Palm webOS and Pleco

[mikelove]

So Palm's just announced their new successor to Palm OS (along with the accompanying "pre" smartphone), which they're now calling "webOS", and while it's very slick it does not look like it will be backwards-compatible with legacy Palm apps. And in fact doesn't even seem to have a built-in stylus, being instead gesture-based like the iPhone.

Assuming they aren't pulling some sort of compatibility solution out of their hats (which seems unlikely - their developer page only mentions data migration and even that wouldn't apply in Pleco's case) this pretty much guarantees that Pleco 2.1 won't be coming to current Palms - it's hard enough to justify continued support with only the Centro and a couple of difficult-to-find handhelds left, and if we don't even have those anymore then that pretty much settles it. To that end, I've just updated the hardware recommendation page to more strongly discourage people from getting Palms.

As far as whether Pleco will eventually support this new webOS, I think that depends a lot on how things shake out licensing-wise with an online version of Pleco; from the description of the webOS programming interface it looks like a lot of Pleco's more interesting functions - handwriting recognition, flashcards, etc - would have to be done on a server instead of on the handheld, since it's basically just JavaScript with some extra APIs to access device functions / store modest amounts of data offline, though if (as I expect) they eventually roll out native application development for the sake of supporting high-end games as on iPhone that might open up a possibility for a full-fledged Pleco port.

If things do come together for an online version of Pleco, though, it looks like a webOS port would be quite straightforward, and in fact considerably easier than porting to many other mobile platforms - of course a Chinese dictionary is a bit less useful if it's dependent on an active internet connection (and the attendant lag time in fetching dictionary entries), but that doesn't mean it wouldn't be worth developing.

 

[benzhen]

new palm phone and os: stab in the back for palm devs?

So in all the hubbub surrounding the new palm pre, and webOS i haven't seen anything about legacy palm support, something i remember was promised way back when. is this the real end of palm os? does that wealth of software have no home on this and other future next-gen devices? i guess it was a real unfortunate choice to stick to palm for pleco, seeing that all that hard work might be confined to the palm TX and treo phones. the new phone looks amazing, especially with the multi-tasking. so, we'll just have to see what will be in store for developers and legacy palm app support. love to hear everyone's thoughts.

 

[mikelove]

TreoCentral reports that Palm actually told them that there wouldn't be any backwards compatibility, at least not out-of-the-box:

http://www.treocentral.com/content/Stories/2315-1.htm

So that settles that, it seems. Putting Pleco 2.0 on Palm was always more about taking care of our existing customers than reaching out to new ones, so it was an unfortunate choice only in that it took longer than we expected and delayed our supporting other platforms like iPhone - we've known for a while that it was going to be a money-loser for us, and even if the Pre were backwards-compatible with old apps they'd look pretty awkward compared to native ones.

 

[Jim]

I think Palm is probably smart to not include backwards compatibility. As Mike said the old apps would look pretty awkward. Most of the more popular apps will be rewritten for the new OS and will be improved to make good use of its advantages. It remains to be seen if the new OS is elegant enough to succeed or if Palm is a dollar short and a day late. How buggy it is and how quickly they deal with the bugs may be a factor in that.

I expect Styletap will work hard to get something out as soon as possible which may or may not work well with Pleco. Also we will have to wait and see how the Pre will handle Chinese characters and input. I would guess the majority of Pleco on a smart phone users will also want to be able to use Chinese in SMSs, Calendar, Contacts, etc.

With most of the early reports being pretty positive I am guessing that Mike is ordering a Pre voodoo doll to stick with pins in the hope that he doesn't have to adapt Pleco to yet another platform.

 

[sui.generis]

Palm has also agreed to work with select partners providing access to deeper levels of the OS (linux) for more functionality than the SDK could provide. What I think this means, for palm pleco users, not pleco's developers, is that legacy support will come to pre 3rd party. Possibly by porting the linux garnet VM that already exists for ALP and Nokia tablets. I only hope that application loading into this emulator is painless.

 

[tianli]

Oh Poop!! ...
Oh well I had a terrific run with my T/X and I will miss it terribly but after upgrading to a 210 series in coming days I suspect I will forget my T/X faster than any 中国字.

