views:

396

answers:

1

Has anyone successfully developed a "native" Palm WebOS app using GWT? How do you setup your development environment and how did you interface with the mojo framework?

I was thinking one setup might be to create 2 projects: a GWT & a Palm OS project. Setup your build to copy the results from you GWT compile into the Palm OS project. Probably want to target your GWT builds at Safari.

I guess you would want to generate Java wrappers for the Mojo framework.

Thanks

+1  A: 

Please understand, I don't mean to offend you, but ... at first glance, this sounds like a really bad idea to me. GWT sacrifices some things to be able to do the cool stuff it does, but some of those things it sacrifices (eg. efficiency) seem very likely to be problems on WebOS. Not to mention that GWT is designed for "normal" (web browser) JS, and WebOS JS has some significant differences (in terms of the core types available and such).

Not saying it's impossible or anything, but just ... there's probably a reason you've had this question up for a week and haven't even gotten a single response.

machineghost
If you watch videos by Google about GWT they go on and on about how using GWT optimizes the javascript to make it so much faster than hand coded JS. I'm not going to argue with you if you have better information on the subject of performance because that is very important.Yes, GWT probably has a lot of incompatible APIs but the thing doesn't need to be 100% Java. The main point is to get basic Java to JavaScript conversion to get the benefits of a strongly typed language and be able to code in Java which is my strongest language.We both agree it's not impossible :) Good!
Cal
With all due to respect to Google (and they are due a great deal of respect; GMail's contribution to the AJAX "revolution" alone was huge), there is a significant body of evidence to suggest they are Java programmers first and JavaScript programmers second (the very existence of GWT itself suggests as much). I suggest reading this article to get a sense of what I'm talking about: http://www.sitepoint.com/blogs/2009/11/12/google-closure-how-not-to-write-javascript/. Like you said, neither of us think what you're doing is necessarily impossible, but I still maintain that it's likely a bad idea.
machineghost
As it turns out, I happen to know someone who works at Palm. This is all unofficial of course (he's just a random coder, not a PR guy or anything), but here's a few quotes from when I showed him your question (just more food for thought for ya): "the gwt thing seems like a bad idea...so much more work for something so simple", "I don't know what gwt is optimized for as well ... technically we're using v8 ... but there might be some idiosyncracies if it's not tailored to webkit", "it's best to just go with the recommended aptana plugins and the like".
machineghost
Thanks for these comments. I think it's in Palm's interests to have WebOS development accessible to a larger cohort of developers (with hardly any effort). Maybe it's not in the interest JavaScript developers to taint their pure JavaScript WebOS world :) You can watch these videos if you want to hear what they say about optimizations http://code.google.com/events/io/2009/sessions.html#gwt Basically from what I gather the optimizations aren't necessarily browser specific things. More like inlining as much code as possible and shrinking/obfuscating the code and "interning" strings etc.
Cal
Regarding the sitepoint article you linked to. I don't think it is relevant since it has nothing to do with GWT as far as I can tell. Please correct me if I'm wrong.
Cal
The SitePoint link was just meant to challenge the "just trust Google" mentality you seem to have ;-) Google is not the master of all things JavaScript ... far from it. So just because *they* think GWT "optimizes" things, does not mean the larger JS community would necessarily agree. But look, you can argue as much as you want for why Palm should lather Google up with massage oil and give them a good rub-down; I was just trying to show you that the chances of that happening are pretty remote, and unless that does happen your GWT plan is probably going to be a bad one.
machineghost
Let's review:(Cal)=> Loves[Google/GWT/Java] Hates[JavaScript]. (MachineGhost)=> Master debater: proficient at straw man arguments and finding guilt by association. Knows *someone* who works for Palm which gives his fallacious arguments a solid foundation. (Using GWT)=> A bad idea since Closure is apparently bad according to some article on sitepoint. Or maybe it's a good idea since Google is the largest search engine on the internet! Which fallacious argument to choose?(The larger JavaScript community)=> Determines what techniques are the most efficient not tests and measurements.
Cal
You should talk to your colleague at Palm to see if he can arrange the required Oil rub down :) Then we can close off this question for good! Oh... maybe he can get in touch with my colleague from Google (who thinks GWT+WebOS is a great idea) to get the lubricant fueled cooperation started. I think your on the right track with the whole oil based solution.
Cal
Look I tried to answer your question, which is more than anyone else on this site has done. Just because you don't like the answer doesn't mean you have to regress to middle school. If you want to keep arguing with yourself here feel free, but I'm done trying to reason with you.
machineghost
At least the record is straight on your claims. Thank you.
Cal
"Just because you don't like the answer doesn't mean you have to regress to middle school."... Nice one, +1 for hypocrisy Mr. Oil Rubdown. I hope you wrote a note to yourself about regressing when you initiated such talk. But, I know you'll remember this thread before you call an idea stupid and back it up with non-sequiturs because I can tell you're a sharp tack. Thanks, it's been fun!
Cal