Symbian devices?

[batoje]

N95

The upcoming n95 from Nokia is seems like the perfect all in one device. It is so perfect that I think I will get one and carry around a simple "Pleco-compatible" device. I really wish there was a pleco version for symbian even without a handwriting recognizer. I can't even find a chinese dictionary for this device.

 

[mikelove]

The latest idea on this is to first release a version for Windows Mobile Smartphone (probably along with Pleco 2.1) and see if sales for that are sufficiently high to justify the more serious rewrite required for a similarly touchscreen-less Java version. Porting from Pocket PC to WM Smartphone takes almost no effort, all we'd have to do is redesign the interface, so it's a good way to see how our product works on the smaller form factor without risking too much time or money on a more involved port.

The Pocket PC version has actually been something of a disappointment sales-wise, to the point where in hindsight we would have been better off waiting and releasing the first Pocket PC port along with Pleco 2.0, so we're naturally very hesitant to spend any more money on porting to a new platform unless we're really really sure that people will actually buy the software. (a few Symbian fans do not a successful product make, sadly)

 

[teriyaki]

All quite true.

However, I would probably keep expectations for Windows Mobile Smartphone modest; I only know of one person who has one, and he's a native Chinese speaker. It just seems like it has no penetration versus Symbian. So even if it's easy, it would probably not be a hot seller. I know several people with Pocket PCs, so if the Pocket PC version isn't selling well, the Smartphone oughtta sell even worse (in my anecdotal opinion).

I think a JAVA version would give you the highest ROI because you can reach out to a ton of wireless devices (maybe even including Windows Mobile Smartphone if they have a JVM). Give a link to some free JVM for each respective platform and then everyone's ready to go, right?

(For what it's worth, I bought the PPC version.)

 

[hairyleprechaun]

Hello Mike,

I just want to put in my two cents worth. I agree with you that if you really decided to port PlecoDict to another OS that it would be much easier and cost efficient, starting out anyway, for you to port it to something like Windows Mobile for Smartphones because of the similarity in that OS with Windows Mobile for PPC. However, aren't there many more Symbian devices out there? And now many Windows Mobile Smartphones and Symbian devices are coming equipped with styluses right? So why even bother with creating/porting a handwriting recognizer-less version of PlecoDict when you can give them a version that has the handwriting built in, and which you already have developed?

On another (and I think important) note. I think there is one more good reason, that maybe you have overlooked, as to why you have seen such poor sales of the PPC version of PlecoDict. The reason comes from your own discussion of which device a person should consider buying to run PlecoDict if they are buying a device solely for use with it. Following is that discussion. Please look closely at the sections I highlighted in red.

Palm versus Pocket PC
The first question for most people is whether to buy a handheld running the Palm OS or Pocket PC (a.k.a. Windows Mobile) operating system. If your only reason for buying a handheld is to run PlecoDict and you don't plan to use it for very much else (except possibly as a personal organizer), we recommend buying a Palm OS system, as we feel Palm OS tends to be faster and more reliable than Pocket PC. Likewise, if you're sensitive about price we strongly recommend Palm OS, as the cheapest Palm OS handhelds are significantly more affordable than the cheapest Pocket PC ones. On the other hand, if you're a "power user" and want to use your handheld to browse the web, play games, compose/edit complex documents, and watch movies, you'll probably be happiest with a Pocket PC. However, the installation process for PlecoDict on Pocket PC is significantly more complicated than on Palm OS, so if you're not very good with computers you will almost certainly be better off getting a Palm. So basically, we recommend Palm OS unless you have a strong reason to favor Pocket PC.

After reading your "discussion" it seems to me that the best choice is to buy a Palm model. No wonder your PPC version of PlecoDict is seeing poor sales. Also let me give you an example between my experience with installing and using PlecoDict on a PPC and my friend's (he is helping me write this right now) instalation and use on a Palm.

*Note, I originally recommended the Palm version to him, based on your discussion as well as many other sites that said Palms are much more intuitive, easier to use, and much better at synching with computers.

I myself bought the PPC version and installed it with no problems, using Microsoft Activesync. Though initially I bought my PDA, HP iPAQ hx2495, just for use with PlecoDict, since then I have found a host of applications that I use with it. In addition I found that using Windows Mobile 5 to be very intuitive, like using my computer, with its drop down start menu and all of the programs located in the programs folder, and all of the settings being located in the settings folder.

