Google Android

mikelove

皇帝
Staff member
Thanks for posting. You're right that it's a big decision, and I get very worried about stretching ourselves too thin - given the choice between making what continues to be a world-beating Chinese dictionary for one platform or an only-slightly-better-than-the-alternatives dictionary for two, I think everyone's better off if we take the former route. Heck, you could argue that the money we might make from a half-hearted Android port would be better off going to a developer that loves and is fully committed to Android and only Android, while Pleco's left to continue pushing the envelope on iPhone with amazing new features.
 

gobi

Member
Hi Mike. I am wondering if you would consider the possibility of opening pleco's source to open source. After all, Android is open source. You still have to make money of course so you can continue to charge for the dictionaries but perhaps the open source community can help in porting pleco to Android. What do you think of this idea?
 

mikelove

皇帝
Staff member
Wish that we could, but there are a lot of problems with open-sourcing Pleco:

1) Much of our DRM security comes through the fact that nobody knows how our DRM system (or even our database file format) works - "security through obscurity" is a fairly sound concept if you're as tiny as we are, the hackers capable of breaking even a weak DRM system have better things to do with their time - so even though the encryption keys would still be private, and even if we perhaps also kept the encryption code private (though that would be tricky, I guess we'd need a BSD license or something that allowed for easy mixing of open-source and non-open-source code) we'd be putting our licensees' content at great risk by exposing the file formats / software internals in that way.

2) The code isn't pretty, to be honest; it's gone through a lot of adventures over the years from its early days on Palm OS, still uses concepts like handles that most developers never have to deal with anymore.

3) We'd still have the burden of supporting the resulting product, since we'd still be selling it, along with guaranteeing it / providing bug-fix updates for it / etc even if the open-source community were helping, but now we'd be doing that with code we didn't know as well / code that hadn't been developed according to our usual conventions.

4) The help is unlikely to be enough to make much of a difference - we'd spend almost as much time coordinating this effort / explaining how everything works as we would just doing the work ourselves. (seriously, there's a TON of work that would need to be done to get Pleco on Android) c.f. The Mythical Man Month.

If you're really motivated to see a better Chinese dictionary on Android, though, why not start an open-source project yourself to develop one? There are a few other programmers in this thread alone who seem equally interested in seeing the Chinese situation on Android improve. There was a project years ago called PADict that came out with a really neat Japanese dictionary for Palm OS (still around at http://padict.sourceforge.net/ actually) - like Pleco, it included handwriting input, stroke order diagrams, flashcards, etc - but I haven't seen anything open-source on that scale for mobile devices since; most of the open-source projects for mobiles tend to be shells for database formats like StarDict and have little ambition beyond that. If you aren't happy with CC-CEDICT, contribute to making it better (checking definitions against other dictionaries, say), or start writing publishers about getting older / more obscure dictionaries released free for open-source use. Flashcards would be easy with a "real" open-source project since you'd have access to the GPLed code for Anki / Mnemosyne / etc; those aren't accessible to us in any case, since incorporating GPLed code would necessitate releasing all of our other source code including DRM.

So honestly, I think the open-source approach could really work great for a Chinese dictionary made from scratch - people get extremely passionate about even tiny UI changes / rare bugs in Pleco, understandable given how many hours a day many of them spend using it, which means you'd likely have an easy time motivating folks to contribute. I just don't think it would work so well for us - juggling content / code module licenses and open-source is very very challenging, and the benefits over a wholly open-source Chinese dictionary project wouldn't be enough to justify the extra burdens.
 
gobi said:
Hi Mike. I am wondering if you would consider the possibility of opening pleco's source to open source. After all, Android is open source. You still have to make money of course so you can continue to charge for the dictionaries but perhaps the open source community can help in porting pleco to Android. What do you think of this idea?

I think it's a great idea!

However, there are also open dicts such as CEDICT, that would make Pleco then completely free and Mike could be this dude: http://s3.amazonaws.com/giles/php_01160 ... r_food.jpg
 

mikelove

皇帝
Staff member
And BlackBerry remains #1 in the US, and Symbian #1 worldwide - never been a big fan of market share numbers. Windows Mobile has 1/2 the market share of iPhone according to this, but for us it's generating only 1/7 of the sales in spite of the total lack of competition.
 

mikelove

皇帝
Staff member
While we're talking market share, though, if you include iPad sales Apple is now the world's 3rd biggest portable computer maker:

http://tech.fortune.cnn.com/2010/08/02/with-ipad-apple-is-no-3-in-portables/

That actually suggests some gains may be coming on Android soon as well once those tablets start to appear - even if Windows 7 manages to grab a decent share of the tablet OS market, if lots of people are buying tablets in lieu of traditional notebooks / netbooks the end result would mean gains for iOS/Android and losses for Windows.

