Should Developers Drop Support for Pocket PC 2002?

[Jason Dunn]

I received an interesting email from Mel Sampat, the man behind <a href="http://www.twopeaks.com/">Two Peaks</a>, and I think this question is on the minds of many developers right now. I'll let Mel explain:<br /><br /><i>"Like many other developers I met at MEDC, I am trying to figure out whether to drop support for Pocket PC 2002 devices now that Windows Mobile 5 is out. The new Visual Studio 2005 only allows us to target WM2003, WM2003 SE and Magneto devices. I'm wondering whether to move forward with using the latest development tools and API's, or continue to maintain broader support including 2002 devices. Can you run a poll or two asking users what percentage of them are still using a Pocket PC 2002 device, whether they would consider a Magneto or 2003SE upgrade, and also what the community in general thinks of developers dropping support for 2002 devices in their apps?"</i><br /><br />In an ideal world developers would be able to support every Pocket PC out there, but given that Microsoft has changed the tools to only support 2003 and above, that's not an ideal world. The more time developers have to spend making sure their applications work on older devices, the less time they have for new features, and in some cases, the more expensive their applications will be. Weigh in by voting in this survey and commenting.

[Rod3]

I'm the one vote for "other." I have a huge collection of PPCs. I do have a Jornada 568 with 2002 on it. It can't be upgraded, thank you very much, HP. :cry: All the others (don't ask; I won't tell :oops: ) have 2003 or 2003SE on them. I still enjoy the Jornada. I'd never sell it; but if it's not supported, I wouldn't cry. It just keeps chugging along as is.

[genetiq]

Keep in mind that the vote only counts Pocket PC users who visit this site. I'm sure there's much more PPC2002 owners. Those guys don't read pocketpcthoughts.com but still buying a lot of software :)

[ddwire]

I also posted other. I currently use a Sprint PPC6601 phone edition with 2003SE, but also still have a Philips Nino 510 color with Palm pc 2.1? With the exception of having good pdf file readability the Nino is very usable. The email client works just as good as my 2003 SE and I have a third pary web browser that works as well as the PIE in 2003SE. The batteries are replacable with AA if you run the rechargeable dead.

When is Microsoft going to stop bloating these devices.
Dan

[ppcinfo]

Keep in mind that the vote only counts Pocket PC users who visit this site. I'm sure there's much more PPC2002 owners. Those guys don't read pocketpcthoughts.com but still buying a lot of software :)

I agree completely. The people that visit this site probably do so to get the latest info on PPC devices and upgrade somewhat frequently. There are still a lot of 2002 OS devices out there, so I certainly would not exclude them.

ppcinfo

[Kathy_Harris]

I support a bunch of ipaq 36xx for family and friends and am still surprised that software works for these old fellas. I mean, geez, they're 5 years old! All have new batteries and some have new stylus locks and yet they keep going.

Personally I have a Toshiba 805 and will not upgrade until there are true VGA devices that work with software as I have it right now. I would consider a Dell 50v if something happened to the toshiba. But the Mobile 5 devices? I'd have to see....

[stevelam]

Personally I have a Toshiba 805 and will not upgrade until there are true VGA devices that work with software as I have it right now. I would consider a Dell 50v if something happened to the toshiba. But the Mobile 5 devices? I'd have to see....

The Dell will be a 5.0 device (when the upgrade is released)

[Jason Lee]

When is Microsoft going to stop bloating these devices.
Dan

Hopefully never. That is what a palm is for.

[Kathy_Harris]

The boyfriend has a 50v and I think it is great. It was what I pushed him to get (actually, I think I bought it for him.) But for some reason I really like the Toshiba. Ugh, I wish I wasn't so loyal.

As a mac user &amp; music lover, the boyfriend appreciates the Media Player 10 interface on the 50v. It's a good thing that 2005 will incorporate Media Player 10.

[farnold]

The problem is somewhere else altogether. As long as it's up to the manufacturer to decide which OS (upgrade) he makes available for a certain device you gotta accept that older versions remain in the market for quite some time. While users in boards like these may be getting new devices almost every year or so, the normal user may look at a new device every 4 or 5 years. So when was the last device with PPC2000 released?

[Paragon]

Like a lot of others I personally use WM2003 SE device(s), but my wife and oldest son both use PPC 2002 devices. I think there are a lot of people out there still using PPC 2002. This is something important to consider....you seldom here from those people. Folks who tend to be involved at this level I think tend to have more current devices, so I think it will be difficult to get the real answer. The 2002 folks are not heard from.