Knowing you you have one Mike is recommendation enough ...unless you are planning to buy an iphone some time soon :roll:

 

[pufftissue]

Palm Pre in Pleco's Future?

Hello,

Now that Palm seems back from the dead with its new supposedly easy to develop OS, any plans on Pleco being "re-ported" back to Palm OS? They said that old Palm OS versions won't be compatible, but a "solution will exist". I much prefer Palm to Apple's draconian "we know what you want more than you do" approach to everything. I would go with whatever platform I like if it weren't for Pleco, b/c I like having my Chinese dictionary anywhere I go.

 

[mikelove]

pufftissue / benzhen - sorry for the confusion, I thought this was more of a "Hardware" topic since it concerns a platform we already support (or its successor, anyway) but since you've both started new threads on it in the "Future Products" forum I've now added a pointer to this discussion in there.

Jim - StyleTap certainly might come to the rescue on this, though as with their developed-but-rejected-by-Apple Palm OS emulator for iPhone there's the issue of the OS just not being designed to deal with small, finger-unfriendly controls and stylus-driven interfaces. And even StyleTap probably couldn't do much to bring over some features like Instant Access, which depend on other software on the phone also running "old" Palm OS.

sui.generis - I imagine they might eventually roll out full, official native development support just like Apple did, but like Apple they probably want to wait a year and let their APIs get a bit more stable before they do so. Pleco isn't really very usable in Garnet VM; some of that has to do with the awkward screen scaling (which isn't a problem on the Pre since it has the same screen resolution as a landscape Palm) but a lot of it has to do with the fact that it doesn't support VFS / flash memory cards yet, so there's not really any way to get all of your Pleco data files loaded into it.

tianli - I already have an iPhone, in fact just this week I've been able to start productively using Pleco on it. But I like WM better for a lot of things, and given both Microsoft's corporate philosophy of obsessive backwards-compatibility and the fact that WM's single biggest selling point for corporations and developers is its use of Windows APIs, I don't think we're likely to see a similar old-software-support-dropping makeover of WM anytime soon. (I have heard murmurs of their switching to .NET only in Windows Mobile 8, but that's a long long way away and wouldn't be as nearly big a shift as Palm's HTML/JavaScript-only move)

pufftissue - a full version of Pleco on webOS is unlikely for the reasons I've described above, though if we get the licenses together we could certainly do a "Lite" internet-connection-dependent one, and if they ever do add native development support we could consider a port then. I've got my fingers crossed that Google will decide they need to start allowing C development on Android in order to complete with the iPhone in areas like gaming, and if they do that would make Android our most likely new mobile platform target.

 

[ipsi]

Damn, that looks very cool. Might just have to buy me one of those... Javascript development though... Bur, that's not exactly making me wild. I'd vastly prefer Java to that. In theory, it should be fine, provided the development environment is good. Most of the pain I have encountered with JS is from a poor IDE, and the fact that each browser has it's own bugs and strange ways of doing things. That, and testing in IE makes me want to kill things (Firefox at least has Firebug). Given that testing should be relatively easy, and there's only one platform to target, that might not be so bad. I also dislike trying to develop and lay out an interface, so the less I have to worry about that, the better.

I may just have to get this phone, regardless of how much I've recently spent on Treo Pro, just because I can.

Things I need to know though: Unicode support? Ease of creating Input Methods? RAM?

 

[ipsi]

From what I have seen in the CES video post on Treo Central, it's what the iPhone should have been, in every respect. If you haven't watched it, you should. The thing that gets me the most is, simply, the synergy. I want to have that on my Treo Pro, but it's simply not going to integrate as well as that. I love the idea of having a single conversation thread between text message, IM and whatnot. Very cool. Notifications that don't steal the focus is a good idea.

I hope this synergy will result in the ability to see my text messages and email in the same inbox (while retaining their own separate ones). Be surprised if it didn't.

Another thing comes to mind: They're allowing people to buy and download Amazon MP3s directly to the device - I hope this means that Amazon is working on making their MP3s available globally. It's always annoyed me that it's USA/Canada (I think?) only.

Dammit, I'm basically sold. I just need Pleco and my eBook reader on it, and I'm set.

