Google Android

mikelove

皇帝
Staff member
sui.generis - thanks!

To be honest I almost think Google are shooting themselves in the foot pushing Android over Chrome OS in tablets - Google wants a world where everything is browser-based, they do a lot better in that world than one with an open-source mobile OS that carriers can muck around with to thoroughly de-Googleify it (AT&T's Android phones default to Yahoo for search, and many Chinese ones to Baidu); Google versus just about any other company in the web-based app / AJAX / etc space isn't even a contest, it's only offline that they become vulnerable to competition. Particularly with the escalating mobile ad wars - how long do you think the carriers are going to continue letting Google take a big fat cut of advertising sales in mobile apps running on their phones? On the web they don't have a choice - if you use Google search or apps then you get Google ads - but what's to stop Verizon from forcing Android apps that run on their phones to use their preferred ad network? (as Apple appears poised to do in OS 4.0)

On Google Editions, I don't see this doing that much for Android simply because you can't yet build a device that has both a good enough screen refresh rate to run apps and a screen that's comfortable to stare at for hours on end; the closest attempt so far would be the Android-powered Nook reader, with a big e-ink display and a little color LCD at the bottom, but nobody really wants to run an app on that that's confined exclusively to the smaller screen. Next-generation, LCD-like e-ink displays could appear on both the iPad and the Google tablet in another year or two, but until then, neither Google's nor Apple's tablet is likely to displace the Kindle for e-book reading. Heck, out of all of the iPad buzz, how much of it actually concerned the iPad's use as a book reader? If you go to the Apple Store and watch people playing with iPads you're not going to see a lot of them reading books.

mfcb - well it's tough to know where everything will land in the timeframe you're looking at; the new iPhone could crash and burn, or Android could fragment itself into oblivion; heck, in three years the dominant smartphone platform could very well be Windows Phone 7 :) (not likely, but three years ago nobody had even heard of Android)

Zeldor - they're literally giving away a free Android phone with every Android phone purchase on Verizon, so this isn't really at all surprising. And as I've said here a few times, market share numbers are a terrible way for us to make decisions - if market share mattered, there'd be a lot more posts here advocating for BlackBerry and Symbian versions than for Android (though BlackBerry at least continues to generate significantly more email than Android does).
 
On Google Editions, I don't see this doing that much for Android simply because you can't yet build a device that has both a good enough screen refresh rate to run apps and a screen that's comfortable to stare at for hours on end

Do you consider LCD-based screen techs like mirasol and Pixel Qi to be problematic in regards to comfort? I refuse to buy eink just because of the cost to utility ratio -- and frankly, because I like backlighting (I read at night every night, and frontlighting solutions suck). But for starters, I think viewing comfort claims about LCDs are overblown (I've been ereading avidly for 7 years; I've eread almost exclusively {sans some textbooks and a few nonfiction that stubbornly refuse to epublish} for 3 years, all on LCDs. It's all about turning the backlight down and avoiding the color white.), and with newer, better transreflective techs, I'm starting to warm to the idea of a good multimedia tablet that would serve as my primary at home reader (where I do most of my ereading anyway). One nice thing about the iPad is that mass market tech coverage has finally become convinced that LCDs can be used for reading. One might have realized this since most people read off LCDs 8 hours a day at work, but whatever.

Impossible to know if Google's own tablet will use something like Pixel Qi, but it would certainly be a nice way for them to set themselves apart from the iPad. Mass market journalists would probably hail it as revolutionary too, giving it an extra boost, even though it's already appearing in laptops and Notion Ink announced their Pixel Qi using tablet last year. It hasn't appeared in any really big name product yet.

how much of it actually concerned the iPad's use as a book reader? If you go to the Apple Store and watch people playing with iPads you're not going to see a lot of them reading books.

