A certain site checking service (who have managed to bamboozle a large number of public sector bodies into paying attention to their ‘league tables’) complains about Very True Things because it claims that –

Tag 'del' may not come under tags 'p' or 'div'
Tag 'ins' may not come under tags 'p' or 'div'

What’s more it claims that these problems make the page invalid and that its validity checks are carried out in accordance with the HTML 4.01 specification.

Very True Things is XHTML 1.0 not HTML 4.01. What’s more it did at the time of the test contain a deprecated attribute which whilst picked up by said checker was not listed as a validation fail, despite the fact that I use a Strict doctype. Any system that claims to report on validity should at least check the doctype and apply the appropriate rules.

More importantly, ins and del are very odd elements.
As the XHTML DTD says:

<!-- these can occur at block or inline level -->
<!ENTITY % misc.inline "ins | del | script">

The HTML DTD is a little more difficult to read as it relies on an SGMLism not much used in HTML:

<!ELEMENT BODY O O (%block;|SCRIPT)+ +(INS|DEL) -- document body -->

But in either case the validator will confirm that ins and del can appear just about anywhere in the document, and can certainly be used within a p and div elements.

So next time your client or boss gets hassled by a salesman from this company (fvgr zbefr) you can tell that they don’t know what they’re talking about and should be ignored.

No Comments - Comments are closed

  1. JackP says:

    Yes… if that company are who I’m guessing they are (and let’s face it, your encryption isn’t exactly 128-bit), then you’re not the only one to have queried them. Look them up on AccessifyForum.com, or search Isolani.co.uk. For starters.
    Having said that, they HAVE raised the profile of web accessibility considerably. Just trying be fair and balanced here…

  2. Steve Pugh says:

    I’m not sure that they have raised the profile of accessibility much. Outside of the public and financial sectors that they target (and accessibility experts) no one has heard of them; and those sectors are amongst the leaders of applying accessibility anyway (though whether they apply it successfully is, of course, another matter).

    The temptation for an overworked web team is to create a site that passes the tests (the same criticism applies to Bobby, Cynthia Says, etc.) rather than one that is actually accessible.

    Eighteen months ago, whilst working on a major accessibility project for a key government department I had to change plans mid-project to improve the department’s ranking on the league table, regardless of whether that actually improved accessibility or not.

  3. JackP says:

    Yes, that’s the problem with any automated tester. As someone I know recently quoted “make sure you measure what you value for you will come to value what you measure”. In other words, if the test ain’t perfect, you’ll be measured against the test, not against accessibility. There’s any number of articles on the subject. If you’re interested and haven’t come across them before, drop me a line. TBH, I’m presuming this is all old news to you…
    Me, I work in the public sector and the name’s quite common there. Assuming of course I successfully hacked your 128 bit encryption.. but you’ve got a fair point – would I have heard of them if we weren’t published in a league table?