Bugs found through looking at the archived Blog...

by Michael S. Kaplan, published on 2014/02/03 08:26 -08:00, original URI: http://sortingtherestallout.blogspot.com/2014/02/bugs-found-through-looking-at-archived.html


So it was just a few days ago that I explained how you can use the Internet Archive to get at the old blogs in my Sorting it all Out Blog.

But then I found out something very interesting.

And that something was quite unexpected.

It was (believe it or not!) a bug.

A bug in one of our more controversial keyboard layouts.

The layout? The Canadian Multilingual Standard keyboard layout.

It was put in as a comment to a blog written years after the original blog was written.

The blog? Well, the archived version of it is Getting all you can out of a keyboard layout #9b

You can scroll down to find the comment reporting the bug, or better yet I will just quote it here:

--------------------------------------------
(((The dead key is on [Right Ctrl + . ] )))

This dead key adds a dot above, changing [ c ] to U+010B [ ċ ], [ C ] to U+010A [ Ċ ], [ e ] to U+0117 [ ė ], [ E ] to U+0116 [ Ė ], and so on.
So one would expect that [ Z ] becomes U+017B [ Ż ], and [ z ] becomes U+017C [ ż ]; which is indeed true about U+017B but not true about U+017C. You get U+0017 instead of U+017C.

I suspect this is a mistake in the KL design, where someone accidentally typed "017" when he or she had to type "017C", confusing 0 and C or dropping the trailing C or something.

Is U+0017 actually what is supposed to be there? I doubt it, because if you really need to input U+0017 for whatever reason, this KL already supports it as [Ctrl]+VK_W.

--------------------------------------------

Now whether you like that particular keyboard layout or not, and I pretty much don't like this kind of multilingual keyboard layout, I certainly wouldn't ever want to have it with such a bug.

But perhaps the fact that nobody ever reported the problem through actual USE would tend to indicate that at a minimum that particular letter is never used!

Anyway, triage approved of fixing the bug, and I was able to fix it in place atop the existing layout. Even at the time, many were impressed that the Blog was finding bugs so obscure that even the French Canadian Standards people didn't notice!

And now the even happier ending -- somebody was able to:

  1. notice the bug report in a comment written in 2008;
  2. repro the bug on some older machines;
  3. be unable to.repro the bug now on newer machines;
  4. realize that the bug was found and fixed by the Blog;
  5. inspire me to write THIS blog about the issue!
And with #5, I write today's blog....
I
no comments

go to newer or older post, or back to index or month or day