Went to a nice talk on “AJAX - Time for The Transformation of Web Applications” presented by CHI*A, an off-shoot of the ACM organization here in Atlanta. Though the talk did have a sales pitch at the end, and some shameless plugs, they were bearable — and actually the company who sponsored, ClearNova, actually has a pretty good product. I’m no fan of frame-works on top of frame-works, but hey if it works (then, it works).

The speaker, Steve Benfield, had a few really good points about AJAX that we’ve all heard, but it was nice to hear them again in good context. The first being, that yes, it’s just a name given to a set of technologies that have been around for a while — that in itself is a “good thing.” Being able to classify a form of rich, user-interactive web-site building tools, however dorky and technically incorrect some feel the acronym may be (see my previous post), is not the important part. The important part is the fan-fare and hoopla, and research, books, frameworks, interest, and talks that ensue. And so they ensue.

The other interesting point was that because it is a new term and a spankin’ new technology, people always wonder, will it stick around. The answer, our speaker says, is absolutely. Main reason being: Users Like It. If a user is shown on one hand, a web-app with nicely AJAX-ified features, and on the other hand a standard form/post/response web-app on the other — it’s guaranteed they’ll go for the more responsive, more intuitive web-app, for sure.

In other news, my pal Bart just got a crazy-wicked job for the Washington Post working on some (yup, you guessed it), AJAX-ified web-app projects that sound really, really cool. Behind all the bitter jealousy, we’re all happy for him. Good luck Bart!

There are some really good meetings coming up soon, and just wanted to blog about a few of ‘em:

AJAX Time for The Transformation of Web Applications: presented by CHI*A Thursday, September 22, 6:30pm at Macquarium. Details: www.chia.org.

PHP Templating Engines: Monthly Atlanta PHP Users Group meeting, presented by yours truly! Thursday, October 6, 6:30pm at New Horizons. Details: http://www.atlphp.org/node/88.

The Monthly Atlanta Ruby Users Group meeting is starting back up, with my good pal Derek Haynes of HighGroove Studios heading up the charge. Details forthcoming, but possibly Late October. Details: http://highgroove.com:2532/atlrug/show/Events

Stay Tuned!

Oh my!

After setting up a IBM Thinkpad T30 with gentoo linux and the works: we’re talking software supsend to disk a la hibernation (works better than WinXP hibernation), wireless/wifi, suspend, power-saving features, and of course a full on mirror of our dev environment (apache2, postgresql, subversion, php4.4, PEAR, Smarty, etc., etc.), I realized that this install was better than the Fedora Core 4 install I had going on a similar (same laptop)!

Of course, compiling KDE and some of the other components (X.org) took a very long time (several hours — overnight even), but it is fully optimized for the Mobile Pentium-4 and fast as heck!

So, rather than just follow the same steps to get my Fedora box up and functioning the same way, I simply made a Stage4 tarball of the system, copied it over to a removable USB storage drive, backed up a few things (homes, etc, root, etc.), then repartitioned, extracted the tar, and boom - out of box, amazing, working laptop with everything fully functioning.

I’ve never done a quicker install of Linux or any operating system. Amazing.

UPDATE: had to create one directory: /var/logs/apache2 — there was a symlink to this direcory that didn’t exist. Makes sense, no need to back up logs.

This is really big news. I’m calling it now.

PayPal Introduces New Micropayments Pricing to Increase the Convenience of Purchasing Digital Goods. It doesn’t take much imagination to stretch this into no just digital goods, but maybe, Digital Services?

Tung also sent me this link from a slashdot post re: Ebay’s plans to buy Skype (for some crazy large number), and I happen to believe in the commenter’s predictions.

Not only do I believe what this person is saying, but I am seeking to foster and promote this promotion of people and services, and rightly compensating those for providing and facilitating!

Just remember, you heard it here first.

Damn! I just wrote the longest entry on how our condo is for sale, including a listing to the unit, my take on the condo market, describing how much work we put into the unit, blah, blah, blah, and it’s gone. The Internet ate my entry.

Seems like my entire life is one big textarea — we live inside these horrible web-apps, and technology is supposed to make things easier, but here I am bitching inside another textarea box, hoping that I don’t hit the backspace key while not focused in this little box, making me go Back a page, and losing all this bitching.

Damn!

Is it possible to create some sort of database schema diff file, but smarter? So, for instance, I know one can take a diff of the the current db schema in use (some scripts to dump the two schemas in question and compare them manually), but I don’t know of any way one can generate those magical statements like ALTER TABLE …. or ADD/DROP COLUMN, in any type of automatic fashion. It’s either all or nothing on a table, right?

I know, I know, the DB Schema was supposed to be set in stone, months ago during the all-encompassing, formal design docs that figured out every single possible data field we wanted stored forevermore and into the future. Riiiight.

So how do people do it? Besides being really careful about altering data-types, and making sure you hand-code up the same change that you want to do in the dev/test/prod database, right? Or simply writing down the exact steps (or using something like phpMyAdmin, phpPGAdmin, or the Oracle Enterprise Manager Java-based piece of s___ — these tools being able to let you copy-paste and hand-jam manual ALTERs and ADD/DROP COLUMNs)?

I mean, yes, I know there are more problems to be had when altering tables like so (what about all the data that’s in ‘em), but I’m just curious to know if anyone has run into making common changes to a db schema, being able to “check those revisions” into a source-code control system (subversion/cvs/etc) and have your developers upgrade, seemlessly.

I know I’ve seen it in some open-source projects — on an upgrade of one of my fav open-source projects (MythTV, I’ve seen the upgrade go through all the revisions and upgrade the db accordingly (and I’ve also seen that break and I’ve had to go in manually and drop and restore entire tables too, though)…. Something to think about.

OK, this is driving me nuts. On a Edgewall Trac wiki and project managment site I set up on a gentoo box (specifically for serving this up), I am consistently getting a “Document Contains no Data” error when ever I post certain words to the wiki.

One night, it only took me 3 hours to determine that a line containing the word “/docs/” was causing the server to immeditately spit back “Document Contains No Data.” It was sitting in the middle of a url, just a link to a page. What in the world could it be? I tried escaping it, to no avail.

It is the most irritating thing, because I found today that I posted a line that contained the words:

$/ usr/local/pgsql/bin ….

It simply dies, yet again. What in the hell is running on my gentoo box that’s killing that kind of request? Is it se_linux (hell no, I didn’t install that). Is it mod_security ? No - it’s possible that it is so built in that I can’t even find it, but that’s not loaded up in the default apache2 ebuild. I can’t seem to turn on enough logging anywhere to find out why only when certain words are posted to the wiki, the server just fires back that annoying message.

Stuck, and clueless, and tired of trying to find out this very strange, very cryptic problem….