A ton of it. Apple has already succeeded in completely upending the way epublishing works. Amazon ruled things before, but since the iPad's announcement, Apple has managed to bolster publishers in such a way that everyone, including Amazon, has had to accept something called Agency Pricing, something largely credited to Jobs offering publishers a ealternative if Amazon decided to get rough, and giving Amazon a reason not to get rough). Now I run in a lot of ereading circles, so my perspective may be skewed a little, but there's no small amount of buzz regarding the iPad as a reading device.

And as I've said here a few times, market share numbers are a terrible way for us to make decisions

Indeed. But Android's momentum is an undeniable plus towards its support. Your RIM/Nokia arguments are fairly flawed, precisely because marketshare alone is a bad measurement. It should be marketshare of what? What's the install base of TOUCHSCREEN blackberries (I could be mistaken, apologies if so, but don't the vast majority of BBs not have touchscreens?). What's the Nokia install base of users for whom English is their native language? What's the install base of both who have purchased at least one 3rd party application?

I love telling people, especially those new to the smartphone world, that Apple isn't the biggest smartphone player, or even second (jaws tend to drop). Who is? yes, that's right, those boring Nokias with virtually no US footprint, followed closely by those boring blackberries (no offense intended to BB or Maemo or Symbian users). And yet, when it comes to 3rd party application sales, Apple is indeed on top.

Android is often pitted against Apple while ignoring Symbian/maemo and RIM for good reason -- the two groups have very different niches they chase. Nokia could care less about backwards compatibility; frustrating users who want to make application investments and developers who want to make sales. RIM users have largely pigeon-holed the device for email, sometimes because their company doesn't want them doing personal stuff on it, and getting a second device for mobile computing fun. Android and Apple are different.
 

mikelove

皇帝
Staff member
Notion Ink's tablet is Android-based too, but it's an interesting question whether Google will want to use Pixel Qi - on the one hand, they're very worried about Android fragmentation, and using an unconventional display technology that developers haven't had much of a chance to test on (and would definitely need to optimize around - all of the brilliant UI rearrangement code in the world can't overcome the fact that graphics / UIs designed in color tend to look like crap in monochrome) isn't going to help on that front, but it would be a way to differentiate themselves from Apple.

I do find e-ink more readable than any LCD-based technology myself - the fact that I spend so much of my day sitting in front of an LCD is precisely why I don't want to then spend even more time staring at one when reading. If it weren't so utterly impractical I'd want to develop a version of Pleco for Kindle just to be able to use it myself on that gorgeous screen.

I just don't think Google Editions is going to be that tied to Android, though - Google's aspirations for it are a lot bigger. Android for all of its buzz is little more than a sideshow for Google's main business of controlling / indexing the world's information; Google wants its iPhone / iPad reader app to be just as spectacular as its Android reader app because Google wants to control the e-book market, and the "information market" in general. If there were some sacrificial altar at which Google could burn Android in exchange for Google-search-like market share in ebooks and mobile advertising they'd do it in a heartbeat.

As far as Apple upending the publishing world, I think that's more a reflection of potential than reality so far - I haven't seen any numbers yet to suggest that the iPad / iBook store is actually starting to occupy a significant portion of the ebook market, though I suppose it's still early to tell. But Apple comes along as the big fish in the music download industry with a hot new tablet on the horizon and naturally publishers fall in line - it'd be stranger if they didn't. Buzz does not equal sales.

As far as Android versus RIM/Symbian, I still haven't gotten a satisfactory answer to my question from a page or so ago about why there are so few Chinese dictionaries for Android - in some ways I view that as a better reflection of Android's potential than anything else we're talking about, because there has to be some reason why there are something like a hundred different Chinese dictionary apps for iPhone OS now and scarcely any for Android, some reason why none of those iPhone developers seem to have jumped on board with Android yet. Are none of the people making Android dictionary apps making money off of them, are Android users uninterested in paying for apps in general, or is there some other reason we're all missing? Is there wisdom in all of that collective decision-making or is every Chinese dictionary developer as irrationally anti-Android as I am?
 
mikelove said:
I just don't think Google Editions is going to be that tied to Android

I don't think it will be tied down to Android at all. I do think it will have a huge effect on Android anyway. As mentioned before, Amazon (and borders has koboware for other platforms, and B&N has nookware etc) makes kindleware for BB and the iPhone, and probably will for Android and WinPho soon too. That hasn't stopped Kindle editions from selling millions of Kindles for Amazon. Google doesn't tie any of their services too closely to Android (I'm sure even Goggles will be multiplatform eventually), but Android does exist largely to ensure Google services can't get shut out of the mobile marketplace through proprietary interference and all android devices benefit from Google's execution/integration of those services.

