Firefox Web Developer ++

ajt on 2004-06-24T12:04:41

Recently I upgraded my browsers at work. I've now got Firefox 0.9, Mozilla 1.7, Opera 7.51 and some IE thing. I only use IE to test things, it's pants to develop on, and I only use it because it's "company standard".

My favourite toy has to be the Web Developer Extension for Mozilla/Firefox. Very cool, it allows you to do lots of things that are normally a pain during development: turn things on and off on the fly, clear authentications, view xhtml block elements, and edit the style sheet on the fly. Opera isn't quite as powerful, but even that has it's debugging and developing tools.

At the moment I'm working on a SAP application server that uses TT2. I've got the Badger book, and I'm slogging my way through TT. I know our HTML guy coudn't grok TT, he hated it and to be honest I can see why. While I like some aspects of TT, and it does seem to work, I don't belive many of the claims made by the fan base. Templates aside, if it all goes well, I've offered brian d foy and article on the experience.


Why?

Juerd on 2004-06-24T15:28:20

to be honest I can see why.

So why?

Re:Why?

ajt on 2004-06-24T20:40:42

Because though it separates some business logic from the presentation end, you still end up with a lot of logic going on in the template. The language is complex enough to confuse our HTML monkey - it certainly confused his Dreamweaver tool. He is comfortable with ASP, but he found the logic is TT2 strange. I'm not saying it's not impossible to use, but it's not a doddle for a non-Perl using HTML head to use.

Personally I find it very Perl like, and I know what's going on, but if I hadn't used Perl before, then it's not as obvious as I think people like to present. Additionally, I find it annoying that it's Perl like but not actually Perl - but that could just be me.

You could argue that it's complex enough to be useful without being too complex that it's a proper language. Which is I think the argument put forward by the mini-language people. You could also say that it's too complex to be simple, but not powerful enough to be really useful.

I think if conditions had been slightly different I may have picked another templating architecture, AxKit being my preferd option I think. However it's a useful learning experience, and I've been quite please with the results so far.

I don't think it's a bad tool, you could say it's an excellent example of a mini-language templating system, I'm just not convinced that mini-language or full language solutions are actually a good idea. Though XSLT has some very bad things in it, I'm instinctivly drawn to it. Before you point out that XSLT is a language in it's self - I'm aware of that too...!