My wishlist for (X)HTML

Posted by admin on Friday, November 17th, 2006 at 3:54 pm in code, html, microformats. No comments.

HTML (and/or XHTML) is not developing as quickly these days as, say, ten years ago; but things are happening nonetheless. I’m no master of the use and history of HTML but nonetheless I’ve got some wishes for future versions.

The wish list

A useful q-element

In-line quotations are used a lot. And there’s no way to mark them up sensibly right now because of user agents’ varying implementations of quotes. Hence this is more a request towards browsers than towards HTML, but still. It would make me happy.

Form requirements

Web forms are very important but I think it’s fine that all but the basic features are the developer’s responsibility, not the user agent’s. Yet, a required-attribute should be introduced so there’s a sensible way to indicate exactly which fields are required and which are not. This will greatly help developers and users alike.

The menu-element

More or less every web page has some sort of navigation-area–usually a list of links to other sections of the site. I mark these up as an ordered list, which is semantically correct but not complete. Why is the menu-element deprecated when it could be used to describe our document’s contents in more detail?

Better use of the caption-element

The caption-element is now only used for tables. First, I want to be able to specify where I want to display my caption: I prefer them below my table rather than above it. I think this kind of a localization-issue (American preference vs. European). Secondly, ‘caption’ is such a commonly used concept, why not make it applicable to images and other embedded content? I’d love to be able to do this:

<img src="pic.jpg" alt="" title="My cat">
  <caption>My cat</caption>
</img>

This is a nasty one though, since it would conflict with existing title- and alt-attributes. But pointing out room for improvement is my job, not coming up with solutions… :-)

Richer mark-up for citations

When writing scientific texts (or any sort of text, really) you often quote other people. The elements we have now (blockquote, q and cite) are not enough for building a bibliography. I would like to see something like this:

<source>
  <author>Gibbon, E.</author>
  <title>The Decline and fall of the Roman empire</title>
  <date>1776--1788</date>
</source>

Now, I admit that this would also be very eligible for a Microformat—but with regards to the original scientific purpose of the web I think this is not too much to ask. Also, the cite-element could very well be gotten rid of in the near future, which I am very strongly opposed to.

Better use of the rel-attribute on anchors

The rel-attribute describes the relation between the source document and the destination document. Great! There are a lot of useful descriptors already, but I feel we could use some more, especially for use in web applications and same-page links.

The line between HTML and Microformats

I love microformats. They add valuable meaning to ordinary text for very little effort. They enhance our mark-up in much the same way as we enhance our page’s behaviour using DOM-scripting. The question arises: what do we need in the core of HTML, and what do we lay on top with Microformats?

We should, of course, be wary of scope creep. We should stick to the fundamentals in the HTML core and add semantic spice on a separate layer. But why then bother with developing HTML at all? We should at least add some basic concepts to the core—even if only to avoid polluting our class-attributes. The whole purpose of HTML—a mark-up language—seems defeated to me if we start putting too much semantics into our class-attributes.

Go forth…

Extending HTML is a risky business, with so many browsers, standards, specifications, working groups and documents out there. But I thought we were the almighty, über-flexible interweb… let’s live up to that ‘reputation’.