Which hurts the case for a desktop version, but if we developed a version of Pleco for Android we might not even need one:

http://thegadgets.net/technology-news/how-to-run-android-on-windows-7-no-live-cd-required/

This would be Android recompiled for X86, not the ARM version running in the painfully-slow AVD emulation environment, so it might actually be usable enough to replace a desktop version of Pleco altogether.
 

numble

状元
In any case, they should start counting "Android Market" devices versus "iTunes App Store" devices, and not simply phones. First, because iPads and iPod Touches (and future non-phone Android devices) are also a part of the putative market, and because some Android devices, like the Chinese oPhone Android branch that Chinese carriers are supporting, don't put Android Market on their phones (or don't let you install from that market, or something).
 

ipsi

状元
The second article is over a year old, and most of the content is no longer relevant. I'm not sure about the 325-character limit, but I've never experienced a stalled download for any reason, they allow screenshots (only allowing 2 seems a bit limiting), Paid and Free apps are in separate categories, and I have no problem with Google Checkout being the only way to pay. I'm honestly not sure what else I would use.

More to the point, the article seems to have been written when the only device available for Android was the G1... The game has changed an awful lot since then, with much, much nicer devices coming out (e.g. Nexus One, HTC Desire, Milestone, and the Samsung Galaxy S series), devices being released to more and more countries, and Android generally being significantly improved upon, to the point where I'm not sure it's really relevant.

First article is more interesting, Some of the problems he mentions in particular seem to have gone (e.g. Tunee doesn't exist, and ringtones don't clog up the Free Multimedia category).

But yeah - not seeing the prices in my local currency is annoying. I guess I can understand the reason - exchange rates fluctuate (possibly over the course of deciding to buy the app), and then people will complain. Not fun. And allowing people to set different prices for different regions is likely to result in people being royally pissed at the fact that they had to pay more than 'those damn Yanks'. Still doesn't mean I don't get annoyed at seeing prices in YEN, GBP, EUR and USD... Can't talk on the AMEX/Carrier billing stuff. It's not an option for me.

The restriction of paid apps to certain countries is insane, and that probably contributes to piracy rates of Android Apps quite significantly. I'm also really annoyed about the fact that the Amazon MP3 Store is geographically restricted, but that's hardly Googles fault.

All in all, though, I find the user experience significantly better on my Nexus One than on my Treo Pro. There are a few things that are missing, but for the most part they're pretty small compared to what I'm getting out of Android. Given that, I'm not surprised Mike's seeing a 7:1 ratio of iPhone:WM purchases.
 

mikelove

皇帝
Staff member
Zeldor said:
Huh, even netbooks coming with Android now, I did not expect it... http://www.slashgear.com/acer-aspire-on ... 5-0596711/

A good sign, because if Google does this then soon Apple may have to do so as well (the Magic Trackpad certainly suggests they're thinking about it), and for a long time now I've been hoping it would be possible for us to keep writing only mobile software but have it also seamlessly work on desktops. (part of the reason we've spent the better part of a decade dragging our feet about a desktop version)
 

mikelove

皇帝
Staff member
To me it shows the dangers of relying too much on the Android "economy" - it's possible for a device to run Android but not Android Market, indeed an enormous number of devices already do so.

