Not signed in (Sign In)

Vanilla 1.1.9 is a product of Lussumo. More Information: Documentation, Community Support.

    On user pages I often see that among the unaccepted answers, some of the vote counts are in bold while most are not. Among the accepted answers, some of the vote counts are black on green while most are white on green. I cannot detect any pattern indicating why. Can someone tell me what these markings mean?

    I admit this is frivolous, but I can't not be curious.

    In the list of questions on a user page, you get four fields of information: number of people who have "favorited" the question, number of votes the question has received, the number of answers the question has, and the number of views the question has.
    If the "number of answers box" is dark green, it means an answer has been accepted; if it is light green, then it has answers, but not an accepted answer; if it is dark red, it means there are no answers.

    In the list of answers on the user page, you just get the number of votes the answer has received:
    The box is green if the answer was accepted and light grey if it wasn't.

    I'm not sure what you're referring to when you say "some of the vote counts are in bold while most are not" or when you say "some of the vote counts are black on green". Where should I look to see examples of these things?


    Ok, I think I understand where those questions came from: you were looking at meta.SE, where instead of green or grey background, they use plain or bold. I see that you posted a question there, and I posted an answer to it. But there still shouldn't be any place where you see vote counts which are black on green; that would be ugly.

    I have sometimes seen black on green vote counts on my user page as well (there aren't any at the moment). If Anton doesn't know about it, it's probably some kind of bug. (Although, for the record, it doesn't look ugly; it just causes people like Jonas and me to scratch our heads.)
    As Pete indicates, it is something that is only sometimes there, so I am afraid that if I give examples they could be gone by the time you look. I see there aren't currently any examples on my page, so I'm looking at the first page of users. On Qiaochu Yuan's page the vote count on the answer to "Proof of “if a^2 + b^2 = c^2 then a*b*c is divisible by 60”" is currently black on green, Anton, on your page the vote counts on the answers to "Sheaf cohomology and injective resolutions" and "intuition about the “section after base-change” for flat descent and exactness of the Amitsur complex" are currently black on green, On Harrison Brown's page the vote count on the answer to "Pseudorandom generators" is currently black on green, On Theo Johnson-Freyd's page the vote counts on the answers to "Is there a Faa di Bruno-like formula for composition of three functions?" and "Nimber multiplication" are currently black on green, Hopefully at least one of those examples works when you look. I see a similar thing sometimes where one of the non-accepted answers on a user page has bold vote-count, but I didn't see any examples in my little search just now.
    Anton, another example I see on your user page is the vote count to your answer to "How to respond to “I was never much good at maths at school.”", which is in bold.

    Drat, none of those examples worked for me. Maybe it's an issue with how the browser is parsing the css. What browser/OS are you using? Could you take a screenshot next time you see it?

    Oh yeah, I'll definitely do the screenshot. I don't know why I didn't think of that. I'm using Firefox 3.5.6 on Windows XP.
    Yeah, I've never seen that happen in opera.
    • CommentAuthorJonas Meyer
    • CommentTimeJan 1st 2010 edited

    Here's link to a screenshot showing the anomaly on your page:

    edit (by Anton): I inserted the image here

    It's definitely your browser then. I've never seen that before. It must be some kind of CSS error like Anton said.
    OK, it was more frivolous than I thought! As Pete said, it doesn't look bad or cause a problem. I thought it might indicate something (e.g., recent views) but if it only appears on some browsers that isn't plausible.
    FWIW I've seen this phenomenon before (black on green) too, almost certainly using a current version of firefox under linux (I can't be more precise because I don't know whether it was at work or at home, but I've certainly seen it, and more than once). Definitely not under XP though; it will either have been firefox under linux or OS X.

    I was actually planning on cleaning up the custom css file for MO sometime soon, since it's a mess, and since making it I've learned a bit more about css. Perhaps this problem will go away, and perhaps some new ones will arise.

    I see this sometimes too. By the way, I find the distinction between light and dark green hard to notice; could the shades be made more different?

    I've been visiting a bunch of user pages to see if I can reproduce this, and haven't seen an example yet. (Including all the ones mentioned above.) Next time someone sees this, could they also save a copy of the HTML code and inspect it to see if anything is different about the ones that come out black?

    Yes, I am using Firefox 3.5.6 as well.

    Ok, I've observed the problem while running Firefox 3.5.6 on linux. Corresponding to the four answers
    is the html

    <div class="answer-summary"><a href="/questions/8338/line-bundles-trivial-after-extension-of-the-base-field/8384#8384" title="total number of votes for this answer"><div class="answer-votes">11</div></a><div class="answer-link"><a href="/questions/8338/line-bundles-trivial-after-extension-of-the-base-field/8384#8384" class="answer-hyperlink">Line bundles trivial after extension of the base-field</a></div>
    <div class="answer-summary"><a href="/questions/1610/pairs-of-shortest-paths/1650#1650"><div class="answer-votes answered-accepted" title="total number of votes for this answer, which was accepted as the correct answer by the question owner">11</div></a><div class="answer-link"><a href="/questions/1610/pairs-of-shortest-paths/1650#1650" class="answer-hyperlink">Pairs of shortest paths </a></div>
    <div class="answer-summary"><a href="/questions/3270/which-rings-are-subrings-of-matrix-rings/3384#3384"><div class="answer-votes answered-accepted" title="total number of votes for 2 answers; one of which was accepted as the correct answer by the question owner">11</div></a><div class="answer-link"><a href="/questions/3270/which-rings-are-subrings-of-matrix-rings/3384#3384" class="answer-hyperlink">Which rings are subrings of matrix rings?</a> (2)</div>
    <div class="answer-summary"><a href="/questions/3278/whats-a-reasonable-category-that-is-not-locally-small/3282#3282"><div class="answer-votes answered-accepted" title="total number of votes for this answer, which was accepted as the correct answer by the question owner">11</div></a><div class="answer-link"><a href="/questions/3278/whats-a-reasonable-category-that-is-not-locally-small/3282#3282" class="answer-hyperlink">What&rsquo;s a reasonable category that is not locally small?</a></div>

    Reloading the page makes the anomaly go away! It must be a bug in Firefox.

    Thanks, Anton and everyone else, for answering my question.

    And indeed, the HTML looks perfectly okay to me. I don't see anything in it that could explain the problem.


    I've updated the css. It's about five times shorter than the previous mess, so it should be easier to maintain in the future. Let me know if you run into any anomalies.

    I have a hypothesis for what's causing the occasional black vote counts. The "answer-votes" and "answer-votes answered-accepted" divs occur within <a> tags, and I think Firefox is getting confused about whether to apply the color for things in <a> tags (which is almost black) or the color for the "answer-votes" div.


    @Anton: I don't quite buy your hypothesis, since nested elements is after all what HTML and CSS are all about. There is no excuse for a browser not being able to handle that. The real reason for the bug is probably more subtle. For example, it could be mutliple threads stepping on each others' toes or something. Oh, and I hadn't quite noticed before, but <div> within <a> is definitely outside the bounds of legal HTML. They should probably have used <span> instead of <div>.


    @hanche: I posted about the DIV inside A problem at meta.SE. It sounds like that actually supports my guess. It's possible that the browser decides to throw out the DIV tags, but it's weird that it happens so infrequently and in such a difficult-to-reproduce way. The only evidence I had for my guess before was that it's the only nearby thing in the html and the color seems to be the right dark-but-not-black (#222222 I think).