I personally think it is far to early to drop support for PPC 2002. I do have to admit, I don't have a good feel for what the additional cost would be to continue that support as a developer. As for Microsoft changing the tools to only support 2003 and beyond, I don't think I'm going to step in that puddle of mud.


I say supprt PPC 2002, and offer upgrades to WM 5.0 under the developers present upgrade terms.

Dave
Dave

[Sven Johannsen]

I look at a bit differently. There is a ton of software out there for all devices, including the older ones. What we are talking about now is new software. I would expect that developing new software and making it backward compatible would limit you to only doing things the old stuff can support. You wouldn't be able to take advantage of new features the new OS allows. That wouldn't make it very competative with stuff that does offer the new features.

A bigger question would be suport for older software. How long should ISVs support older software with at least tech support and/or bug fixes. New revisions with new features would fall in the above paragraph I would think.

I have to wonder if folks with older models are really a big market for state of the art software. Not the folks here, but the average Joe with a 568, 3600 or an EM500. Do they have anything non-OEM on there at all?

[wkspear]

In theory I am with the developers, in that it "makes sense" to focus on the dominant operating systems (WM 2003/WM 2003 SE). There are unfortunately many conflicts of interest in the Windows Mobile world within which they will be caught, such as the OEM control of upgrades for the operating systems. If I could buy the upgrades from the OEM, I would, but they will not let me. It's unlikely however I will buy a WM5 device because, to be candid, the hardware developments have been very disappointing with few exceptions, and I don't anticipate that will change. My iPAQ 22xx is still more appealing to me than most of the newer mid-level (where my budget resides) devices. In fact, I consider it perfect for my needs, and if I were to upgrade I would be looking basically for a 22xx with the newest OS. In other words, all I want is the software side, please kind Sirs.

My point is, the OEM blackmail strategy of using software updates to sell hardware is going to get very hard to sustain, I think, and developers may and will suffer when people who've had Enough of it simply choose to live behind the latest incremental WM tweak (also known misleadingly in most instances as "a new operating system" - will the next OS "upgrade" consist of new Pocket Office icons, a renaming of Connection Manager, and the ability to edit PowerPoint? More or less, I expect).

Probably the way out of this will come when Windows Mobile fragments into a multi-platform operating system, giving OEM's a broader range of income streams. We'll be able to buy the OS upgrade (but not from and at the mercy of the OEM), and the OEM's can sell a broad range of WM appliances (with the fancy new hard drives and memory chips), freeing them from the pressure of coming up with and marketing through OS blackmail a new line of PDAs every few months because they have too.

This would be much better for all of us than the current sorry state of affairs.

[Paragon]

Do they have anything non-OEM on there at all?

I think they do Sven...but your point is well taken. I would wonder how much new software they are continuing to add onto them though? I'm glad I don't have to make this decision. ;)

Dave

[Jon Westfall]

Keep in mind that the vote only counts Pocket PC users who visit this site. I'm sure there's much more PPC2002 owners. Those guys don't read pocketpcthoughts.com but still buying a lot of software :)

I'd debate that a large portion of our readers are not the most current-up-to-date gotta-have-it-will-sell-firstborn-to-buy-it kind. The most active contributors and the editors are (Thank God I don't have kids yet or it might be tempting), but I'd estimate that the bulk of our 22,000+ members are frequent readers &amp; lurkers, and infrequent contributors. They can vote, but infrequently do. Right now we have around 2% of users that have voted, so the results are probably skewed as well.

Back on topic, I think developers should support legacy devices as long as is feasible for them. If their user community and consumer base wants new features over legacy support, then the shift should be made, on a developer by developer basis.

[ctitanic]

Keep in mind that the vote only counts Pocket PC users who visit this site. I'm sure there's much more PPC2002 owners. Those guys don't read pocketpcthoughts.com but still buying a lot of software :)

I agree completely. The people that visit this site probably do so to get the latest info on PPC devices and upgrade somewhat frequently. There are still a lot of 2002 OS devices out there, so I certainly would not exclude them.

ppcinfo

100% agree.

[PocketPC Addict]

I've spent 2 days programming workarounds for a 3 users who have 2000/2002 devices and who are having trouble. I have to say, it is a bit of a challenge to go back, pull out the old ipaq 3635 (2000) and the old 3975 (2002) and remember everything needed to make them work again. Where is that blasted sync cradle for the 3635?!?! The old devices are slower, more prone to error with new code and consequently more difficult to program for if you keep up to date with the tools.

