Anne van Kesteren

Tables for indentation

Dave Winer, inventor of at least several versions of incompatible RSS is experimenting at wordpress.com. He is well known for using tables for layout and just made some changes which make the statement WordPress is a state-of-the-art semantic personal publishing platform not entirely correct anymore… In wordPress.root version 0.3 he writes:

<p>4. Indentation. <i>Done.</i></p>
<p>
<table cellspacing="0" cellpadding="0">
<tr>
<td width="25">&nbsp;</td>
<td>When you indent in the outliner, it&#8217;s reflected by indentation in the blog post. </td>
</tr>
</table>

Oh well, yet another reason to use real standards instead of some supposedly interoperable format called RSS. Not really because of the format, but because of the guy it came from. (Which is not to say there is not a whole of a lot things wrong with the format as well.)

Comments

  1. Now I read that source code. It seems that WordPress is outputting non well-formed markup again. That is probably the price you pay for doing this.

    Posted by Anne at

  2. Excellent, Anne. I haven't seen that in years. I remember doing stupid stuff like that back when I was 13, learning how to do HTML for the first time. That'd be a technique you'd expect to find on a website that has rainbows for a background image or stars that trail after your cursor while it plays some sort of stupid midi in the background.

    Posted by Dustin Wilson at

  3. Cool, a table inside a p element. :-P

    Posted by minghong at

  4. What a strange post, opening up a flame-party. Did you read where he said that because of the inbuilt CSS he couldn't do it in another manner? Functionality for the user has to trump semantics.

    Posted by Firas at

  5. Not to say that his is the best way to go about it, just that this is a needless ad-hominem attack. He did try other ways; he failed.

    Posted by Firas at

  6. This is a case of "biting the hand that feeds you." It doesn't reflect too well on yourself - or other advocates of Atom - that you resort to this kind of drivel. Standardise your own work, but never degrade someone else's work just because you have an obvious personal agenda.

    Following on from Firas' initial comment. It's true, "Functionality for the user has to trump semantics". I'm not sure you're aware of that considering the fact that I have to wrap P tags around my comment before I can post it.

    Posted by Nathan de Vries at

  7. All of you bashing Anne, stop. Dave is an idiot, just read this and you will see what I mean. As far as I'm concerned, if Wordpress.com doesn't support indents, don't do them. When developers start whining about "I can't indent my text! Oh noes! Someone help me, I'll die if I can't indent my text!" then we know there is a problem.

    Posted by Christian Montoya at

  8. If you actually read through Anne's post thoroughly you could have noticed where he mentioned a quote that appears on wordpress.org that claims that WordPress is a state-of-the-art semantic personal publishing platform. If WordPress was state-of-the-art and semantic it wouldn't use tables to do something so simple as to indent text. So to make this clearer Dave Weiner made WordPress indent text by using tables when you indent in the outliner. This makes WordPress neither state-of-the-art nor semantic.

    Writing out comments in XHTML isn't the only way to ensure validation of Anne's weblog. It's a good choice considering his reader base. Anyone who doesn't understand markup probably wouldn't get past the first sentence of any given post on his weblog.

    Posted by Dustin Wilson at

  9. Disingenuous much? WordPress itself has nothing to do with the issue. It'll publish whatever you push into its db.

    The whole issue is about how Dave Winer chose to solve a problem. Quibbling about the claims of the publishing platform is a feint.

    Posted by Firas at

  10. This is a case of "biting the hand that feeds you." It doesn't reflect too well on yourself - or other advocates of Atom - that you resort to this kind of drivel.

    This is about horrific markup, the whole Atom/RSS thing is tangent to that. Seems to me you're suggesting that Winer should be exempt from criticism because of he has a successful syndication format.

    What a strange post, opening up a flame-party. Did you read where he said that because of the inbuilt CSS he couldn't do it in another manner? Functionality for the user has to trump semantics.

    The only place I see him say that because of the inbuilt CSS he couldn't do it in another manner is where he says he tried to abuse a <blockquote/> first - just as bad. I very much doubt that WordPress won't let you specify your own CSS.

    Posted by Brendan Taylor at

  11. Firas, he could have asked. p.nested { margin-left:25% } would work. (Not that the paragraph is actually nested, but that is besides the point.)

    Posted by Anne at

  12. Firas, what Anne said in #11 is exactly what my first thought on this issue was. What Dave Winer used is more similar to what I could possibly have used like 6 years ago before I knew you could use CSS for more than just colours.

    I don't agree that this is a valid reason not to use RSS, but I do agree that you shouldn't use RSS. I do think that RSS stands to Atom as a P element with a class stands to that TABLE element Dave Winer used. So concluding, actually I do agree with Anne more than I first admitted to myself.

    Posted by Frenzie at

  13. Cool, a table inside a p element. :-P

    The table is a sibling to the p element.

    Posted by zcorpan at

  14. Yeah, exemplary. But fortunately, there is the code police.

    Posted by Jens Meiert at

  15. Dave doesn’t grok “semantics” or “interoperability” – film at 11.

    Posted by Aristotle Pagaltzis at

  16. I was always partial to:

    ;-)

    Posted by Ara Pehlivanian at

  17. Ara, cool! Who needs tables ;).

    ~Grauw

    Posted by Laurens Holst at

  18. Did you read where he said that because of the inbuilt CSS he couldn't do it in another manner?

    Last time I checked, even if everything else had failed (which is not the case here, there are many things he didn't try) inline CSS still overrode page-wide CSS.

    The table is a sibling to the p element.

    It's not, another paragraph is opened right before the start of the table, the table is in the 2nd paragraph of the statement.

    Posted by masklinn at

  19. It is actually. A table element start tag closes the p element implicitly.

    Posted by Anne at

  20. It is actually. A table element start tag closes the p element implicitly.

    Does it? Nice, I have to write that one somewhere (and check what firefox produces on that kind of code next time).

    Thanks Anne

    Posted by Masklinn at

  21. It is actually. A table element start tag closes the p element implicitly.

    Not in Opera 8.5x though (but it seems this is fixed in Opera 9) ;)

    Posted by Tino Zijdel at

  22. Ugh. Horribilistic!

    Posted by Ryan Bergeman at