My friend on the other hand, went ahead and bought the Palm OS version of PlecoDict, for a Palm TX that was also purchased for the sole purpose of using PlecoDict, based on your discussion above, as well as on my recommendations. He is not very computer savvy and he liked the idea of a simple, uncomplicated install, and being able to easily use the program without having to have any computer knowledge. Unfortunately, when it came to installing PlecoDict on his Palm my friend had several problems because he could not get Palm Desktop and the Hotsync application to work properly. In the end he was finally able to sync the PlecoDict software but not without many hassles, such as downloading updates for the Palm software and drivers. On top of the syncing problems, which he still has, he found using the Palm interface so unintuitive to use, especially finding the locations for changing different settings, that he recently switched to the PPC version after his Palm got run over by a car.

All in all my friend had a lot of problems with the installation process of PlecoDict on his Palm, and he finds the Palm interface, especially when it comes to finding and changing various device settings, not very user-friendly as many of the settings are located in different places under different headings/windows/tabs. On the other hand the PPC experice seems to me, and my friend, to be much more intuitive (very much like using a Windows PC). Additionally, there were no problems with syncing and installing the PlecoDict to it, espcially since the process is pretty much automated except for a few dialog boxes that pop up on the device asking you whether you want to intall portions of the software in the main memory or on the memory card. Of course simply selecting a radial and clicking on "OK" gets the user past the pop up dialog boxes.

 

[mikelove]

teriyaki - yes, we wouldn't be expecting WM Smartphone to be a huge sales boon for us, the main idea would be to get user feedback, see if there was any real interest at all in a touchscreen-less version of our software before devoting a lot more resources to a BlackBerry port. So "sufficiently high" in that context would probably just mean 100 orders a month or so, just enough to suggest that a decent number of people were buying, using, and liking the software.

hairyleprechaun - I'm well aware of our (my) comments steering people towards Palm OS over Pocket PC, but that really doesn't affect whether the Pocket PC port was successful or not. As far as justifying the cost of a port goes, if a customer has no particular attachment to Palm OS or Pocket PC and is happy to buy whichever type we recommend, then that's a customer we would retain regardless of whether or not we offered a Pocket PC version. So, to be completely cold-hearted about this, they don't really factor in to whether or not the port was a good idea financially; we're getting their money either way.

The main reason to add a new platform is to support people who either can't or won't use a handheld/smartphone running one of your existing ones. And I'm simply not convinced that there are a lot of people out there who aren't buying our software because we don't offer a Symbian version. I am willing to believe this for BlackBerry, because such an enormous percentage of the people using BlackBerries were issued them by their employers and hence really don't have the option of using another platform, which is why releasing a WM Smartphone version to "test the waters" for a BlackBerry one makes sense.

I'm glad that the PPC version installed so easily for you (and sorry that the Palm OS installation was so difficult for your friend), but that's really far from typical; going by our support requests at least, a much higher percentage of PPC users than Palm users have trouble installing our software, in large part because of ActiveSync's poor handling of large data files. (particularly with WM5) We've actually designed a brand new customized installer system for PPC to help alleviate this problem (essentially we bypass ActiveSync's built-in file installer completely and copy the files over manually) but we're not sure if it will be release-quality when 2.0 comes out, so it may not make it into our software until 2.1; still, when it does we'll likely tone down the Palm versus PPC comments somewhat. But until then I continue to believe that the installation experience on Palm OS is considerably better than on PPC.

 

[david.dencker@gmail.com]

touch screen symbian devices becoming abundant

I have argued for Plecodict on symbian devices before – I still think you should be proactive and start developing plecodict for touch screen telephones (or somebody else will do it!). Strategy Analytics, a consultancy, predicts that 40% of new mobile phones could have touch screens by 2012. The vast majority of mobile phone OS is still symbian - not window mobile... Mobile phones is a far larger and very different market than PDAs so you can hardly use the PPC version's low ROI as an argument for not porting plecodict to symbian - a market analysis would be in place. PS: I think the handwriting recognition is too central to be left out. David

 

[mikelove]

Symbian's statistical dominance comes mainly from the touchscreen-less Series 60 devices, nearly all of which are made by Nokia, and most of which make little or no effort at doing anything really "smartphone"-ish (like, say, offering a regular keyboard), so I'm not convinced that the Symbian software market now is really that big. Before we released a Pocket PC version we were getting maybe an e-mail a week asking for one; the last e-mail we received asking about Symbian was back in November, and one every couple of months is pretty much the norm. We've already gotten far more total e-mails about the iPhone than we've gotten about Symbian, in spite of the iPhone only having been announced 2 months ago.

I appreciate that there are a lot of people who are very fond of Symbian, and I'm not ruling out the idea of doing a dedicated Symbian UIQ version eventually, but for right now I think WM Smartphone, BlackBerry, and iPhone (should Apple eventually release a development kit) are the three most interesting handheld platforms for our purposes. Plus regular Windows, which is increasingly appealing as a handheld platform; UMPC shipments are projected to hit 7.8M units per year by 2011 and we'd certainly love to get Pleco running on those.