Already can think of several features for Facebook that could work very handily with the Pre/WebOS - notice the search boxes when typing? Why not update Facebook Status from there as well? Also, designating special albums for automated upload to Facebook Photos or Flickr or whatever. Dammit, need to buy one.

I wonder how easy it will be to integrate other social networks in a similar fashion to Facebook? Or Google? Assuming the expose the appropriate API, I hope it's easy to make it as seamless as Facebook et-al.

All I need to do now is find some friends so I've got an excuse to use this.

While I'm here, what is CDMA's 4G technology? Listening to Sprint's CEO, I wonder what they'll do when AT&T or T-Mobile start rolling out 4G networks... I know that our local incumbent (Telecom) is rolling out a 3G network and dumping its CDMA network.

Sorry if the above seems a little disjointed, kinda wrote it while trying to listen to the CES Keynote. And the Keystone is another wonderful gadget that has been too long in coming. Other companies had better start coming out with them damn quickly. A bit sad that it doesn't come standard, but there we go. Can't have everything.

Anyway, as far as I can tell, the only thing that the iPhone has that this lacks is actual C [well, Objective-C] development, the app store, and access to iTunes. Admittedly, those are pretty big things to be lacking, but hopefully they'll be remedied. Email support seems better, multi-tasking is present and available for everyone, it supports multi-touch, seems to support more gestures, and in general seems better. It's significantly more open, which I like a lot. MMS as well, but that's a non-issue for me. A flash for the camera, but again, a non-issue for me.

Anyone else care to weigh in with stuff the iPhone has that this doesn't, or that this does that the iPhone doesn't? Every phone these days seems to be compared to the iPhone, but this one seems to be made for that - it looks similar, and is heavily consumer-oriented with fully-integrated social networking and all the rest. Yeah, I'm fairly impressed.

 

[radioman]

4 G technology, if you do a google search on "3GPP LTE path" you will find a lot. LTE is the path to which many GSM operators around the world are evolving. Verizon, a US based carrier CDMA (IS95) technology, has decided to follow the GSM world and move to LTE as they see better opportunity and scale there. Also, it helps that they are part owned by Vodafone, a major global operator (as you probably know) who's systems are generally based on the GSM standard. Sprint is CDMA, but I believe is going down a WiMax path. http://news.cnet.com/Sprints-WiMax-dile ... 12618.html has a recent article on WiMax and some challenges for Sprint.

As for the iPhone, I am a huge evangelist. As for the Blackberry Storm, an interesting recent competitive product, according to many of the reviews, it seems firmware is not there quite yet, and they do not have the iTunes-like store (e.g., mega convenient, some call it "the killer app" for the iphone). For my Blackberry, I need to go out, find the software, find a review of the software, then try to get it downloaded. It is just not simple or fun. But RIM makes good products and mobile software. I am hopeful they will make improvements - I am all for competition.

If you are an experimenter you can get MMS and multitasking on the iPhone by jailbreaking it, but you risk your warranty and some jailbroken apps are unstable. But a lot of people have gone this route.

As for Palm, If they can get momentum, develop a great store, etc. all the better.

 

[Dr.Grace]

As a longtime Palm phone user, and a new convert to Pleco (it's fabulous--can't live without it!) I'd add my plea for a version for the Pre, if at all possible. I understand the awkwardness of not having an official stylus, but there are capacitative styli available for the iPhone. I would NEVER get a windows mobile phone unless it were the only PDAphone left on the earth. However, I will probably hang onto my Centro as long as possible and not buy a Pre unless there's a version of Pleco that runs on it.

Don

 

[mikelove]

Character set support is one of the single biggest questions on the pre, actually - I assume since Apple includes full Unicode fonts that the pre will too, but if it doesn't that would essentially kill the possibility of anyone releasing a decent Chinese dictionary for it; you might be able to come up with something where you'd locally store bitmaps of characters and retrieve them when you pulled up an entry, but it would be extremely awkward and slow. Handwriting recognition should be possible regardless thanks to WebKit's support for the <canvas> tag, but that would pretty much have to be server-dependent (or done in native code in Linux); even if JavaScript performance were fast enough (and I suppose it could be for a rudimentary engine - ran OK on 16 MHz Palms after all) there's no way anybody would put their code for that in a format where it can be easily read / copied by others.

