TamsPPC - the PocketPC Blog

A colleague of mine over at PocketPCThoughs.com has freaked out about Microsoft’s recent announcement stating that it may create iPhone applications in the future. Wooh; the Windows Mobile is in Danger. Wooh, my turf is in danger. Wooh Wooh!

Unless Darius Rey, I do think that Windows Mobile and the iPhone devices can and will coexist - after all, Windows and Mac OS have coexistied peacefully for some time. The reason (or rather, the reasons) for this is(are) simple:

The platforms target different audiences
Darius probably never really used an iPhone. Trust me, I did(and I own an iPod touch) - and I can tell you that a stylus interface is SUPERIOR to a finger-based system whenever small controls come to play. Excel sheets. Word-by-word editing. Transport Tycoon(yes - there is a port): these apps are impossible to do pointfully on a device without a stylus.

The iPhone is a very mediacentric device. It is not a workhorse…

Apple
Being completely honest: I do not feel that the folks at Apple’s have the potential to be market leaders. As bold as it may sound, but their revenue model and company culture is based upon NOT being the market leader. OK; they lead the MP3 player market to some extent - but in the end, their products are way to expensive to really hit the masses.

Also, Apple is one company. Windows Mobile is endorsed by hundreds. There is superiority in numbers…

I personally predict that Microsoft will make some money selling iPhone apps - and that the two platforms will coexist peacefully. I know, that doesn’t sound dramatic(and doesnt make good headlines) - but it’s what I think that we’ll see!

In the last years, Handango has grown to become the dominant ESD – in fact, the folks felt strong enough to do an insane price hike recently(which will probably reduce their importance significantly). Their Yardstick reports have always provided an interesting insight into the mobile software market – here’s my take on the 2007-yardstick.

First of all, let’s look at the number of new applications/software houses added. Symbian has dominated the list, the Palm OS grew the slowest(suffering significantly from last year):

The average application price and order size have both decreased slightly – which is interesting, as I predict the recent rate hike to increase prices significantly:

From a top-seller point of view, SPB’s domination of the PocketPC market has not changed – in fact, SPB now has of the top-ten titles coming from its portfolio. Other platforms show a more balanced app/house ratio, although games are notably absent from the top-ten lists this year(except for a few casual-gamer exceptions):

Overall

Palm OS

1. PocketMirror Std. (Outlook synchronization) - $29.95
2. Agendus Professional Edition (PIM enhancement) - $39.95
3. SplashID (information security) - $29.95
4. Ringo Pro (ringtone player) - $29.95
5. Agendus Premier Edition (Outlook synchronization) - $59.95
6. Snap (data entry tool) - $9.95
7. Agendus Desktop Edition (Outlook synchronization) - $39.95
8. Diet & Exercise Assistant (_tness assistant) - $24.95
9. PocketMirror Std. Upgrade (Outlook synchronization) - $19.95
10. SplashWallet Suite (personal ID manager) - $59.95

PPC

1. Spb Pocket Plus (today screen plug in) - $29.95
2. Spb Mobile Shell (today screen plug in) - $29.95
3. Pocket Informant 8 (today screen plug in) - $29.95
4. MobiTV (streaming television) - $9.99
5. SPB Backup - $19.95
6. MIcrosoft Voice Command (voice control) - $39.9Microsoft 39.99
7. Spb Diary (today screen plug in) - $19.95
8. Spb Time (clock) - $14.95
9. Spb Phone Suite (phone features) - 19.95
10. Spb Brain Evolution (brain trainer) - $19.95

Windows Mobile® Standard

1. Ringtone Megaplex (ringtone library) - $19.95
2. SBSH Facade (home screen plug in) - $14.95
3. MobiTV (streaming television) - $9.99
4. HandiTV (streaming television) - $24.95
5. Fizz Weather (weather monitor) - $16.95
6. SmartphoneNotes (note manager) - $17.95
7. Agendus Standard Edition (calendar manager) - $19.95
8. Agenda One (calendar manager) - $29.95
9. PocketStreamer Deluxe (streaming media) - $24.95
10. IM+ (instant messenger consolidator) - $29.95

