It seems that motorola has decided to take the bull by the horns and start the ball rolling for the MIDP 3.0 specification. Given the advances cellphones have made over the last year, it's a good thing that the ball has gotten rolling with this.

Looking at the proposal, one thing that really jumps out at me, and something that I doubt will make it into the final spec, is the idea that multiple, concurrent midlets should be enabled.

I have to say, that shocked the hell out of me for a number of reasons. First of all, I take it for granted that you can run multiple midlets. I do it on my Series 60 phone all the time. But then, that's something you'd expect a smartphone to be able to do. For a feature phone, this is new territory. I can just picture the 17 year old with a pink RAZR complaining to the guy at the Cingular store that her phone keeps giving her "Out of Memory" errors because she's got Spongebob and NY Nights running at the same time. Don't laugh, it'll happen.

I seriously doubt Motorola, or anyone for that matter, is going to want to beef up the low end feature phones (and again, look at the RAZR for the poster boy for this category) to the point where it can run multiple applications. The market doesn't want it, the benefits really aren't evident, and it just won't happen anytime soon.

Additionally, the proposal calls for increased functionality in a number of areas that really need help. Specifically, it states that a goal would be to "Improve UI expressability and extensibility". Hallelujah, it's about fucking time. As I've mentioned before, the MIDP 2.0 really didn't improve the lcdui package much (at all?) over the MIDP 1.0 specification, and it's badly in need of some lovin. It's sad to say, but you can't set the background color of a form, change the text color on a form, or anything else you'd take for granted, without having to use a third party library like J2ME Polish. It's hard to believe that you can write a better looking Java application for a Psion Revo+ (which hasn't been produced this century), using it's ancient JVM, than you can with J2ME and the Midlet 2.0 spec.

And the problem of the horrible looking lcdui forms and form elements becomes worse when you look at midlets on QVGA phones. And it seems like more QVGA phones are coming out all the time.

The proposal also calls for the ability to "Enable MIDlets to draw to secondary display(s)", which is another good thing. I know you can do that with some phones, again I'll point my fingers at motorola and their IDEN libraries for IDEN phones. But a universal standard is always a good thing. Otherwise, the proposal is pretty much the standard stuff (better performance, better large screen support, better gaming, IPV6).

I really look forward to perusing the specification when it's published.