Contact
Lifestream
Normally I don’t care much for valid html code. Perhaps it was my recent encounter with WAP / WML that made me think twice. I guess we have all become very comfortable with out highly tolerant PC browsers and have forgot past experiences.
Anyway, I spent the better part of the afternoon trying to get the site to validate using HTML 4.01 Transitional. In the end, it proved to be very time consuming and somewhat picky. But there were also some more fundamental errors and general sloppiness that I’m glad to be rid of.
The most frequent mistakes I’d made:
- General: Sloppy use of html code, i.e. forgetting end tags or putting a quotation mark in the wrong place.
- Still more instances of ABSMIDDLE.
- Having IMG tags with no ALT=”" property.
- The return of the freakin ampersand (&). Instead of using & in URIs I should have used &. None of the scripts or software that I use seem to be able to convert ampersand and thus I will have to be careful in the future.
- Having news items with " in the description instead of ". Or not at all. It turns out I had posted loads of items to the Pasqlnews script with this faulty property.
- Using <UL> with no nested <LI>. I used to utilize <UL> tags to indent text in blog items. Now I have moved that to CSS as well and use a <DIV> tag to set style margins around the text paragraph.
- Having a rich and pointless <body> tag, including margin settings that have been obsolete for ages.
- Having XML style tags in place despite not being XML, i.e. <br />.
- Trying to use VALIGN=”" in TABLE instead of TR and / or TD.
- onSubmit events that were in the INPUT tag instead of the actual FORM tag.
- Javascripts that weren’t properly declared by type, i.e. <script type=”text/javascript”>.
No surprise really. Most were old sins that have followed me through the years. But I’ll be sure not to repeat them in the future. As for standards I’m gonna stick with html. Even if all the trendy people go for pure XHTML. I don’t feel the time is ripe just yet. And any changes in standards will mean I’ll have to rewrite all the 3rd party scripts myself. Just fixing the errors above forced me to edit both the Sympoll, Pasqlnews, Shoutbox and all the SSI snippets (the gallery- and forumboxes).
So for now the document validates. I doubt it will last very long though since the site is made up of so many different scripts and just one of those can tip the balance. But the work I did will pay off. I did after all reduce the amount of validation errors from over 200 to 0.
As for the rest of the site, and mostly those pages that are not generated by Nucleus, they can’t validate at the moment. But then again they are nowhere near [as complex] as the portal document.
