Monday, January 23, 2006

on AJAX and Web 2.0/3.0

Last summer I spent a considerable amount of my free time learning AJAX (for those who don't know what I'm talking about, see here and here). The goal of my very small 2-person team was to produce a tree-like navigational web application. Actually, the real goal was to experiment with AJAX and non-relational databases (read U2 and multi-dimensional) instead of the ubiquitous LAMP (Linux, Apache, MySQL, PHP). One of my computer science profs was all about ditching the relational database, an effort which I think is in vain, since they're here to stay. Plus the alternatives are much harder to learn, so we're done with that point.

Anyways.

Using AJAX turned out to be not-so-simple, and made me remember the infancy of the interweb, the late 90's, when using JavaScript to detect browsers was commonplace. Ugh. Like anything else, it's what you know (or not) that makes it harder than it should be. In this case, I very much didn't know enough to get that groovy navigational app going... but the appeal was so strong! Think Google Maps, or GMail, or Google Suggest, or Flickr, etc.

Alas fate had other plans in store for me. I got hired as a server-side programmer and went back to Java, XML and other J2EE technologies.

Until now. Having a little bit of experience with afore-mentioned technology, I got put in charge of coming up with an AJAX framework for the next version of our software. This is exciting news indeed, since in my opinion we should all be embracing it (of course, as long as we remember form follows function). Step number one was to pick up the latest AJAX book. Step number two was to read as much as possible in 5 days, and start coding. Step number three was to stumble upon a Jeffrey Zeldman article, on ALA, that mentions "wireframing AJAX is a bitch"? Surely it can't be that tough... And even more puzzling, a bunch of ALA members posting agreeable comments???

Maybe the coming together of server-side programming and web design that AJAX represents is making some web monkeys afraid, because of their lack of business logic skills (as well as other non-related computer hacking, bostaff and ninja skills). Maybe Zeldman forgot about all the third-party libraries that will put together nifty AJAX apps. Maybe I never really tried coming up with a production-ready framework (this is true) and it WILL be a bitch to wireframe.

In any case it's here to stay. As for the raging debate re: Web 1.0/2.0/3.0, who cares...

4 comments:

scott hendric said...

sorry...what?

nokomis said...

yeah..... i need your mailing address, I have a really late christmas present for you. you can email it to my old dordt account or briellenokomis@gmail.com

Ismael Florit said...

AJAX rocks. Well, they were better off with Davids but then again he's a bit of a bumble bee.

Anyway, all that nav asx.2asp YTO.dmG = form plus if then this that jpeg S1H2I3T4 is highly confusing but extremely exciting.
(you're making google-impact then?)
ps I'm sure Mr. Zeldman doesn't appreciate you calling him Shirley - O_O
ps Mark Ray at the rncm is also GAY - O_O
ps (ps is lonely today so here's another one) I think I'm going for Carole Pressland Can you tell how much mom loves her?

--dawn said...

Even if a lot of folks are stuck with SQL DBMS tools for now, your computer science prof might be onto something in trying to leave them behind ;-)

It is more intuitive and easier, not harder, to learn to model data in non-1NF (as done with XML). Then you can develop using end-to-end AJAX (non-1NF and two-valued logic from end-to-end).

See www.tincat-group.com/mewsings (Read the ripple delete entry for an example of end-to-end AJAX)

Prof. Wolthuis (aka --dawn)