I have been looking into Atlas and Ajax.net for a while and I have yet to see why I should waste my time. It seems like such a raw, immature implementation of a sophisticated technogy that I am better off spending $500-$1000 getting a 3rd-party Ajax component like ComponentArt, or rolling my own nasty Ajax code. Earlier this year I built an entire website with heavy Ajax functionality (www.lagunacollege.edu), and they were all server side 3rdParty controls that I dropped on a page. A couple of lines of JavaScript was all it took for me to have callbacks and all sorts of cool things.
Because, honestly, this Ajax.net stuff is a sandbox. I downloaded the latest beta and got nowhere. There were so many compile errors and bad references that I gave up. Its not like I can't quickly deliver Ajax features from another vendor. When I browsed the code, I saw tons of client-side code that I did not even want to get into. Isn't the whole point of Ajax.net to abstract the nasty internals of Ajax so that a server-side developer can get up and running with high-fidelity web apps with spending his days writing JavaScript?
I was far more productive using a 3rd-party control that had real word examples like load-on demand tree view, images strips, splitters, etc. and spared me the headaches of scripting. These Ajax.net demos are utterly lame. A password strength widget? An accordion thingy? Rounded corners? Give me a break. And what is with all these examples where you have to build a web-service? What is the point of that complexity? When you look at the samples of the 3rd party vendors, they a real world scenarios, like Ajax grids, slick menus and tree-views, etc.
All that I can assume is that this Microsoft's attempt to keep the competion guessing until it comes up with a coherent way to deliver rich UIs for the web. For now I think this Ajax.net stuff is the domain of hobbyists and tight-fisted hackers with plenty of time on their hands.
Yeah, I wrote about the same few posts too. I was required to remove Ajax.net libraries from the solution yesterday because I'm tired to read forum every day and looking for a solutions for many problems... and I haven't time to play with it in the real project. The last thing was that after a new release the webparts stop to be working... Looks like that these controls are in the "Not Supported" list. So I asked in the previous posts - why do I need 300k of additional javascript code? UpdatePanel that supports label with timer? I just for an experiment installed MagicAjax.Net library for asp.net 2.0 (one DLL, few lines in web.config) and... webparts are working very good!!! Why?! I mean why Microsoft team is writing library that can't support own standard ASP.net controls, but 3rd-party solutions can? Lets wait for the release and I hope that there will be no a 300k of javascript code, I do not neet it, and as minimum standard asp.net 2.0 controls will be supported. Otherwise I don't know why Microsoft spent so many months to develop this library...hello.
TunnelRat:
Isn't the whole point of Ajax.net to abstract the nasty internals of Ajax so that a server-side developer can get up and running with high-fidelity web apps with spending his days writing JavaScript?
well, you can look at that thought that perspective too. so, just drop some updatepanels everywhere and you're ready to go.
you seem to describe yourself as an AJAX programmer with lots of AJAX sites developed, but as i've seen in lots of other threads (so, you're not alone), you don't want to learn Javascript to build better pages. You just want to drag-n-drop controls into the pages and that's it. the last time i looked, the J standed for Javascript...to me, this means that if i want to build cool performant pages, i'd really need to learn Javascript. but hey, that's just me...
Hi Luis, thanks for the answer, but why did you decide that we don't know or do not use javascript? Heh, we use it very well, if you mean about XmlHttp - the same, and we have created own AJAX controls - it is not a problem. Let me to describe my point of view and why I'm waiting for some AJAX solutionfrom Microsoft. The one from many examples - lets look to the nice controls int the ASP.NET 2.0 - WebParts. They are good, but.. if don't want to refresh whole page during every minimize or maximize of any webpart (this is really very bad) - I need to put web zones to a some AJAX panel. What it really means? To implement support of the webparts in the own AJAX panel - we need to wrap (hardcode!!!) some ASP.NET client objects - "__wpm" etc. Lets propose that Microsoft will fix some issues or implement some new features and will create new update for the ASP.NET 2.0 (.NET 2.0 framework). In this update they will rename "__wpm" to "__webPartManager" for a some reason - just for example. Do you think that this is not real? Real! And what? All our AJAX panels will stop working... Do you follow my thoughts...? That is why I like idea of Microsoft to create ASP.NET AJAX extension for theown ASP.NET controls, but guys, please do it normally, please think about developers, real environment, investigate and compare existing AJAX solutions...
hello.
Hi Luis, thanks for the answer, but why did you decide that we don't know or do not use javascript? Heh, we use it very well, if you mean about XmlHttp - the same, and we have created own AJAX controls - it is not a problem.
well, my previous post was a reply to the 1st post of the thread. well, if you read it, you'll see that he's complaning about not having server controls that you can drop in a page and get automatic partial refreshes. to me this sounds like not wanting to learn javascript...
They are good, but.. if don't want to refresh whole page during every minimize or maximize of any webpart (this is really very bad) - I need to put web zones to a some AJAX panel. What it really means? To implement support of the webparts in the own AJAX panel - we need to wrap (hardcode!!!) some ASP.NET client objects - "__wpm" etc. Lets propose that Microsoft will fix some issues or implement some new features and will create new update for the ASP.NET 2.0 (.NET 2.0 framework). In this update they will rename "__wpm" to "__webPartManager" for a some reason - just for example. Do you think that this is not real? Real! And what? All our AJAX panels will stop working... Do you follow my thoughts...? That is why I like idea of Microsoft to create ASP.NET AJAX extension for theown ASP.NET controls, but guys, please do it normally, please think about developers, real environment, investigate and compare existing AJAX solutions...
yes, i follow. web parts have some integration problems with updatepanels and i think this will only be solved after a rewrite of the web part components (it's easy to understand why it doesn't work...just think about the new way to insert scripts in a page during a partial refresh). i understand the pain of migration (i really do). but if they've changed something, i'm really sure that they've only done it because it was really needed and not just to upset the guys that use the platform.
btw, why don't you create your own client framework for building something similar to web parts? instead of relying on server side programming, why not do it only in the client side? ajax extensions give you a lot of help in these scenarios.
No comments:
Post a Comment