Finnamatkat just launched their redesigned site. With skepticism in the front of my mind I decided to see if web standards were given any consideration on the new site. I know that I’d used the old site at some point of time, but I couldn’t remember anything about it.

The new site looks surprisingly clean and ad free. Or maybe it’s just my broser settings with adblock set to filter quite agressively. (Update: an unfiltered Mozilla gives the same clean interface, so no problems there.) Even Flash is used in a moderate fashion and there aren’t too many intrusively animated objects to fill my screen. Changing the font size causes some problems, but nothing major. Even turning off images doesn’t cause any problems. So far, I’m quite impressed. It even looks like the developers of the site have known what they’re doing. (Now doesn’t that sound arrogant? Remember, I’m being a skeptic here ;).

Looking at the code causes some concerns. To start off, I was surprised to see that the DTD used is XHTML Strict. Correction, I was positively surprised. However, the developers appear to have forgotten to change the language decleration of the html-element. Swedish isn’t a good option for the language setting of a page with content only in Finnish. The code is not clean or aesthetic and clearly some sort of development tools have been used to create the site, the element identifiers are so convoluted that they weren’t written by hand.

The site almost validates, the front page gets eight validation errors, all of which are minor and easily corrected. There is a minor point that the content encoding declared by the server and in the meta-element conflict. Another small issue that is easily corrected. Validating the CSS causes more errors and a slew of warnings, mostly about the use of the background-property. Nothing major jumps at me though.

My main problem with the site is that it appears to use quite a bit of scripting. At least the page code has a declaration that if JavaScript is turned off, users should change their browser settings. I’m not against using scripting etc. but am a strong proponent of unobtrusive JavaScript. A travel agent’s site should (I’d go as fas as saying must) work without JavaScript. In reality, after some fiddling around with the site I can’t even figure out what JavaScript is really required for. Other than developer lazyness.

What annoys me the most though is that the site doesn’t offer clean URIs. Even the front page is redirected to an insane URI. However, I don’t have any idea if any ASP.NET tools and servers have similar functionality to Apache’s mod_rewrite. If they don’t, well just another reason to choose good tools for the job…

All in all, I’m positively impressed. It appears that standards based development may be on the rise even in Finland. What taints the site is that a bit more care and a simple run through the W3C validators would leave an even better impression. Since accessibility isn’t my strong point (at least yet), I’ll leave accessibility evaluations to others.

Chris Heilmann’s post about suitable replacements for CAPTCHAs reminded me about the little tricks I’ve used to avoid comment spam in the Life of Jalo. While the solution is exceedingly simple it is effective:

1. the comment form has a hidden field with an autogenerated value and
2. the server checks for the validity of the hidden fields value when submitted.

Currently the autogenerated value is static, but can easily be changed to something a bit more dynamic, especially if session or such are used. By using this approach LoJ hasn’t gotten any comment spam. A matter that was highlighted when I used the same commenting script to allow feedback in Anna’s comic and forgot this little piece of protection. For a site that gets relatively few visitors (about 20 a day then) it took only four hours for the first spammers to use the form. Implementing the spam protection outlined above dropped the amount of spam to zero immediately.

I don’t know if this kind of approach could be used to foil robots and ensure that a human is filling the form, but it is an effective way to avoid comment spam. I know that the current value that’s used would be very easy to hack if spammers had any incentive, but I’d be interested in figuring out if a more complicated key generation would work.

Maybe I’ll get around to testing some approaches sometime later…

A while back I started writing about this topic. Now reading throught the RDF-IG mailing list archives I came across some discussion on how to use the xml:lang-attribute in RDF-documents. HTML documents have a similar attribue, lang, that can be used.

In many cases, the language of an HTML document is set in the opening html-tag as generally the whole document will be in one language. There are some cases in which other languages are used in parts of documents in which the identification of different languages would be semantically correct. However, I seriously doubt that most authors do not use them. One of the problems that authors face when using the lang-attribute (in either or both forms) is deciding when to use it.

Continue reading →

The interview of Andy Clarke has really started a buzz about web professionalism that sounded so good to me. While we need to take our education about web standards out of the collective herds of standards users (think preaching to the choir) and educate the coming generations better (as Rob Dickerson Holly Marie Koltz aptly points out), what we really need is to consider what web development really is about.

Both Molly and Roger are heading in the right direction by bringing up and emphasizing the word craft. This struck a nerve with me as I was talking with a friend of mine on how standards based web development is a form of craftmanship instead of just doing the work. Which then lead to a whole discussion of trying to translate craftmanship into Finnish which isn’t an easy task. But I think that Kevin Leitch is in the right direction when he points out that a professional can be defined as anyone who earns money from the work they do.

Continue reading →