What I do think will happen is that alongside the Google Editions store Google will highlight a Google-branded tablet, raising Android's profile and bringing in a lot of new blood (perhaps unknowingly) into the Android-application capable install base. They haven't yet managed to sell a ton of Nexus Ones direct, thanks to US habits concerning cell purchases (it's easier to buy from carrier, with subsidy), but ereader purchases are different. I wouldn't be surprised to see Google partner up with WalMart for tablets either. Google is frustrated with carriers, no doubt. Tablets are an arena where they can exercise a little more independence.

I'm not trying to browbeat you into agreeing with me (even I am only a little convinced I'm right), just wanted to point out that my highlighting of Google Editions wasn't because I thought the store would be locked into Android, but that it would bring heavy dividends to the platform notwithstanding the ability to buy books to read on other platforms.
 
mikelove said:
As far as Android versus RIM/Symbian, I still haven't gotten a satisfactory answer to my question from a page or so ago about why there are so few Chinese dictionaries for Android - in some ways I view that as a better reflection of Android's potential than anything else we're talking about, because there has to be some reason why there are something like a hundred different Chinese dictionary apps for iPhone OS now and scarcely any for Android, some reason why none of those iPhone developers seem to have jumped on board with Android yet. Are none of the people making Android dictionary apps making money off of them, are Android users uninterested in paying for apps in general, or is there some other reason we're all missing? Is there wisdom in all of that collective decision-making or is every Chinese dictionary developer as irrationally anti-Android as I am?
Last time I checked there were about 20 different devs offering offline chinese-english and/or english-chinese dictionaries on the Android Market. Judging by the download numbers of the paid apps it certainly seems that none of those devs are making much money. Admittedly most of those are using generic bilingual dictionary engines - i.e. are not specifically geared to support things like searching by pinyin (but then again KTdict doesn't do this properly either).

How to explain this? Well, last time I checked, the AppStore has about 5 times as many apps as the Android Market and about 10 times as many devices (read: more money up for grabs per app). You say there are about 100 chinese dictionary apps for iPhone OS but when I checked I only saw a few that were reasonably popular. So there might as well be only a few there. I was reading an Android vs iPhone article the other day and they noted the number of apps as a major advantage for the iPhone. But, for me, its really a red herring. iPhone apps win not because of the numbers but because of the quality and that comes largely from the investment from big companies. Android apps will get that level of investment soon too and when it does you will see the best apps on Android out do the best iPhone apps because of the advantages the Android platform has to offer.

But anyway, back to your original question, chinese dictionaries make up about 0.1% of both markets so I don't think this should be very surprising.

Out of interest, can one of the iPhone guys here explain why the chinese dictionaries get such bad ratings (never more than 3.5 stars)? Clearly they are more decent than that. Are rival devs marking down apps etc? Also, why does Pleco have so few comments? Some of the Android chinese dictionary apps have hundreds of comments even after being out for only a few months.
 

numble

状元
Most iPhone Apps have lower scores because users are asked to rate it when they delete it from their device, and users that delete apps usually would rate it lower.
 

mikelove

皇帝
Staff member
Well that's true, most of those iPhone offerings aren't very good - one of them actually seems to have been so impressed by Pleco that he's been busily ripping off our user interface, using the same buttons in the same places to do the same things. I disagree that big companies and big investments are the key to successful mobile apps, though - the situation with us spending an order or two of magnitude more money and time on our products than any of our competitors isn't typical elsewhere in the mobile space, and it may not be as much a matter of the big mobile developers deciding to invest more in Android as it is a matter of more individual developers learning how to produce high-quality Android apps; the improvement in quality among little put-together-in-a-week-or-so iPhone apps compared to even just a year ago is pretty massive.