A little bit of a silver lining: StyleTap are apparently working on a Symbian version of their Palm OS emulation system, so at whatever point that's released, that would give you a way to run PlecoDict on a Symbian phone without any help from us.

 

[batoje]

There are options besides Palm and PPC!

I previously thought that there were no other devices to accomplish what I wanted in just one device. My first concern is to have a device that runs Pleco dict. For this reason I did not venture to look outside Palm or Pocket PC.
Palm finally lost my business when they refused to keep their technology up with the times. Pocket PC seemed good for awhile and I was glad to have other more technologically advanced options. I actually prefer the PPC version of Pleco dict over the Palm one. The biggest problem with PPC is that they do not have any good/stable option for running chinese (with an IME) and english at the same time. I was very disappointed after I purchased CE star. It was very buggy and they offered no tech support. I found myself wishing for the Palm equivilent of CJKOS on PPC.
THE ROKR E6 HAS IT ALL! I have had it for a few weeks now and could not be more satisfied. In a perfect world Pleco dict would be installed but the included Chinese-English English-Chinese dictionary is not that bad at all. Actually the main function for me in Pleco dict was handwritting translation. I need more then anything to write a character and have a the definition pop up. I can do all this on the Rokr and much more.
The Rokr perfectly integrates Chinese and English. I can switch the imput methods with one tap and zero lag time. It feels and functions like a real phone too! Battery life is awesome! Snappy performance, excellent commuication options, Chinese GPS beijing map included, business card scanner, super good build quality, and ALL the STANDARD connections (SD card, mini USB, 3.5mm headphone jack) bundled into one svete phone with no need to download anything!
The ROKR has really made me realize what we should expect from a phone in 2007. Palm and PPCs seem like a rip off now. I wish there was at least a JAVA version of Pleco so that I could really have it all! I miss so many of the features in Pleco. Especially the Flashcards. However until there is some manufacture out there that makes a device as small and capible as the Motorola ROKR E6 or Pleco releases a version that will work on this device, I will be quite content with this as a long-term interim! :wink:

 

[mikelove]

batoje - interesting to hear that the ROKR's handwriting recognizer is that good. Does it work with cursive handwriting? How well does it handle mistakes in stroke order? Does it handle simplified and traditional characters or only simplified?

The problem with PPC Chinese input methods is really that they try to do too much - if they just paired a Chinese input system with a font link in the registry, that would do about 80% of what their current system does while not having any noticeable impact on system stability. But by insisting on hacking the system to make Chinese work absolutely everywhere they end up making things painfully unstable. If we ever end up turning PlecoDict's text input system into a standard Pocket PC input module (not something we're planning for 2.0, but possible in a later version) we could actually consider doing this ourselves, though I'm not sure if it's really a market we'd want to get into.

But on the ROKR, I should warn anyone here keyed up about a possible Java version that it really wouldn't be nearly as nice as the native Palm/PPC ones; there are severe limits to what you can do in mobile Java, and lots of what we'd be doing would come with severe performance costs, so Pleco for Java would have maybe half of the features of the non-Java version and be much much slower. (unfortunately, at least according to Motorola there isn't any way to develop for the ROKR other than with Java, so regardless of how successful its US launch may be we'd still be stuck)

 

[mikelove]

Small update on this. In yet another one of my habitual changes of heart, I've been reading up a little on the Symbian development situation and actually it's not as difficult as I had thought; they've come a long way in the last couple of years, to the point where a Symbian port probably wouldn't be any more complicated than porting to Pocket PC was.

I'm still waiting for a compelling market reason to do it, but if Nokia eventually manages to break into the US smartphone market (say by buying Palm, as it's widely rumored they might do) then a Symbian port might not be that far off after all. It would probably only be S60 at first, though; the lack of a touchscreen would save us mucho dineiro on handwriting recognizer licensing fees, and S60's installed base is an order of magnitude bigger than UIQ's, plus however well-designed their phones may be it seems doubtful to me that Sony Ericsson has the market muscle to launch a really successful smartphone here.

 

[teriyaki]

Well, I've got a Nokia E62 (Series 60) here, and would love to Pleco on it. And the E90 looks really sweet; I'll definitely get it when it comes out. For me, I've been a long Treo fan and put up with all the crap (constant reboots and data corruption) since the Treo 300 (then 600, then 650, and now 750), and after trying my best to stick with Palm and Windows Mobile for many years, I've got to say that I'm just tired of all the weird problems and instability. Like, really tired. I'll take anything that is stable, has a keyboard, has Pleco, can surf the web and do SMS/MMS. Would be really happy if Pleco, even a stripped down version, was available for Symbian. The E62 has a built-in dictionary, but it's pretty limited and doesn't have pinyin (i.e. it's designed for Chinese people who already know hanzi.) Now that Palm's up for sale and Nokia might be a buyer, I'm hoping that the situation you've outlined happens!