Symbian

1. SBSH Papyrus (time manager) - $19.95
2. Handy Calendar for S60 (calendar manager) - $39.95
3. LCG Jukebox (music player) - $24.99
4. Pro_Mail (email viewer) - $27.99
5. Quickoffice Premier (document manager) - $49.95
6. Handy Weather for S60 (weather monitor) - $19.95
7. Advanced Call Manager (call manager) - $19.99
8. WorldMate Professional (travel assistant) - $69.95
9. Ultimate Voice Recorder (voice recorder) - $14.95
10. IM+ (instant messenger consolidator) - $29.95

From a sales/device perspective, the trend is clear. The adoption of new devices has slowed down somewhat, especially in the established platforms but also in the Symbian space. Interestingly, the Treo 650 still is the single-strongest sales getter. As for application categories, producitivity apps and gaming stuff seem to take the lead this year(see above).

For me, the trends in this yardstick are clear. The established platforms have begun to see market saturation(for PPC, the Palm OS situation should rather be called catastrophic). Windows Mobile Smartphone is currently at a turning point; and Symbian and BlackBerry are growing insanely fast.

What do you think?

When WinCE4/Dust hit the road back in 2004, its author decorateds thjis proof-of-coincept virus with the following words(before sending it out to a few AntiVirus vendors):

This code arose from the dust of Permutation City
This is proof of concept code. Also, i wanted to make avers happy.
The situation when Pocket PC antiviruses detect only EICAR file had to end …

However, McAfee now reports about a new “virus” called InfoJack. It has been distributed in a bundle of (probably pirated) applications. McAfee claims that this self-updating application currently has the following features:

• installing as an autorun program on the memory card
• installing itself to the phone when an infected memory card is inserted
• protecting itself from deletion, copying itself back to disk
• replaces the browser’s home page
• allows unsigned applications to install without warning

Additonally, it sends some data about the phone back to its “owner”.

While this tiny-teeny virus is harmless, it nevertheless shows that Windows Mobile is perfectly vulnerable to virii. This will not only generate loads of revenue from AV vendors, but also clearly shows that piracy is dangerous.

Please don’t get me wrong - I am not supporting the criminal works of virus writers in any way. However, more virii definitely leads to more wannabe pirates burning their hands and stopping to do pirated apps - something that developers will definitely appreciate!

I recently met a person thinking:

Microsoft’s .NET framework is beautiful. MSIL(it’s assembly language) is even nicer. Not being able to precompile CF applications is even nicer. Thank you, Microsoft, for making my job easy. - Joe Cracker.

And indeed, applications compiled to MSIL can be decompiled, giving people who have access to your exe file access to your source code. We had something like that with Visual Basic 3 - but apparently, some people at Microsoft’s are very warm to the idea of open sourcing applications by force.

How else can you explain that it apparently is impossible to obfuscate an application in the build process? Yes, you can link a cab project to a regular one; and create a nice little cab file for deployment. Sounds easy…but Joe Cracker has a great time decompiling it.

If you feel like protecting your intellectual property, you instead need to break up the process into three pieces according to this MSDN discussion. First, you create the .exe file from Visual Studio. After that, you need to manually open DotFuscator - and after that, you need to get back to Visual Studio in order to generate a CAB file using the obfuscated file(NOT the project’s standard output). Got it? No? I am still struggling with it, too….

In the end, it is very impressive to see that a company like Microsoft(who appears to be very worried about people copying its wares) is completely unmotivated when it comes to protecting its developers. Obfuscating must be built into the compiler IMHO - forcing developers to make multiple steps by hand is an unacceptable procedure.

Probably, someone at Microsoft’s is living by the German mantra of keeping his own door clean - as long as it’s just third parties getting fu**ed, its not my problem. I can only hope that someone at Microsoft’s finally gets off his hide and makes obfuscation easy - millions of developers would instantly love him forever…

P.S. For all those of you who don’t have too much existing .net CF code yet(and don’t wish to give it out to Joe): AFAIK, ArianeSoft’s PPL is free of obfuscation issues…

A thread on PocketNow.com comparing the home screen of various versions of the Windows Mobile UI have made me think about what really disturbs me about Windows Mobile. After quite a bit of thinking,. I think that I have found my culprit…it’s the annoying top and bottom toolbars which IMHO are way to wide(especially in landscape mode):

If the image above wouldn’t be so sad, one could laugh about it for hours. A spreadsheet cell is tinner than the bottom toolbar, and the same is valid for the bottom one.

