ObfusCo > The List > jscript

Dynamic Web Blog

Ajax de rigueur:

• Open Rico.
• Prototype.

• 63) An attempt to demystify Atlas, which is apparently Microsoft's AJAX, er, I mean DHTML.
• 62) Screw client-side JavaScript, do everything on the server!
• 61) Getting concrete about Ajax development, in this case with TIBCO's 'General Interface'.
• 60) Check out 59) The world of AJAX libraries is heating up, with Yahoo! getting in on the action. I'm glad; I would not find writing all that stuff fun, myself, but using it would be OK.
• 58) Adobe's Spry might be cool.
• 57) Nice demo!
• 56) Another Web 2.0 approach, this time in Scheme.
• 55) Continuation-style programming has come home to roost; now available in client-side javascript.
• 54) Yup, Google has released some AJAX stuff for Java developers - it is apparently cool in that you write Java code which then gets run through some Google tools to spit out the AJAXy stuff for your browser (Although it is unclear to me exactly what the deal is with the license it is under - I ain't no lawyer.) Some tutorial articles are around.
• 53) Please don't do this. Thanks.
• 52) Check out a video about django, which is the Python alternative to Ruby on Rails, it would seem.
• 51) You got your closures in my AJAX. The whole thing just makes me think that while JavaScript and Perl are cool because they easily allow for so many different ways to tackle problems, it is frustrating for somebody who just wants to get work done to have to spend hours reading up on which OO library hack they should be using. One wastes so much time just getting the language you want in place, rather than having some language that is more restrictive and narrow so the rubber meets the road sooner. If you catch my rutabaga.
• 50) Check up on your mutual exclustion.
• 49) OpenLazlo supports both Flash and DHTML now.
• 48) Microsoft's Atlas stuff sounds nifty, and in theory it is relatively agnostic and not tied at the hip to lots of MS-specific systems.
• 47) Finally, somebody who has the ounce of sense required to see the middle ground.
• 46) Yahoo! has released their shiznit to the public.
• 45) For those of you venturing forth into AJAX, beware there be dragons, potentially.
• 44) In the ever on-going saga of "Just How the Hell Do We Do Inheritance in JavaScript Anyway?", it is worth reading what Harry has to say, along with the commentary. Check out even yet another style, too. It makes me wonder why everybody is bending over backwards to do all this? Why do we have to force JavaScript to be like Java? What is wrong with getting into the JS mindset and doing things without weird language hacks? I'd wager that either everybody is too lame to learn how to do JS right, or that prototype based JS is just a fundamentally broken language design. I dunno which it is.
• 43) Another Ajax system is echo2.
• 42) IBM has some tutorials on calling SOAP from Ajax [1] [2]. Personally, I think SOAP is sorta XML from Hell, but that doesn't mean you can't get things done with it.
• 41) The Ajaxian site has some useful security talk, and a myriad of other great things.
• 40) SACK purports to be a simple system.
• 39) Another supposed stud-muffin is phAtJAX, although their only fully-featured server examples use PHP, yuck.
• 38) At a pretty low leve, there's JSON-RPC-Java to let your client JavaScript call out to server-side Java.
• 37) ZK has a slightly different take, currntly using XUL.
• 36) A take on supporting history in AJAX, also described in an O'Reilly article.
• 35) Even more ways to shove AJAX up your... orifice. A little dose of warning and common sense doesn't hurt, and generated some ideas about handling network suck.
• 34) Ruby on Rails has kicked off some Pythonic versions which can dove-tail with AJAX. Note that there's a plug for Web Stack in an attemp to address all the wheel reinvention going on.
• 33) Myghty is a toolkit for getting AJAX working with Python.
• 32) Another offering is SAJAX which purports to easily hook you up to whatever back-end language you want. Unfortunately, it kinda looks like it requires PHP? I'm pretty sure I've been told to stay the hell away from PHP, if only for security reasons.
• 31) Misc reference material.
• 30) A fun toolkit for Ajax.
• 29) Check out haXe for generating JavaScript and other Web-oriented languages.
• 28) Everybody wants a version of getElementsByClassName(). So after some reading, here are what seem to be interesting refs: Glazblog 1, Glazblog 2, getElementsBySelector(), and maybe something about Behaviour.
• 27) Another AJAX library: Surebert.
• 26) On doing closures right in JavaScript.
• 25) A JavaScript Standard Template Library (also parts 2 and 3).
• 24) Example snippet of retro code for old browsers.
• 23) Yet another Ajax intro, this time influenced by Rails.
