Google Android

[character]

Depending on who you believe, Google either was losing the fight in China against Baidu or not:
http://news.softpedia.com/news/Google-D ... 2465.shtml
http://business.rediff.com/report/2010/ ... months.htm

If it was losing, then pulling out makes some business sense, as their censorship combined with all the other controversies such as that over Google Books, their 'don't be evil' slogan was viewed with derision.

Worst case for Android is it gets spun off to the Open Handset Alliance and loses some Google marketing muscle.

 

[mikelove]

westmeadboy - Apple's already playing ball on that front. My whole point with gray-market imports was that if Chinese manufacturers were discouraged from making Android phones, there'd be less likely to be a particular Android phone that was enticing enough for people to be willing to shell out hefty premiums to get it as they are with iPhone.

The point of Android, as for all of Google's many money-losing businesses (so much like Microsoft in its heyday), is to push people towards their highly profitable search business, so no search => little reason for Google to be interested in improving Android's support for Chinese. Which I suppose could actually create a market opportunity for Pleco if it means that Android never gets, say, built-in Chinese handwriting recognition support, but it also increases the odds that Chinese support on Android will continue to be heavily dependent on third-party software, and we all know how well that's worked on Windows Mobile...

 

[gato]

Which I suppose could actually create a market opportunity for Pleco if it means that Android never gets, say, built-in Chinese handwriting recognition support, but it also increases the odds that Chinese support on Android will continue to be heavily dependent on third-party software, and we all know how well that's worked on Windows Mobile...

Well, Android is open source, so it shouldn't be so hard for a third party (say China Moblile) to add Chinese support, as it is for Windows Mobile. Right?

This comment has a good chance of being right:

http://gigaom.com/2010/01/14/will-andro ... navigation
Knows China Thursday, January 14 2010
Google already has lost control of Android’s destiny in China. The biggest carrier, China Mobile, uses their own version of Android, called OMS. They do not have to use any Google services on their devices to use the Android stack. China Mobile and Borqs (their developer) have hundreds of developers working on Android and OMS. Now, remember that while on the surface the phone companies look independent and competitive they are in reality all run by the government. The government can have China Mobile “license” its version of Android to the other suppliers and the Chinese are on their own branch and development path, free to pull Google improvements, but not dictated to or dependent on Google. Handset suppliers will put whatever OS the carriers want on the device but more interesting is the fact that Chinese manufacturers are building a LOT of Android devices.

In the end, China gets what ti wants: independence form US companies, a software base licensed through China and a market where the handset suppliers all are Chinese or all the devices are built in China. Pretty clever way to close the market without violating WTO rules.

By the way, source.android.com is blocked in China right now, while android.com is not. See ping chart below.
http://just-ping.com/index.php?vh=sourc ... vhost=_&c=

 

[mikelove]

Looks like it's connecting now, so might have just been a temporary server problem.

A Chinese Android branch seems almost destined to break compatibility at some point, particularly since they derive so little benefit from maintaining that compatibility (and indeed improve the situation for local Chinese software developers if they make it harder for foreigners to get their software working on OMS).

 

[numble]

The Google pullout looks like its just pulling the plug on the self-censored google.cn website. Chinese users would probably continue to access a Chinese-customized version of google.com, as well as google.tw and google.hk. Below is a quote from Google CEO Eric Schmidt:

http://www.newsweek.com/id/231117

"And please understand, we will still have engineers, programmers, and others in China. We love China and the Chinese people. This is not about them. It's about our unwillingness to participate in censorship."

 

[gato]

The question is whether the government will block google.com after the censored google.cn shuts down.

 

[numble]

Well, I didn't think it would have affected Android in China, but...

"Google Scraps China Cell Phone Launch Amid Dispute"
http://www.nytimes.com/aponline/2010/01 ... oogle.html

 

[mikelove]

Interesting; seems to increase the odds of Android development in China shifting over to China Mobile's fork of it, anyway. And makes it clear how closely Android and Google's online services are linked as far as their product planning goes. It'd be (delightfully?) ironic if Android phones in China end up using Baidu as their default search engine...

 

[character]

China says Android can stay

"A Chinese Ministry of Industry and Information Technology spokesperson has today delivered a statement affirming China's willingness to allow Android devices to operate within the country without restriction so long as they adhere to the nation's laws."

