Not signed in (Sign In)

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

  1.  
    While reviewing my revisions to an old post I noticed that the diff fails to highlight a change in emphasis that I made (which, in fact, was the primary reason that I made the revision). Namely, notice that in version 2 in [1], in the last sentence of my post, I emphasized the word "highly" in the phrase "highly recommend". But the diff does not show this change (no change markup). Obviously this is of little consequence in this instance since it's my own change. But it wouldn't be so innocuous if the change was made by another user. So I thought it worth mention that such changes might easily go unnoticed. Is this a known bug? Are there other known problems with revision diffs?

    [1] http://mathoverflow.net/revisions/31457/list
  2.  

    I'm confused, what is the problem exactly? It didn't highlight the word that was newly \emph'd, but it does display the word differently in the two revisions. I'm pretty sure this is a feature, not a bug. :-)

  3.  
    I thought the problem would be clear. There is no sort of markup whatsoever to bring the users eye to the fact that a change was made at this point in the text. The changed text should be marked up in some way as are all other changes. If I hadn't pointed out that change it's highly likely that you would have missed it upon quickly scanning the two texts for changes. In fact, even armed with the prior knowledge that I had made some such change at some point, I still had to scan the revision history a couple times to find that precise spot. If it wasn't obvious to the original author then it's highly likely to be overlooked by a more casual observer.
  4.  
    I should add that this is not only a GUI problem. Examining the HTML source shows that the newly emphasized text is not marked-up with any of the "diff" tags employed for other differences. So, e.g, if someone were to implement a post-processor that summarized the changes then it too would miss this change (without resorting to an additional diff to discover such).
    • CommentAuthorAndrea
    • CommentTimeJul 21st 2010
     

    I agree that this should be considered as a bug, or at least a missing feature. But you may want to read past discussions on meta about the switch to the StackExchange 2.0 platform. Short story: we are not getting any new feature in the nearby future. :-(

  5.  

    I agree that this should go in the bug bin. As Andrea said, we won't be able to fix it in the near future, but it may be worth reporting this on meta.SO.

    For what it's worth, if you wanted to implement your own diff, you can do it by following the "view source" links on the revisions page, which shows you the markdown source (in this case, the revisions in question are link and link). Then you can use whatever diff tool you like. It's not a very good solution, but it does the job.

  6.  
    Thanks for the info. In fact I do have my own custom diff tools (based on work I did long ago as an XEmacs developer, and some consulting on a very abstract revision system) but they'd be overkill for this purpose. However, if MO should need coding volunteers at some point please do not hesitate to contact me.

    I haven't yet had a chance to explore the SE/SO platform. Where is a good place to start?

    Why can't we fix bugs? Is it not open source?
  7.  
    It's not open source. There is an open source alternative called OQSA. Stack Exchange is going through a change at the moment (from 1.0 to 2.0) and we will at some point soon need to decide whether to move over to 2.0, stay with 1.0, or move to something like OQSA. SE2.0 has some major advantages (network effects with similar sites, dedicated and competent full-time employees to deal with issues), but also some possible drawbacks (which we don't fully know yet because the terms of the migration are up in the air at the moment). But SE1.0 sites are no longer going to be given new features by the SE people (though I think they'll still fix bugs and major problems).