With the VS 2005 coming, there will probably be somewhat of a "forced conversion" of code to the new version. What that would mean is to continue legacy support, one would have to maintain two separate development environments. - thus have to do more than twice the work to produce the same 1 program. :| The time used for that would be time lost for building enhancements and building new programs.

That's not to say I won't do it, just pointing out what we are facing in the coming months.

[lapchinj]

Keep in mind that the vote only counts Pocket PC users who visit this site. I'm sure there's much more PPC2002 owners. Those guys don't read pocketpcthoughts.com but still buying a lot of software :)Very, very true

Jeff-

[lapchinj]

I've spent 2 days programming workarounds for a 3 users ...But I've found that when apps do not have to be fixed then a customers with an older PDAs do not always want to upgrade hardware that works fine for them.

Jeff-

[surur]

I have to wonder if folks with older models are really a big market for state of the art software. Not the folks here, but the average Joe with a 568, 3600 or an EM500. Do they have anything non-OEM on there at all?

I agree with this view. The people who are not actively upgrading the hardware are probably not buying new software either. Also I bet a lot of old ppc's are lying in drawers somewhere and not in active use. How many 5 year old PC's are still in use. I know a lot, but >30% of the installed base? I dont think so. PPC's turn over even faster.

Surur

[Pony99CA]

My main Pocket PC is an iPAQ 5550 running WM 2003. My daughter has a Viewsonic V36 running WM 2003, and my wife has my old iPAQ 3870 running Pocket PC 2002, but neither of them use theirs much. I also have my old iPAQ 3650, but don't use it much, and have a Mobile Crossing WayPoint 200 running WM 2003 that I use in my job.

I think that developers should try to support Pocket PC 2002 (and Pocket PC 2000 :!:) if possible. Ideally, they keep track of what operating systems registered users have so they can kill support for older versions when too few people use them.

Steve

[hoodmeister]

Keep in mind that the vote only counts Pocket PC users who visit this site. I'm sure there's much more PPC2002 owners. Those guys don't read pocketpcthoughts.com but still buying a lot of software :)

As already stated, what use is a poll which only takes the enthusiast community for a demographic?

2003SE, for those who care.

[Pony99CA]

The problem is somewhere else altogether. As long as it's up to the manufacturer to decide which OS (upgrade) he makes available for a certain device you gotta accept that older versions remain in the market for quite some time.
Actually, I'd argue that that isn't really where the problem is. The problem is implicit in this discussion -- Microsoft keeps dropping support for older versions of the operating system with each new release of the developer tools.

If Microsoft didn't do that, third-party developers would have a much easier time supporting older devices. Yes, they wouldn't be able to support some newer features perhaps, but I don't think all classes of software need those features.

Steve

[Pony99CA]

I look at a bit differently. There is a ton of software out there for all devices, including the older ones. What we are talking about now is new software. I would expect that developing new software and making it backward compatible would limit you to only doing things the old stuff can support. You wouldn't be able to take advantage of new features the new OS allows. That wouldn't make it very competative with stuff that does offer the new features.
Not true. Software compiled for newer versions could take advantage of newer features, while software targeted for older versions might be missing some features, but would still offer value to customers. Look at Resco, for example -- they still support Handheld PCs. :!:

Also, as my previous post mentioned, I disagree with your implication that new software has to use such new features. I bet there are plenty of classes of software that people would buy that don't require features only available in Windows Mobile 2003 or later.

Steve

[Pony99CA]

As already stated, what use is a poll which only takes the enthusiast community for a demographic?
Well, first, a developer specifically requested it.

Second, doesn't it make sense that enthusiasts would be some of the more frequent software buyers?

Steve

[lapchinj]

I have to wonder if folks with older models are really a big market for state of the art software. Not the folks here, but the average Joe with a 568, 3600 or an EM500. Do they have anything non-OEM on there at all?

I agree with this view. The people who are not actively upgrading the hardware are probably not buying new software either...
As a developer I'm sort of stuck in the middle of this. While I don't do much of any development on PDA but I do quit a bit on servers and Desktops. Where I keep my own hardware current and would love would love to keep customer's systems also currently I find that 75% of my customers are not interested in hardware upgrades that do nothing for their bottom line. But when a customer is looking into new hardware they will buy the latest and greatest.