 

[westmeadboy]

Hi Mike - just found this post about the OPhone supporting the Windows Mobile API. Sounds impossible to believe, but seeing as you have loads of experience with Windows Mobile, I was wondering if you can shed some light on what it might mean????

 

[mikelove]

I'd have to see the documentation to know, but the most likely possibility is that they've done something like Wine (perhaps even something based on Wine); created a compatibility layer that runs Windows Mobile style ARM executables and simulates the core Windows Mobile DLL functions / background processes. Either that or they've built some sort of cross-compiler that lets you rebuild WM apps on Android seamlessly translating Windows Mobile API calls into Android Linux ones, but that area seems to be less well-explored (and, if anything, more likely to cause bugs / incompatibilities).

 

[Lurks]

Well, I've had a Nexus One for awhile now. It's so superior to my Touch HD it isn't even funny. It feels incredibly retro to pull out the Touch HD just for Pleco. I'd really like to be able to retire this. Flash cards on Pleco on iPod would be awesome. I love how you can just go into the market place and install a different Chinese IME and use it right off the bat. I prefer the Google one meself.

Tell you what though Mike, there's a bunch of Chinese apps in the Google marketplace already and they're pretty much all garbage

I find it quite pleasant to make stuff work on the N1, I've knocked up a little word of the day widget to sit on my home screen.

 

[mikelove]

Chinese dictionaries on Android may be mediocre now, but even if we started an Android version today, by the time we actually had it ready, they'd have improved greatly - iPhone Chinese dictionaries were all utter crap too when we started working on the iPhone version of Pleco. (though most of them still are)

I'm still not sure what the deal is with WM APIs - some have suggested now that this may just be referring to ActiveSync support, which wouldn't do that much for us.

Anyway, between the iPad, desktop and/or online version, and WM7 there's really no chance of an Android port in the near future at least (absent someone coming along with a giant wad of VC money) - too much other stuff to work on.

 

[westmeadboy]

About the whole Google phone debate, I came across this article which suggests that Google's Nexus One-style approach may improve the fragmentation issue. Actually, I still maintain that fragmentation is not nearly as much of an issue as some people make out. AFAIK, all of my apps worked immediately on every new device released in the last 3 months (i.e. since my apps were released). The only exception was one person had some trouble when using a custom ROM.

About that article, its true that OEMs don't really have an incentive to update existing devices to the latest SDK and that the Google Nexus One strategy should speed them up a bit. I hope so, my Hero is still running 1.5!

I'd still love to see carriers have less of a (read, "no") say in phone-pricing in the same way that electricity companies don't have a say in vacuum cleaner-pricing! Its just difficult to change people's mindset. Its difficult for many people to believe that a 530USD unlocked phone works out cheaper than a 180USD locked phone.

Ooooh - my 100th post, I'm so excited

 

[Lurks]

Sorry Mike, I guess it sounded like I was advocating Android for you again. I know it's not on the cards and completely understand why. At least on the Apple platform people can buy a relatively inexpensive thing (that isn't a phone) to run it on. I've pointed a bunch of people in the direction of Pleco for iPod and returning to uni (old timer goes to uni to study Chinese - urk!), I'll doubtless keep doing so.

Paradoxically with Flash 10.1 coming to Android real soon, I'll likely be spending a lot of time this year using Skritter on the N1. The heavy Pleco flashcard action has been a little too effective, my reading comprehension is ridiculously advanced of my ability to write so I think I need to put some time into rebalancing that... Well, it's not an option if I want to actually pass exams and stuff.

Desktop Pleco would also be awesome to run on a netbook... then Skritter and Pleco in one place. That would be truly awesome

 

[character]

There's a series of videos on Android development O'Reilly is selling:
http://oreilly.com/catalog/0636920001690 Looks fairly basic, but perhaps it's a painless way for someone to get started.

Android Native Developer Kit:
http://developer.android.com/sdk/ndk/1.6_r1/index.html

 

[thenrik]

Hi:

I'm wondering what your take is on the upcoming Windows7 phone? It sounds like the Palm experience again, dropping previous compatibility and coming out with a product that has the same name but a completely different OS. I have a Samsung Moment Google Android phone but still use my Palm T/X pda.

Do you think the Windows7 OS changes will make Pleco incompatible with the new and future phones?

Still hoping for a Pleco Android version. I'm never buying an Iphone and wouldn't touch "Palm, nothing to do with Palm" Webos.