I agree completely about the app count not mattering - in fact I think that may become a problem for Android down the line, Apple at least has a framework in place to reject low-quality apps / raise their standards with time (and has already taken the associated PR hit for doing so - at this point it's unlikely that any more app rejections are going to make much of a difference in terms of iPhone-to-Android switchers) while Google is philosophically committed to approving everyone; in a few years time there could be 500,000 Android apps and still be 200,000 iPhone apps, assuming Apple uses the OS 4.0 / OS 5.0 / etc upgrades as an excuse to weed out abandoned / incompatible apps like they did with OS 3.0, further increasing iPhone developers' profits and reducing Android developers'. At some point Google is going to have no choice but to start rejecting apps, or at least offer a "premium" Android market that requires an extra, harder-to-obtain level of approval.

iPhone apps seem to get fewer comments than Android apps in general - I'm not sure why, it might have to do with Apple's aggressively pushing the star-rating interface to a point where people then don't bother to actually type up a review to go with their rating. The fact that the star rating count gets reset with every update (not something I necessarily disagree with - big redesigns often do make software worse - though it would be nice if Apple differentiated between major, design-changing updates and bug fixes), and that in desktop iTunes only reviews for the latest version appear by default, might also have something to do with it. Does Android Market separate comments by specific local markets? If not, that might be another explanation right there - our reviews are scattered between the US, Germany, China, HK, Singapore, etc app stores.
 

mikelove

皇帝
Staff member
Nice things, but very very carrier-dependent - they'll be great in countries with less tightfisted telecoms, but tethering in the US always comes with a hefty fee, as does WiFi hotspot service. If this pushes carriers to loosen up on that stuff, though, that'll ultimately help iPhone users too - if Verizon were to offer, say, free WiFi hotspot service for every data plan, AT&T and Apple would have no choice but to do likewise.

Faster Java execution should be nice for certain applications, but I don't imagine it'll make much of a difference for most reference apps since they're already using the native-code SQLite. Plus they tend to be more I/O- than processor-dependent; we aren't doing a lot of complicated math, but we are randomly accessing several hundred MB of data. What we really need is for some of the technological leaps they've been making with SSDs to trickle down to mobile devices - not quite feasible power-wise yet, but in another year or two the pace of processor speed advances will slow and they may need to start looking at optimizing I/O to keep improving performance.
 

mfcb

状元
mikelove said:
free WiFi hotspot service
available almost everywhere in my hometown since 2 years (Austria,Linz, the city of "Ars Electronica") hehe

westmeadboy said:
There are some great things coming to Android in 2.2 very soon:
...
The USB tethering ...
does this mean connecting your PC to the internet via the phone connected with USB? even my nokia 6120c can do that... and my touch pro can do it via BT, USB, WLAN (but the WLAN option even when connected to power source empties the battery within 2 hours, hehe)
 
mfcb said:
does this mean connecting your PC to the internet via the phone connected with USB? even my nokia 6120c can do that... and my touch pro can do it via BT, USB, WLAN (but the WLAN option even when connected to power source empties the battery within 2 hours, hehe)

Yes. It's also been available on most Android devices but was never part of core Android. AFAIK, its not available on the iPhone (unless you jailbreak of course).
 

mikelove

皇帝
Staff member
Kind of ironic, actually: back when nobody was paying attention, we could do all of these cool things on our Palm OS and Windows Mobile phones and nobody would care - I personally tethered a couple of Treos and HTC Touches on AT&T without being asked to pay for anything more than a regular data plan. Then we lost those abilities in 2007 when the iPhone came out and now we're slowly getting them back bit by bit...