Which brings me to another problem with pre development - since it's all in HTML/JavaScript, the source code is there for anyone to see. I suppose Palm might rig up some sort of system to compile JavaScript into bytecodes and/or encrypt it, but I'm not sure if I'd trust it with sensitive code for something like DRM; after a year-and-a-half Apple still hasn't managed to stop people from unlocking / jailbreaking iPhones, and their engineers have years of experience with this stuff thanks to the iPod. This won't stop people from developing applications for the pre, but I think you'll find that most of the really cool / innovative ones are either open-source or heavily server-dependent - people aren't going to put their painstakingly-developed algorithms in a format where anybody can just rip them off. For our purposes (and I suspect that of most of the eBook reader companies) there'd only be one or two dictionaries we could make accessible offline due to the lack of DRM - most of our license agreements specifically require that we keep the data encrypted and there'd be no way to do that on a pre except by storing it all on a web server.

Personally, now that it's actually getting to the point of being semi-usable (at least in the dictionary part) I'm getting really excited about the iPhone version - there are things we can do (and are already doing) in that we can't even do on Windows Mobile, let alone a pre, and assuming the iPhone continues doing well market-share-wise (as I suspect it will, particularly given the likelihood of a US $99 model appearing at some point this year) and Apple doesn't give us a lot of grief with approvals / feature limitations I think we may stick with that, Windows Mobile, and their two desktop equivalents (Windows and OSX) for a while - other platforms like Android, BlackBerry, and webOS could be supported with a light / web-dependent version while we wait for the market to sort itself out and a clear winner to emerge (for which we can then develop a dedicated / more powerful version).

 

[Jim]

The best aspect of the Palm Pre may be that it spurs Apple to improve the iPhone hardware, firmware and licensing.

 

[Azabu]

Some interesting insights from Pandora on developing on webOS at PIC:

http://www.palminfocenter.com/news/9685 ... for-webos/

Well worth a read.

 

[ipsi]

There's going be no compiling, I suspect, as there have been screenshots of the logs from the IDE posted and it's being powered by Jetty. Which is good for news for us J2EE developers. Not so good for those hoping for C/C++ development possibilities.

As to DRM, I guess we'll have to wait and see, but as long as it's sufficiently hard that would be ok, I imagine. DRM can't ever be perfect, and Apple will never be able to stop people jailbreaking their iPhones, provided they're sufficiently determined. But that's getting very far off topic.

Anyway, I hope the phone does well, and I may end up buying one for myself, if only to play around with and show off.

I'm also wondering how much of an opensource community will pop up around it - there are very few bits of OSS for WM or Palm. Slightly more free programs, but still not many. Not even close to what's available in desktop-land.

 

[mikelove]

Jim - amen on that; personally I'd love to see them either officially support jailbreaking (with some sort of a configuration setting, perhaps requiring a phone reset) or streamline / lift the 200-device limit on Ad Hoc distribution. Given how overcrowded the iTunes store is now, this would give them a way to streamline that catalog while still letting people access any application they want as long as they're willing to go out and look for it.

ipsi - "Sufficiently hard" is the main problem, I think - if your DRM algorithm is available in cleartext JavaScript code it'll likely be almost trivially easy to break. Now that Nokia's put Qt under LGPL I think there's a chance they might actually be able to get a decent number of open-source developers on Symbian, though in general I think the prospects for Symbian are grim, and for a whole lot of reasons it's a terrible platform for our purposes.

I don't think that open-source is going to take off on mobiles anytime soon, though. Mobile software coding isn't very much fun, to be honest; it's a lot harder, there's a lot less cool stuff you can do with it, and between the locked-down nature of some platforms and the difficulty of finding / installing software for even the more open ones, the download numbers can be very disappointing - there are few things sadder in the software world than a beautiful, well-designed open-source project that nobody uses.

The pre's prospects specifically for open-source are pretty grim, most of the interesting things one might want to do with the pre are going to rely at least in part on a web server somewhere, and while an offline open-source project can be posted on SourceForge and cost you essentially nothing to maintain except your time, with a server-dependent one you could end up stuck with a big hosting bill too.