My personal request to the Windows Mobile team can be expressed easily - but the toolbar size by half. They are still tappable(case frames aid inaccurate users) - and we could get a nice extra bit of information on-screen!

The news struck me like a bombshell - an industry alliance called Open Handset Alliance announces a new mobile OS to be licensed under the Apache license(aka no need to give back derived code). Among the members of the committee: Google, a variety of carriers and semicon boys - and, most importantly - HTC.

Many of my colleagues think that this will affect Nokia(S60 OS) and Palm(Palm OS) the worst - however, I think that Microsoft is the one who really has a lot to loose here. Both Palm and Nokia have hardware and software in their hand and can thus afford to loose all other licensees(Palm has already lost most, as for Samsung….their cool-looking S60 phones have yet to hit the road). However, Microsoft doesn’t have an own hardware division doing smartphones(forget the Zune, boys).

For them, HTC has always been the ‘reference client’; delivering devices with everything that Microsoft deems fit to carriers that feel like distributing the machines. But HTC has gone on an ‘ego trip’ some time ago, starting to push its own brand at the expense of ODM/carrier jobs.

Looking at Microsoft’s current offerings, I consider Windows Mobile for Smartphone nothing short of a catastrophe. The OS is incredibly clumsy, requiring loads and loads of scrolling and tapping. Windows Mobile 6 for touchscreen has also got quite a bit of flak from journalists(strange…I really like WM5 on my hp ipaq rx4240).

IMHO, should HTC drop WM Smartphone, Microsoft will have serious problems with this part of its OS. Why should developers continue to support an as-good-as-dead platform(even if .NET CF makes it pretty easy)? Why should carriers give their clients devices with a hard-to-use OS if they can have an easy-to-use one backed by a brand name which is just as big(Google)? Nothing good for Microsoft here…

HTC, on the other hand, has little to loose by killing the smartphone runt - it finally gets a fully open OS that it can change around as much as it wants, and will save a load of cash on licensing fees. Microsoft cannot afford to tamper with HTC’s licensing deals(touchscreen devices selling well) anyways.

Peter Chou, CEO of HTC Corp., is quoted saying that “…integration of the Android platform in the second half of 2008 enables us to expand our device portfolio into a new category of connected mobile phones that will change the complexion of the mobile industry and re-create user expectations of the mobile phone experience.” Looking at the emphasis(added by me), this translates to: devices get easier to use…something not easy to accomplish with the runt.

As for the classic PocketPC - as it looks now, this platform will also be left alone mostly. I think that HP definitely won’t join the Open Handset Alliance while HTC is a member there - last but not least to avoid pissing off business customers who use HP devices due to their excellent build quality.

What do you think?

Nowadays, all PocketPC’s support ClearType rendering. ClearType is Microsoft’s name for sub-pixel rendering, a technology that makes text appear “smoother” by using a specific property of LCD displays. However, ClearType usually works only in portrait mode - while a hack is available for enabling landscape mode, I think that this usually doesn’t pay out.

In order to understand why I’m skeptical, we need to take a deeper look at how LCD screens actually work. Feel free to follow me along(yes, the figures are hand-drawn; nope, they can’t be given away…they already have a new owner):

First of all, here, we see a LCD screen(2×2 pixel only). As you can see, it is made up of pixels(points), who are then made up of so-called sub-pixels. A sub-pixel is like a small red, green or blue bulb that can be dimmed in steps. Now, a pixel consists of one R, G and B subpixel; together, they can create every color:

Now, a regular LCD would render a line from left bottom to top right as shown below. If the pixels are big enough, a “step” effect will be visible:

And this is where ClearType locks in(explanation simplified). The program enables sub-pixels adjacent to the line(shown in blue) and possibly also dims corresponding sub-pixels at the other side in order to make the steps less prominent:

So, a line that goes mostly perpendicular to the sub-pixel grid(horizontal) and has just a light slant to the left or the right, benefits a lot from ClearType(look at the average font…most lines are vertical). A line that goes parallel to the sub-pixel grid doesn’t benefit at all.