I think that the endgame for Android fragmentation may be a decentralized system more like Windows, where applications are purchased on individual developers' websites rather than through a central catalog - even on iPhone it's getting harder and harder to sell apps without an already-established brand / market, so Apple's 30% cut is becoming more of an entrance fee than a sales commission, but unless Google follows up on today's "evil" no-net-neutrality-on-wireless-networks announcement by officially locking down future versions of Android to prevent the installation of non-Market apps, developers who aren't relying that much on Android Market to generate sales are going to start to abandon it just like all of us Palm / WM developers started to abandon PalmGear / Handango when they got to be more trouble than they were worth.
 

character

状元
Mike, you would probably need to avoid the market due to Google Checkout limitations. There's no excuse for Checkout to work in fewer countries than Apple's payment system.
 

mikelove

皇帝
Staff member
That's what I'd been thinking, yes - not supporting the Market would cut off AT&T customers, but if you're OK with a phone that's a) locked-down and b) on AT&T you're not limited to Android :)

Apple has a big edge international-payment-wise because they already had to spend years building out an international payment infrastructure (which really is pretty complicated; we get wires every month from four or five different banks) for the sake of iTunes music sales; app sales can use the same infrastructure, and the music sales going through the same system help to justify maintaining the infrastructure for app sales.
 

radioman

状元
This is an interesting problem.

Trying to think of any other way this could quantified. Without the store, the business is surely a different animal. Apple has done a very good job of providing pretty much everything in their "closed system" - product creation , development, payment, etc. The closed system frustrates some, but it works and works well.

From the perspective of selling software online, it would seem to me that Pleco would have a good a perspective as any, based its history/current selling model as it relates to Palm and windows. But at least for me, the whole thing is tough to judge. If I'm not mistaken, Apples current mindshare well exceeds that of where the PCs and Macs diverged back in day - a bit of a different landscape.

Can at the model of not using an Apple type store really compete? Perhaps worst of all, if there is no convenient store to purchase from, what will that mean for the add-on business?
 

character

状元
mikelove said:
Apple has a big edge international-payment-wise because they already had to spend years building out an international payment infrastructure [...]
Yes, but Google has shown the ability to build or buy whatever it considers important to its business. Making a viable Market for Android developers to earn money is clearly not important to Google.

You're in a position to route around the Market (though you could put a lite version up for free so you have some presence there) so it makes sense to do so. Android phones and Chinese learners are in many more countries than the few Google Checkout supports.
 

mikelove

皇帝
Staff member
radioman said:
Can at the model of not using an Apple type store really compete? Perhaps worst of all, if there is no convenient store to purchase from, what will that mean for the add-on business?

For us, not too much - we can support PayPal, regular (not 30%-Android-commission-paying) Google Checkout, Amazon, our own account system, and various other methods to make it only slightly more complicated than buying add-ons on iPhone; instead of entering a password, you'd pick one of those systems and enter an email address and password, then you'd only need the password for subsequent purchases. What we lose in speed of purchasing we make up in transparency / ease of order retrieval (no more of the dreaded "sandbox" messages for jailbreakers) and in not having to pay big commissions.

There'd still be stores, I think, there'd just be several of them. There already are a few alternate stores on Android, actually; heck, there's even one for jailbroken iPhones. Carriers would each have their own (a scenario I've speculated on a few times here). Directing people to a company's own website to download an app is only a bit more involved than directing them to an App Store / Android Market page, and the installation process can be similarly painless. What percentage of the iPhone / Android / etc apps that you download did you discover while just browsing around the market? For me at least I find out about far more apps online at various review sites / blogs / etc.

character said:
Yes, but Google has shown the ability to build or buy whatever it considers important to its business. Making a viable Market for Android developers to earn money is clearly not important to Google.

Yes, another thing I've talked about here in the past - Google doesn't really want an offline "app" world, they want everything online; Android Market very clearly favors apps that are a) free and b) small (there are all sorts of conspiracy theories about why they waited until 2.2 to support SD card installation), apps that serve as front-ends for web-based services, web-based services that can then sell Google ads / be listed in Google search indexes / etc. Google doesn't want you using Pleco, they want you using online Chinese dictionaries like MDBG and nciku and online flashcard apps like Anki and Skritter (and the still-largely-theoretical Pleco Online).

character said:
You're in a position to route around the Market (though you could put a lite version up for free so you have some presence there) so it makes sense to do so. Android phones and Chinese learners are in many more countries than the few Google Checkout supports.

The problem with lite versions is that technically, according to the Android Market agreement you're not allowed to use a lite app to promote a paid product that's sold outside of the Market. But other apps do do it, and there's safety in numbers - the last thing Google wants after the recent net neutrality flap is to start getting in trouble for banning apps, a few more bad PR moments and the true believers are going to start deserting Android for MeeGo or some other Linux-based OS from a company they feel less negative about.
 
Top