So I definitely see mfcb et al's point: occasionally-frustrating UI notwithstanding, a 2006-vintage Windows Mobile smartphone could multitask, run unsigned apps, tether by Bluetooth or USB, run applications off of an SD card, be programmed in native or managed code or any reasonable combination thereof, and it even had a very nice version of Pleco, all of this when the iPhone was just something whispered about on blogs and Android was a little startup that got bought by Google and then vanished.
 
mikelove said:
Kind of ironic, actually: back when nobody was paying attention, we could do all of these cool things on our Palm OS and Windows Mobile phones and nobody would care - I personally tethered a couple of Treos and HTC Touches on AT&T without being asked to pay for anything more than a regular data plan. Then we lost those abilities in 2007 when the iPhone came out and now we're slowly getting them back bit by bit...

Yup, even on junkier dumbphones. I used a samsung a500 as a 3G modem back in the good ole days when carriers barely knew what tethering was. At that point, it was the fastest internet I'd ever had. Towards the end of my relationship with Sprint, they started doing some screening for that type of thing, but weren't very sophisticated about it. I got cut off once while using a super cheap data plan, called them up and was back on the same plan within minutes.

I wouldn't even attempt something so unofficial these days for fear of fees and what not. But times were different back then.
 

mikelove

皇帝
Staff member
Have any of you Android developers played around with SWIG? It seems like that might automate NDK coding to the point where it could actually be less work than rewriting the whole program in Java, though from what I can read there are a few areas of Java (specifically something called "weak global references") that SWIG requires and Android doesn't implement yet. Unless they're now doing so in 2.2.

Incidentally, while Pleco for Android is still no more than a maybe, and nothing about 2.2 seems likely to change that, I can say right now that if we do do a version of Pleco for Android, it will almost certainly require 2.2 or later - with the timeframes we look at for development there's really no reason for us to go to the trouble of supporting anything but the newest version of Android in our first release (whatever version that happens to be when we release it - I suppose 2.3 or 3.0 or something could be out by then, in which case even 2.2 might not be compatible).
 
I've never used it, but I checked the 2.2 release notes and can't see any reference to the, er, weak global references. I assume you've already googled for others who are already using swig with android.
 

mikelove

皇帝
Staff member
character - I saw that post, but there isn't anything in the NDK r4 or SDK 2.2 release notes to suggest that they actually added it.

I tried running Pleco's cross-platform engine code through SWIG and it seemed to churn out Java interfaces quite happily, but the result is really an unholy mess - lots of awkward arbitrary not-quite-Java-standard objects, and after several hours I couldn't even get it to the point where it would tell me whether or not it was using Weak Global References that it wasn't supposed to.

However, the project to get the native-code Qt framework running on Android seems to be picking up some steam - might even be getting some official Nokia support - so that's another possibility for an Android port; supporting Qt would also let us do Maemo and Linux versions (and maybe even Symbian with the new Qt-based release they're working on), and it looks like it's implemented as a regular NDK library so it should be compatible with almost any Android phone. (normally I'd be wary of a non-standard UI framework like this, but in Android's case the built-in UI controls aren't exactly thrilling me anyway)
 

Giraffe

举人
I'm another person who created an account just to post to this particular thread. I've been using Pleco on my Tungsten C for the past year and it has quickly become central to my Chinese learning effort. But I'm also getting ready to upgrade my gawdawful Windows phone to a new Android (maybe the HTC Incredible) and I'm struggling with the fact that I'd still have to drag around my clunky Tungsten because I can't run Pleco on Android. I don't use all the fancy features, but I really do need the ability to add flashcards straight from the dictionary (and adding flashcards from the Reader module is also sweet).

For me, buying an iPhone is out of the question. Although I think the Apple designers are totally brilliant I despise all the proprietary crap that surrounds the platform - the cables, iTunes, the App Store, the censorship etc.

Having worked in the software industry myself, I also sympathize with Mike's situation. Supporting multiple platforms can a big suck on resources, even for the big companies, so it's not a decision that can be taken lightly. But I would be very happy if it did happen.
 
Top