As said before, most fonts are made up of entirely vertical lines. In portrait mode, the sub-pixel grid is horizontal. Thus, most lines do benefit. However, once your PocketPC is in landscape mode, the sub-pixel grid AND most of the lines are arranged. In landscape mode, ClearType can’t do much, as most of the lines can’t really be optimized. Thus, Microsoft chose to disable the feature…probably to save CPU performance…

Now that we have the physical and electronics background, please let me know what you think! Do you like ClearType’s output when in landscape mode?

A pretty big PocketPC developer has just posted an open letter to pirates to his forums:

http://www.pocketinformant.com/Forums/index.php?showtopic=11368&st=0

After having worked in the Palm OS sector for a few years, I have come to the conclusion below - feel free to debate me as much as you see fit:

For me, the solution is not teaching the masses, writing open letters or doing anything similar to change the people’s behavior by free will. Instead, the solution is to force people to go legitimate. Brick and mortar shops don’t have a preacher preaching about how bad theft is…they have a security guard and RFID tags to protect their goods. For software developers, I think that a different approach is in order:

Cooperate on DRM
Many developers have decent DRM systems in their applications. If a group of developers would gang up and work together on creating powerful DRM systems for PocketPC, the pirates would have a harder stand as they need to fight shared expertise. There’s a cryptographer here, and a hacker there…together, developers can for sure create a kickass DRM system that is extremely difficult to hack!

Attack pirates and warez communities
While I still stand behind my words about attacking legitimate users, I meanwhile do think favorably about developers creating “logic bombs” that attack user’s hardware/software. The trick behind this is that users must be shown that the crash/damage they are currently experiencing is NOT caused by your app, but rather a virus.

I know of at least one Palm OS house that is currently developing a method that destroys Palm hardware irrevocably, and think that a similar approach is also possible on PocketPC devices.

Releasing such an application into the PocketPC/Palm warez ecosystem will immediately shake up the masses and reduce piracy to a select few elite pirates; soft of like how many users are afraid of using Series 60 warez due to the many virii around.

Have pirates cooperate with you
At Tamoggemon’s, we have a “house pirate”. He gets access to all Tamoggemon apps before they get released, and attacks them. In exchange for the privilege of being the first to get the app, he gladly forgoes distributing his crack and even shares his findings with me…although I didn’t implement much of his hints yet(yes boy, I KNOW that you want me to release that Vampyr Zyklon DRM NOW, yes boy, but I have a girlfriend…).

Bring a server into the game
Last but not least, a company I consulted a long time ago(4 years), simply made their apps “web clients” that needed a server for an essential bit of computing. The server checked the legitimacy…and cya, pirate.

This approach probably won’t work for everyone…but if your app already does some kind of web related service, why not hit it for good?

What do you think?

When you buy a boxed, non-system-builder version of Microsoft Windows (e.g. Vista), you can usually ask Microsoft for help with the product you just purchased. However, asking Microsoft for help with the Windows Mobile OS on your PocketPC usually ends with them forwarding you to the manufacturer…why is that so?

The reason behind this is that the relationship between a PocketPC and Microsoft is different from the relationship a Windows box purchaser has with the company - if you buy the box, you buy a license from Microsoft. On the other hand, if you buy the PocketPC; you don’t directly buy a license from Microsoft. Instead, you buy a product containing code licensed from Microsoft by the manufacturer.

The manufacturer looks for an OS that can power its new hardware. In the ideal case, he goes for Windows Mobile straight away - or he could just be persuaded by one of the free evaluation cards that swamped austrian embedded designer trade shows a few years ago. Microsoft then helps the manufacturer get his OS running on his box - and that’s it.

The manufacturer, on the other hand, acts as a man in the middle. He receives support from Microsoft for the technology, and in exchange handles end user support for his personal pet mutation of Windows CE(NOT developer support).

So, to cut a long story short. If you have a non-developer question about your PocketPC, go to your manufacturer! He is the only one who can help you…

Thank you John for the idea!