Tom

 

[mikelove]

Lurks - reading that post actually makes me think about the continued need to learn writing; it seems like this point there isn't really much of a practical reason to learn to write Chinese characters other than for tests / exams, handwritten notes are no more common in Chinese than they are in English nowadays; if you can speak and you can read you can probably use a Pinyin input system just fine.

But of course there's still the fun / aesthetic pleasure of learning to write Chinese, which is probably why people like Skritter - that sense of writing practice being an enjoyable activity in its own right (as far as possible from tediously copying down a character hundreds of times in those darned paper grids) may be something to try to do more of in future updates to our flashcard system, make both the UI and the actual activity a bit more game-like.

character - I've looked at the Android NDK, but unfortunately it's much too cumbersome to make it a realistic option for getting our old code running on Android - there are too many jumps between our cross-platform engine code and OS-specific UI code to make it practical to go through all of that setup each time. But with the apparent downfall of C development on mobiles a full Java rewrite starts to seem more and more plausible.

thenrik - see http://plecoforums.com/viewtopic.php?f=13&t=2173, it does seem unlikely Windows Phone 7 will be compatible with Pleco but we won't know anything for sure until mid-March.

 

[Lurks]

reading that post actually makes me think about the continued need to learn writing; it seems like this point there isn't really much of a practical reason to learn to write Chinese characters other than for tests / exams, handwritten notes are no more common in Chinese than they are in English nowadays; if you can speak and you can read you can probably use a Pinyin input system just fine.

Oh absolutely. So as you say it comes down to the joy of learning to write Chinese. Also it seems to have some other benefits, by forcing a greater knowledge of exactly what is in each character. At least I've found that so far.

... may be something to try to do more of in future updates to our flashcard system, make both the UI and the actual activity a bit more game-like.

For sure! Makes all the difference in the world. In Skritter one thing that surprised me was how the little noises push me along. Like the one it makes when you get a character wrong or give up and get it to show you what the character looks like. Compared to the "great!" message when you get something right. It pushes the buttons of those little pleasure/reward centers which I think are important to learning. Not got any complaints with Pleco's flash cards but it's awfully straight in comparison

Another thing I quite like from Skritter is the progress stats. I think this has been raised before but it doesn't seem very easy to get Pleco to tell you how much stuff you know to gauge progress? With a single flashcard list maybe it'd be easy enough but when you have several... and how many you've learned in a given time frame, retention rate etc. It opens up a whole world of possibility if you get that info up into the cloud for sharing on web. With Skritter you actually get emailed progress reminders and your time studied, characters learned etc are nicely graphed right on the front page. That'd be awesome for flashcards.

Oh and before it sounds like we're saying Skritter has all this stuff Pleco doesn't, I find flashcards great well beyond just learning words. The wealth of information there when looking at words, characters, example sentences etc. Ode for Skritter to even have anything that approximates an actual dictionary entry for each word...

Mat.

 

[mikelove]

Pretty graphs are certainly doable, though we're in a slightly more constrained situation storage-space-wise than a web-based flashcard program would be and so there are some significant tradeoffs to recording more precise statistics about retention rates / number of cards learned recently versus time studied / etc.

As far as general polish / lack-of-boring-ness goes, that suffers mainly because we have to spend so much time juggling multiple platforms; if we'd been able to stick with Windows Mobile and not launch off on the iPhone port the WM version would probably have a beautifully refined flashcard testing UI now. Improvements to things like dictionary search / flashcard selection are easy to justify because they're cross-platform; we've got beautiful pieces of code for those sorts of things developed originally on Palm OS that are now working equally well with very little modification on iPhone. Improvements to the UI, however, tend to be very OS-dependent and difficult to port over.

The iPhone flashcard UI may be a little bit prettier than the Palm/WM UI, but it won't have lots of new sound effects / reassuring graphics, at least not initially, and before we can get around to adding them we may have to turn around and start porting to yet another new platform. Which is the reason why web-based software is so appealing, since you can keep refining it for years and years. Though in Skritter's case, given Flash's rapid fall from grace they may actually have to deal with their own platform-juggling problem in a year or two...

But that's the basic conundrum - you can either have a beautiful / polished version of Pleco, or a version of Pleco that runs on Super Awesome Mobile Platform 2010, but unless you're willing to pay twice as much for it as you do now (and convince all of your friends to do the same) you probably can't have both.