 

[mikelove]

I've heard Symbian has occasional stability problems too, though probably not on the scale of Palm's and Microsoft's; it makes sense that they'd be more stable, really, they're not running on a 10-year-old core like Palm and they're not crippled by corporate politics like Windows Mobile.

Anyway, as I said Nokia needs to make a big splash in the US smartphone market before we seriously consider this, but assuming that eventually happens Symbian would probably be third on the priority list behind WinXP/Vista and touchscreen-less Windows Mobile.

 

[batoje]

rokr

Mike,

I must admit I am quite excited even to hear of the possiblity of a product for a different platform. I really would love to have option other then palm or PPC.
As for the handwritting recognizer yes to all of the above. Being primarily a Chinese targeted phone they really have a slick interface for Chinese imput. The handwriting is very similar in feel to Pleco and the Pinyin imput is very similar to CJKOS. You can smoothy and quickly switch between imputs. The Chinese friends that have used it seem to love it.
About the traditional characters...I don't know many tradtional characters so I am not a good tester on that one. However the phone does have a choice of traditonal or simplifed characters in the phone setup options. I am sure that if Traditional is assigned then the handwritting recognizer would be geared towards those characters. As for stroke order it is really forgiving and even if you don't get it right the first time the 6 most probable choices are available on screen (like Pleco).
I am all for even a stripped Java version. The one thing about the onboard dictionary in the Rokr is that it does not allow pinyin look up. There is obviously no real comparison or replacement to Pleco dict but at least I have something to get me by until there is more choices available either for phones or Pleco platforms.

 

[mikelove]

Interesting that it handles both - if you write a traditional character in the simplified-version recognizer, does it give you the simplified equivalent? (very specific question but an important one for our purposes)

Still, after my recent exploration of the Symbian development universe I'm starting to think that it would be a lot easier to expand to that than to Java; Java would require a near-total rewrite of PlecoDict, whereas it looks like with Symbian we could actually leave most of the engine code (the non-platform-specific stuff that does all the heavy lifting database searches / flashcard management / etc) unchanged. The best chance for a Java version would be if we made it an online-only app, with the heavy lifting done by a server somewhere and a slim Java front-end.

 

[sfrrr]

If you produced a symbian version, I would seriously consider returning to the Nokia fold. Oh, and if Nokias synced with Outlook.

Also, Mike, in case you might miss me, which I seriously doubt since I don't even miss me, I'll be lurking for the short term. I won't have the time to post much. I'll be back as a loud mouth (loud finger??) as soon as my family life settles a bit.

Sandra

 

[hairyleprechaun]

If you produced a symbian version, I would seriously consider returning to the Nokia fold. Oh, and if Nokias synced with Outlook.

Hey Sfrrr,

I bought a Nokia 5300 last October, and though it isn't even considered a smartphone, it syncs with Outlook on my computer and on my PDA.

 

[batoje]

Interesting that it handles both - if you write a traditional character in the simplified-version recognizer, does it give you the simplified equivalent? (very specific question but an important one for our purposes)

Still, after my recent exploration of the Symbian development universe I'm starting to think that it would be a lot easier to expand to that than to Java; Java would require a near-total rewrite of PlecoDict, whereas it looks like with Symbian we could actually leave most of the engine code (the non-platform-specific stuff that does all the heavy lifting database searches / flashcard management / etc) unchanged. The best chance for a Java version would be if we made it an online-only app, with the heavy lifting done by a server somewhere and a slim Java front-end.

Mike,
Looks like when I imput the traditional character for "men" (door) it will come up as "men" in the simplified. When I switch the phone options to traditional it will just come up traditional. I don't think it will display traditional and simplified together.

 

[mikelove]

Unfortunate, but we were probably going to want to use our own recognizer anyway so it's not really a deal-breaker I suppose.

 

[John]

Just another voice amongst the few (?) asking for a Symbian port!

I've been a loyal Pleco customer (Oxford, Berlitz and Pleco) but both my Palm T3 and mobile phone are getting a bit long in the tooth. So I've just bought a Nokia N95 which seems an excellent phone to take travelling, e.g. wifi, quadband, 3G etc ... but no PlecoDict so I'll be carrying around a Palm for the foreseeable future it seems.

 

[mikelove]

The N95 is an interesting device, but it doesn't really seem like a particularly good PlecoDict platform since it has no touchscreen and only a numeric keypad - without a keyboard, Pinyin input seems like it would be tedious at best. The E series are a lot more interesting for our purposes as Symbian devices go, though once we had a S60 version it would presumably work on the N95 as well. Anyway, thanks for your thoughts on this.