When Dell’s last Axim was axed for good, many commentators in the PocketPC arena wondered themselves why this series was ended - after all, it was very successful, scored rave reviews and also was popular with developers(it was the top device from a sales perspective for Resco in the last TamsPalm interview). No reason to kill the series - but Dell still did. IMHO, they didn’t do so because they wanted though…they did because they had no other choice. Sounds mad? Yep, but it isn’t - please follow along…

Lets begin our journey with a visit to the Axim’s source - HTC. The Dell Axim essentially is a design by HTC(which, according to some industry insiders was rejected by HP); Dell just stamps its brand logo onto it and manages the distribution. So, HTC produces and Dell makes the big margins… . This combination stands behind many PocketPC smartphones(with a carrier instead of Dell) for sale in Austria and has worked well so far…

However, HTC is not dumb. Why should others make the big money if they can do it themselves? This kind of thinking has become prevalent there - an industry insider I met at the CeBit(by ill luck in a press taxi :)) working for a major distributor told me that HTC is on a ’sort of ego-trip’, wanting to promote its own brand. Palm’s Treo 750 manufacturer swap is an example of that new policy.

With Dell loosing HTC, it essentially had nothing. Nada. Zilch. No hardware design, no PCB layouts, no nothing. A lot of the applications on the Axim series were written by HTC(just check the About sections…), and lets not forget all the drivers and other custom stuff. So, Dell would have needed an enormous amount of R&D cash to get their Axims rolling again - cash that they didn’t want to spend for one reason or the other…

Of course, this is all just speculation - but with the raising market share of PocketPC’s and Dells classically good market shares, it looks very plausible. There still is money in high-end PDA’s…not the kind of money other companies made years ago on a funky device called Vx; but still good enough money. Essentially, the only one who knows more is Dell - but they prolly wont tell. And since Austrians say that stuff that rhymes is good…what do you think?

In the Palm OS world, the topics “decline of the handheld” and “move to smartphone” are permanent topics as there is only one real manufacturer left(sorry GSPDA, but you don’t count). PocketPC users are better off with a variety of manufacturers alive and kicking…and IDC’s latest study allows us to look deeply into the manufacturer’s pockets.

Looking at these figures, we see that all manufacturers shipped less “handhelds” than in the year below. Palm still is the number one manufacturer, but HP, Mio and Dell alone have more market share. HP still is the number one of PocketPC’s, but Mio is closing in significantly, having almost tripled its market share.

Interestingly, the top 5 contains two walking dead - both Sharp and Dell have announced to exit the handheld market and are currently clearing out their inventory.

Last but not least, handheld shipments have dropped by 1/3 overall. IMHO, this is a matter of customer migration mostly…manufacturers like HP and Palm have recently begun to aggressively market their smartphones at customers instead of PocketPC’s…

What do you think?

Source: IDC

When the first VGA PocketPC’s hit the market 2004, they dethroned Palm’s HiRes+(320×480) screen systems from the crown of highest resolution in a PDA. I was impressed by the quality of the displays and the amount of data on-screen when I saw such a handheld a few months ago(HTC Universal)…but nevertheless, our local electronics store didn’t have any VGA PocketPC. Brighthand users currently discuss the lack of VGA PocketPC’s - and indeed, the lack of VGA displays is appalling.

Yes, VGA is cool - but there is one critical flaw with increasing display resolution:

Human eyes are limited when it comes to seeing small objects
Look at the image below:

It shows a more-less 1:2(your screen shows the text about two times as big as the Treo) image of a HiRes(320×320) screen showing text in the smallest font possible. While the text may be somewhat readable on your screen, trust me that a dirty little TFT on a PDA will affect readability adversely:

Now, that VGA PocketPC has text that’s 50% smaller than the already small text above - that goes beyond the vision limit of the average user. Both Microsoft and PalmSource kept their big default fonts when they upgraded their screen resolutions…and did so for a good reason. The average user will not benefit much from VGA, so why force him to pay extra?

To cut a long story short, VGA PocketPC’s rock(I would love one)…for us. But the average user doesn’t benefit much from VGA - and since companies tend to target average users rather than freaks, VGA will probably stay a rare feature for the time to come…