Random bits
Rapid fire opinions, devoid of value. Here we go:
1) Documentation: Yes good compared to other OSS projects. Yes can be better. Trolltech's API documentation and surrounding info is nice, and you should expect that when you work with a company that gets paid to make a product. Amazing how KDE continues to prove that OSS and commercial offerings benefit from one another.
So let's be proud of what's already documented and thank those that have put in the hours, but there's always work to be done. Other groups should be saying, "We need to get to a KDE-level of documentation" for depth, breadth and clarity. And I'll try to help over the next year, to avoid being just another person throwing their opinion around.
2) Ajax: Momma mia. When will it join the Push-Technology/Cue-Cat/VRML Death March? Talk about a solution in need of a problem. Ajax would make sense if there were no alternatives to client-server communication efficiency and presentation. But to hype a new option that seems inferior to current solutions *and* has a less room for growth?
I worked on a team in 1999 that implemented Javascript/DOM dynamic-client rendering and processing connected to servlets and RMI on a server. Looked cool with the added benefit of being the most illegible code in existence. And this was with extremely talented programmers; I pity those new to the paradigm. Why are we working toward browser-based separation of presentation and logic only to wind up with Ajax? This falls firmly into my 'can do' versus 'should do' rant.
"You know that tiny, tiny corner case where you want way less graphics than Flash, less network traffic than ASP/JSP/PHP and way less processing power than a fat-client or even an browser-based embedded app, and you don't care whatsoever about variant DOM-support and Javascript/ECMAscript-support or code mantainability because it's static? Oh and using a slow interpreted ill-formed language? Well, Ajax is your answer. Hey, wait, where are you going?"
3) Google-Sun: Talk about user profiling. Fits perfectly here. If you're someone that doesn't care about content privacy, loss of ownership or power-user (cliche alert!) application options, this is for you. Obviously any sane business would scoff at using this product, but maybe students who are by nature nomadic, dealing with mundane content and have access to decent supported bandwidth, this would be awesome.
As always with any free storage offer, like gmail basically, I always just think about opening up accounts and embedding my ogg files in there, and seeing if they'll ever notice, giving me a free music server. Talk about giving your account info out and getting in DMCA trouble. Yeah yeah yeah, I'd probably at least zip them up - the lazy man's obfuscation via file extensions.
But this is interesting in the continuing saga of client processing power and network reliability. In an age where OpenOffice still takes 30 seconds to kick open on my AMD64, I won't hold my breath for awesome browser-based apps. Unless of course they're built with Ajax. Then it's Good Times (tm). Should be only about 6 months until we have a browser-based Autocad, Avid and Blender, right?
I wonder where the pendulum of binary creation vs transfer will slow down on network-based apps this decade.
4) I was the one who shit in tuxipuxi's cereal. You got me.
1) Documentation: Yes good compared to other OSS projects. Yes can be better. Trolltech's API documentation and surrounding info is nice, and you should expect that when you work with a company that gets paid to make a product. Amazing how KDE continues to prove that OSS and commercial offerings benefit from one another.
So let's be proud of what's already documented and thank those that have put in the hours, but there's always work to be done. Other groups should be saying, "We need to get to a KDE-level of documentation" for depth, breadth and clarity. And I'll try to help over the next year, to avoid being just another person throwing their opinion around.
2) Ajax: Momma mia. When will it join the Push-Technology/Cue-Cat/VRML Death March? Talk about a solution in need of a problem. Ajax would make sense if there were no alternatives to client-server communication efficiency and presentation. But to hype a new option that seems inferior to current solutions *and* has a less room for growth?
I worked on a team in 1999 that implemented Javascript/DOM dynamic-client rendering and processing connected to servlets and RMI on a server. Looked cool with the added benefit of being the most illegible code in existence. And this was with extremely talented programmers; I pity those new to the paradigm. Why are we working toward browser-based separation of presentation and logic only to wind up with Ajax? This falls firmly into my 'can do' versus 'should do' rant.
"You know that tiny, tiny corner case where you want way less graphics than Flash, less network traffic than ASP/JSP/PHP and way less processing power than a fat-client or even an browser-based embedded app, and you don't care whatsoever about variant DOM-support and Javascript/ECMAscript-support or code mantainability because it's static? Oh and using a slow interpreted ill-formed language? Well, Ajax is your answer. Hey, wait, where are you going?"
3) Google-Sun: Talk about user profiling. Fits perfectly here. If you're someone that doesn't care about content privacy, loss of ownership or power-user (cliche alert!) application options, this is for you. Obviously any sane business would scoff at using this product, but maybe students who are by nature nomadic, dealing with mundane content and have access to decent supported bandwidth, this would be awesome.
As always with any free storage offer, like gmail basically, I always just think about opening up accounts and embedding my ogg files in there, and seeing if they'll ever notice, giving me a free music server. Talk about giving your account info out and getting in DMCA trouble. Yeah yeah yeah, I'd probably at least zip them up - the lazy man's obfuscation via file extensions.
But this is interesting in the continuing saga of client processing power and network reliability. In an age where OpenOffice still takes 30 seconds to kick open on my AMD64, I won't hold my breath for awesome browser-based apps. Unless of course they're built with Ajax. Then it's Good Times (tm). Should be only about 6 months until we have a browser-based Autocad, Avid and Blender, right?
I wonder where the pendulum of binary creation vs transfer will slow down on network-based apps this decade.
4) I was the one who shit in tuxipuxi's cereal. You got me.
0 Comments:
Post a Comment
Subscribe to Post Comments [Atom]
<< Home