Anne van Kesteren

Just add those 14 bytes of accessibility

Yes, that is correct just 14 little bytes (even Google wouldn't notice...) and your site will be easier to navigate. And you won't have to add them on you most frequent visited page. If you homepage is the most visited one that is.

You could have a weblog, just like me. And while you are serving and reading an item (or you could be more interested in all those wonderful comments, which should be well-formed, I'm watching you Minz ;)) and you want to go back to the homepage. Since I don't have that usability weblog, I don't have a link at the bottom of my article straight back to the homepage. I do, have a quick and easy solution for you, which generates 14 bytes extra every hit. The accesskey attribute. Just press ALT + 1 (you did know about the kbd element, didn't you?) within your Mozilla browser. It can take it (This could be different for none Windows XP platforms).

That was all about accessibility. I didn't have time to add the 'new accessibility method of Tom Gilder', I just use display:none (Yes I don't like JAWS). The second reason I didn't have time to do such thing, was because I was busy making my site validating again from comment horror Minz Meyer, who is named again, because of his entry on inline XML. Help him out! It should be noted that the styles are imported with Mozilla 1.5b (latest fresh edition).

Last, but not least: New look from Simon Jessay, the next day: Redesign explained. I like his fixed techniques. I also like his RSS feed. When I have time, he gets a XSLT file from me (I want one myself of course :p), which makes the links clickable and transforms the whole document into XHTML.

And is a bottom note: we agree with Jacques Distler (shouldn't we all?). Reasons:

The only thing that is difficult is to reach his weblog. If you are not at home and you don't have your floppy with bookmarks with you, you will have to go to Google: q=jacques+distler. His URI is just too long.


  1. Anne, drop me a note if you are interested in further beta-testing (...uh, just kidding, ya know).

    I think I got it now. Promise change for the better. Go ahead, hoping I am your only comment nightmare :)))

    Oh, and by the way, there's a little typo in the Add A comment section....should be are available.

    Posted by Minz Meyer at

  2. Do you want me to throw some severe tests at your comment system?

    Having just set up bulletproof XHTML comments on my own site (I hope), I've developed some quite nasty testing examples.

    Posted by Gary F at

  3. Don't test. Just enter well-formed/valid XML, including the p element.

    Posted by Anne at

  4. Certainly, if that's what you want. But the chances of everyone else who visits doing the same is pretty much nil, particularly about using the p element. Everyone assumes it'll be taken care of for them (and rightly so). Perhaps you should look into textile?

    Posted by Gary F at

  5. Gary,

    Both of you have pretty user-unfriendly comment systems. Neither has a preview button, and neither offers the poster an opportunity to learn of and correct any XHTML errors in his comment.

    In Anne's case, the comment is just accepted, crashing the page. In your case, the comment is rejected, and the poster has to retype it in its entirety.

    For a relatively simple comment, like this one, maybe the burden on the poster is not too great. But for such a simple comment as this one, XHTML formatting was superfluous anyway.

    Posted by Jacques Distler at

  6. Jacques, you are, of course, quite right. The error handling is done (all of the XHTML errors are caught in an array), I just need to re-write the form to re-accept failed entries. Previewing should be trivial to implement as well.

    The one thing that I'm not sure of when it comes to previewing is where should the preview be shown: above the form, or on a separate page? Should potential erros be shown?

    Enough excuses, though, it has been checked into the urgent queue of stuff to do, so will be dealt with within a few days.

    Posted by Gary F at

  7. I would put the preview above the form, with potential errors highlighted (in whatever fashion strikes your fancy). That way, the poster can scroll down to the form, correct the errors in-place and resubmit.

    PS: Please don't misconstrue my criticism. I'm glad to see you are making great strides in the right direction.

    Posted by Jacques Distler at

  8. Oh, and Anne, please reset your Ping timeout for a longer value. Every time you rebuild this page (which seems to be several times a day), I get four trackback pings from you, which I then have to go in and delete by hand.

    If your timeout were longer, your MT installation would not be convinced that the ping had failed and would not resend it with each rebuild.

    Posted by Jacques Distler at

  9. Is there an easy way to do that? I actually didn't intend to ping you, but Movable Type discovers links in my post and sends pings without asking. I already noticed the trouble I made on you weblog.

    I'm sorry about this. The reason it happens is that this is the first time I work with pings/trackbacks and I have no clue what Movable Type does.

    Posted by Anne at

  10. RTFM.


    PingTimeout 30

    or whatever number of seconds seems reasonable. 15 is too short, 30 seems to work OK for me, 60 might be a bit long to wait but would certainly be safest.

    Posted by Jacques Distler at

  11. Jacques, constructive criticism is always welcome. :) I think we're all trying to do the same thing and a nudge in the right direction is appreciated. If you have any more, I'd be glad to hear it.

    Posted by Gary F at