Thanks for the insight into the looming battles. I was especially interested in your analysis of the various directions of JavaScript frameworks. There is a fifth alternative which is about to hit the market. It is called JavaScript Synthesis Technology (‘JST’) and is developed by a company named Morfik. This technology will be launched as part of the Web 2.0 Conference next week. See http://www.web2con.com

In a nutshell, Morfik allows the programmers to implement the business logic of their application in a high-level object oriented language of their choice. Morfik compiles this code into a JavaScript AJAX engine. The process is a true compilation and avoids boilerplates or code snippet libraries. The source code is put through a parser which includes a tokenizer and syntax analyser. The parser output is then passed to a semantic map builder which crates a detailed semantic map that conveys the entire ?meaning? of the application. This is a technique in widespread use by CAD systems. Finally a synthesiser uses this map to create JavaScript code which is semantically identical to the original source and conveys the same ?meaning?.

Since the output is neither an executable in machine-code nor a one to one translation of source code nor a collection of predefined code snippets, this Morfik process is referred to as JavaScript Synthesis Technology.

Siamack Yousofi
Chief Technology Officer
http://www.morfik.com

Riiiight
Another MS sponsored article

I find it quite funny that you think this is MS sponsored, because it doesn’t mention XUL????

I apologise for not going into detail on XUL, but there are only SO many technologies that we could cover in this article.

XUL is great technology, and there are other great technologies that we couldn’t mention here. This is a high level article :)

Cheers,

Dion

ruby on rails needs to be able to scale to big high traffic sites before it will become mainstream, and i don’t mean bitplayers like backpack, think more like ebay, amazon, yahoo, google, etc: sites that get serious traffic. that said, i am a huge ruby fan, the language seems to have supplanted perl as the cool hackers language, but it has performance issues.

python could also be resurgent if it is intergrated into mozilla as certain project pages seem to indicate, amking it an alternative to javascript, a language that simply is not acceptable as the future of application development.

what we need in the browser is some sort of ABI so i can do perl,python,ruby or javascript coding, all of which gets compiled at page load time into an object file the browser can execute. why should dynamic web pages with ajax like behavior be limited to one language? it makes no sense, particularly if that language is a weak toy like javascript. its good to learn javascript now but remember language diversity always comes back – look at sql!

“We can’t discount PHP. A lot of “serious engineers” (read: anyone who isn’t a PHP developer thinks they are serious) poo poo the PHP world.”

I guess I’m offended by that line. I for one am a PHP developer and I think I’m serious. It’s not the language itself but the scriptkiddies that might give PHP a bad name. However, one can do pretty serious AND scalable stuff in PHP as long as it’s done right. I even wrote a book about the serious type of PHP development.

Java may still be in the corporate ‘hot seat’ but from a developer point of view I really think this is thanks to excellent marketing by Java based companies. I’ve yet to see the first large scale Java implementation that performs well and doesn’t suffer from deadlines being crossed big time, going over the budget by a lot and engineers close to being burned-out. I guess I’m amazed by the fact that the corporate world still buys this garbage. As a language it’s exellent but implementation is your worst nightmare.

Ruby seems like the exciting new kid on the block indeed. I agree wholeheartedly on that one. However we’ll still have to see what will be left of it after ‘the hype’.

All in all a great write, except for that line about PHP… :P

I see a future for JS 1.6 in mainline browsers and embedding situations. Things like e4x, Rhino’s continuations, and some of the new spidermonkey functional programming primitives point to a direction where where it’s not necessary to add lots of new syntax to get the kinds of capabilities that JS2.0 was supposed to provide.

JS2.0 is dead. Long live JS.

Regards

I don’t think that JS will be taken seriously anytime soon. That said, I think that AJAX and it’s increased usage will speed that up quite a bit.
I’ve worked in too many shops where you have client/server developers put onto web projects. As you know, going from a stateful environment to a stateless environment is a difficult transition.
While the .NET stuff is very very good – it allows developers who’ve never done any kind of web work to dive in and create all sorts of pages that abuse the hell out of postback. What you end up with is a page that wants to act as a client /server thing but fails…miserably.

I’m really hoping that AJAX technology becomes the technology of choice to get rid of the unnecessary chatiness that products like .NET are introducing and finally make the stateless environment issue go away. At least, by making it appear to be stateful!

If that happens, then yeah Javascript will have to taken a lot more seriously. That means testing, automation, functional specs, etc.

I’m hoping to make it to the December 8th conference. My boss hasn’t responded yet about whether I can go or not, but my fingers are crossed!

I am not trying to make fun of PHP developers there. The opposite. I am making fun of the OTHER developers who consider themselves BETTER than PHP developers for no actual reason.

I apologise for that not being obvious. My english sarcasm is hard to capture in text sometimes ;)

Dion

That’s ok, I guess it’s a tad ambiguous maybe then :)

I agree that PHP has it’s issues though!

P.S: Love this blog! There was no need to repost on my site because I read everything on this site ;)

WHAT-WG’s specs take a lot of what uses Javascript now and turns it into declarative HTML, with higher-level Javascript hooks (e.g., a validate event instead of just an onsubmit event). It doesn’t invalidate Javascript at all, but just gives us a higher-level way to handle it.

In actual implementation, it will have to be implemented in Javascript itself (to be backward compatible, which is a primary goal). However, I think it has more weight than the current Javascript frameworks, in that it is limited and well defined. I hope current frameworks use it as an opportunity to simplify themselves, rather than seeing it as a competitor.

I am a huge fan of what the WHAT-WG is doing. I really hope that the IE team gets totally onboard and we move forward on all fronts.

Cheers,

Dion