even if it were, Cancel should fix the "dirty" flag for the page.

[Walter Roberson]

Clicking Cancel on an Answer will always immediately generate this message now.

If you click Cancel on a Comment, then (by design choice) the partly-entered comment is stored in the page FORM data, there in case you choose to click Comment again, in which case what you had already entered before re-appears. This apparently sets whatever flag is use to determine whether the page is dirty or not, so when you go to leave the page it asks you if you really want to.

A canceled Edit might possibly cause the same behavior; I would need to test.

[Image Analyst]

Thanks for the explanation.