 

[ipsi]

If developing for the Pre is as easy and as powerful as Palm claims (I'll wait until I see the SDK before accepting that), then it may be quite easy and fun to develop for. The iPhone is probably the easiest platform available today to develop for, as I understand the SDK is large, powerful, well-designed and well documented (right?), and there's two platforms: iPhone and iPod touch. WM is rather the opposite, in many ways, and there are dozens, if not hundreds, of devices you need to buy and test on to make sure your application works as you expect. This is especially true if you're doing any sort of system hacking, or making use of the various wireless connections available.

If it does end up that most cool applications need a back-end server then yeah, little to no OSS stuff, and most stuff will involve a monthly fee.

The majority of it [OSS] will, I suspect, be for the various social networking sites that aren't supported out of the box - MySpace or LJ or whatever. Assuming they don't develop their own clients for it.

I might also disagree that there's less cool stuff you can do with - mobiles might not be as *powerful*, but they've evolved a lot more than desktops have. I still interact with my desktop the same way I did 10 years ago - with mouse and keyboard. Sure, I can do more, store more and generally get more out of it, play prettier games and all the rest, but mobiles have acquired touch-screens (as the primary interaction mechanism), something that would be very hard to put on a desktop PC or laptop. They've started integrating stuff like GPS, accelerometers, proximity sensors and more. Mobiles have added a lot of cool stuff, relative to what desktops have over the last few years. I don't know if that offsets the added pain of developing for a mobile but even so, I'm not sure I'd say there's a lot less cool stuff that can be done a mobile. Admittedly, it does depend on what you define as 'cool', but there we go.

As soon as someone can invent a battery that has a significantly better capacity for its weight and volume than a LiOn battery, then mobiles will be able to push a whole lot of interesting stuff much more heavily than they can now (sure, we could run GPS, Bluetooth and 3G 24/7. If you don't mind charging your battery every 2.5 hours).

Also, Palm is showing promising indications - they've put up a developer blog here. Maybe that's an indication of more openness. Maybe it's just a sop to people clamouring for *something* from Palm. I like it though

 

[mikelove]

The iPhone's easy to develop for in the sense that it has a lot of powerful capabilities under-the-hood (SQLite built-in, lots of wonderful POSIX functions like mmap() that not only work but work well, excellent and extremely-powerful 2D and 3D graphics rendering interfaces) but it still has a lot of issues - there are some things that would take literally one line of code to do on WM that require extremely complicated workarounds on iPhone because they don't have APIs for them yet, and the documentation still has a lot of errors / references to things which are only available on regular OSX.

I see your point about "cool stuff" on mobiles, but the problem with a lot of these new capabilities is that they're very hard for programmers to take advantage of. Even on extremely-open Windows Mobile, hacking the OS to add new features is difficult enough that programs like CE-Star still don't work very well. Most of these capabilities also tend to be extremely device-dependent, and no open-source developer is going to want to have to buy a dozen different phones just to get their software working on enough devices to get a decent number of users. It seems like every mobile platform is either a) extremely locked-down (iPhone, BlackBerry, Pre) or b) runs on too many different devices (Windows Mobile, Symbian, soon Android), so your choice is to either not be able to add the new features you want to add because Palm / RIM / Apple won't let you, or not be able to to add them because all your users are asking for is for you to add / improve support for their particular phone model. And of course the mobile market itself is so fragmented that even if you support half of the devices on a particular platform that's still only a tiny little slice of the entire market.

 

[BERBINAU]

I purchased "Pleco" a couple of years ago, as had learned Chinese, and go to China regularly.

I went to China on business last summer, and I used "Pleco" everyday on my Palm T3 ! (in fact every 15 minutes or so...). Can't live without it... "Pleco" is a wonderful product, powerful, very well designed and user-friendly.

Mike, I BEG you not to give up on webOS (I was thrilled by the announcement from Palm to release a product like the Pré- at last...), and I am planning to buy a webOS-running device once it it released, to replace my aging Palm T3.

Palm is not stupid, and hopefully there will be some way to run Garnet apps on webOS, or ways for developpers like you to program securely for webOS. Furthermore they can't ignore the Chinese market, so I am sure they will have Unicode fonts on webOS. I guess Palm has chosen to make such announcements later, at a more relevant time.

Best regards and good luck,

Pierre BERBINAU (from France)