• 22) Doing transparency in a browser. another take.
• 21) Doing graphics in JavaScript/Ajax. Seems like one should just port some kind of already existing, production-quality 2D system to a Div array, instead of reinventing the entire stack of poo?
• 20) Fun with animations .
• 19) An LtU hit talking about using CSS to pump-up JavaScript, with some links to interesting work.
• 18) This is killing me. How can I use a variable as the pattern in a regexp? I don't want to use a literal, I want to use the contents of a string variable, ya know? Gah!
• 17) Yikes. Convert your reams of Lisp browser scripts into JavaScript! Continuing with the theme, check out The Little JavaScripter.
• 16) A somewhat anti JavaScript rant, food for thought.
• 15) Blah, blah, blah, Ajax, blah, blah, blah.
• 14) A tutorial care of Peter-Paul Koch.
• 13) If you thought you might want to do some interactive JavaScript, you might want to think again. "If you don't really understand the difference between addEventListener and attachEvent, stick to the traditional event registration model: x[i].onmouseover = doSomething; Don't experiment with methods you don't understand. You'll do yourself a favour in the long run." (Some possible fixes are the Prototype or AddEvent Manager libraries.
• 12) Most of the stuff from webreference.com on JavaScript is confusing, but the article on inheritance is still a possibly useful read, with things that are obvious yet worth saying like the fact that using prototypes is faster than using constructors.
• 11) On performance:
  1. 10) Okay, I gotta like a page that mentions big-O notation!
  2. 9) Some maybe old school notes on JS performance and style.
  3. 8) 10 possibly useful performance tips.
  4. 7) Another perf document, noting that << isn't so hot.
• 6) An educational comparison between Perl & JavaScript, handy for people like moi who know Perl first.
• 5) The fact that it is so whacky flexible when it comes to inheritance is pretty cool, but it also means that trying to learn how to impelment it the first time around is kinda cognitive overload. Pick any given Google hit for javacript + inhertiance, and its code will look pretty different from anothers. Heck, that's true just of how to set up objects at all, never mind doing inheritance.
• 4) Ah. Maybe Venkman will help make JavaScript development a little less hellacious. (There's still the fact that it isn't going to check all the code until I execute it... Yes, I just have a fundamental dislike of purely runtime languages.) Also, it is vaguely funny to have Venkman on because you see how pretty all web pages have broken JavaScript in them!
• 3) [Update: OK, I think the deal is that you cannot call document.write() while in the <head> section, since this test doesn't generate errors.] I'm attempting to learn JavaScript, and am banging my head against the wall. When I use document.write(), which I thought was something everybody used all the time in their code, I get errors. Here's my proof, which results in: Error: runC is not defined. Source File: http://www.obfusco.com/js-broken.html. Line: 11. (Seen in the browser's JavaScript log.) This is with Firefox 1.0.6 running on an Ubuntu Linux power-pc (G4) machine. No, I have not yet had the chance to try it with e.g.: MSIE. Is this something everybody already knows about? I can't find anything helpful via Google because the search terms are too prevalent. Am I just doing something horribly blatently wrong and I just can't see the obvious?
• 2) Another way in which I hate JavaScript (at least in Firefox) is that you have to run the code to find any errors - there is no compilation step, and there is no interactive interpreter. (Well, maybe there is, like Rhino or something?) Anyway, trying to develop in the browser means you end up with a language that has all the drawbacks of a scripting language with almost none of the benefits, you know? (Like early XSLT was all the trouble of functional programming, without the gain, because it didn't have lambdas.)
• 1) JavaScript in Firefox is killing me because the only error it seems to know is "XYZ is not defined" so it doesn't actually help you figure out why the interpreter decided XYZ wasn't defined. I figured out that I shouldn't use document.write() in the head section of the page (next bullet down), OK. But now I'm not doing that, but I'm getting a new "XYZ is not defined" error from other code that looks perfectly fine. Maybe IE has better, or at least different, errors? This is kinda killing me. I cannot fathom how anybody has ever written anything remotely serious for the web. How can Ajax possibly be a powerful and un-hateful development idea? Where are the IDEs? Where are the interactive debuggers? Where are the libraries that make all browsers behave the same for your code?