While I just bought an Axim x50v because 'I had to...' 8) when I mention the benefits of this new hardware and software I get an answer like "I don't need an Axim to use AF , Billrate or Resco Audio Recorder". End of discussion. But at the same time they will upgrade the software they have loaded religiously.

As a developer I know very well the issues with supporting older platforms. But that's the feeling of the customer base especially when they are not computer oriented businesses. When I dropped the COM world for .NET I was very surprised that I had a lot of resistance to a platform that is more or less transparent to the customer. (didn't cost them anything just made my life easier.)

New server hardware that will cut a customer's processing time from 6 hours to 45 minutes at the end of a month will be bought in a heartbeat. But even the latest versions of AF, Billrate, Phatware and Resco's Audio Recorder (for audio notes) will run very nicely on the older hardware.

But it seems from this thread that the home market is more willing to spend the bucks.

Jeff-

[Sven Johannsen]

Also, as my previous post mentioned, I disagree with your implication that new software has to use such new features. I bet there are plenty of classes of software that people would buy that don't require features only available in Windows Mobile 2003 or later.

Steve

I guess my point is, if the software runs on older systems now, won't it continue to do so? If you have the development suite that created it and you need to do bug fixes...hang on to it? If you are upgrading your program just for fun, is it that unlikely that it won't run under WM2005. There is still PalmSized PC stuff (CE2) that runs under WM2003SE. Yea the menus show up at the top.

If you are adding functionality specifically because of WM2005, wouldn't that be targeted at WM2005 plartforms? If you are gearing up a killer app for Longhorn, why do you care if it will run on Windows ME? Probably worth your while to make sure it runs reasonably on XP. Probably one that already does that for XP though.

[SteveHoward999]

Ultimately you are going to get better backing for dropping support from people who do any kind of software development. we at least understand the issues from a daily-toil perspective.

'Ordinary' users who never do any development have a different perspective - they don't want to be left on the shelf.

Then there are the enthusiasts. They probably appluad support for older devices in principle, but since they most likely have old and new devices to play with, they don't really care if support for older OS versions is dropped.

Personally I think if you have been developing an application over a number of generations and find that it is becomming attractive to drop support for older OS versions, then go ahead and do so - but continue to make it possible for customers using older devices to purchase the last version developed to be compatible with their OS.

[Perry Reed]

This is a topic near and dear to my heart. I am one of the rare breed of enthusiastic Pocket PC users, and avid follower of the latest technology, and a user of an "ancient" Jornada 568 running PPC 2002. In fact, it is my newest PDA at something like five years old now.

While I've followed closely the intrcduction of newer PDAs, I have not upgraded, in part because of the high cost (and a wife who is unwilling to let me pay it...)

I suppose I have been lucky thus far that most all of the software I run, and I do run plenty of non-OEM applications, runs on my Jornada. Some of it runs VERY slowly, but it runs.

I can certainly understand the desire by developers to limit the number of platforms they support. Especially when Microsoft eliminates support in a platform in the development environment, it is increasingly expensive to maintain support for old versions of the OS and/or the old hardware (since in the Pocket PC world, they're related).

So what to do? Those of us on older platforms certainly don't want to be left out in the cold, but at the same time I recognize that you can't stop progress and at some point developers have to move on. Heck, I stopped hoping for updated software for my old Jornada 850 clamshell (which I also still use) a long time ago. :)

I think SteveHoward999 actually had the best idea. Developers should go ahead and announce an end to support for PPC 2002 on the latest versions of their products, but should continue to make available, perhaps at a discount, an older version that would still run on those devices. Presumably the defects in those older versions would have been fixed by virtue of those versions being older, so continued software maintenance on those older versions should not be requried, for the most part, so the cost to the developer to keep that old version available should be negligible.

However, a developer who commits to continued defect fixes to that old version, even without any new features without upgrading, would probably receive a great deal of gratitude from customers on the old platform and that gratitude could turn into revenue once the customer does decide to upgrade to a new PDA.

[alex_kac]

Whoa whoa whoa here. Why the fuss?

There is no need to drop WM2002 support. None at all. VS2005 supports WM2002...with a tiny bit of editing of the project configs. Nothing any more serious than what was required in eVC 4.

So enjoy your latest development tools. And let your customers enjoy WM2002 and up compatibility.

As for APIs, sure there are SOME that are WM2003 and up only, but I've actually found that most APIs that *I* have used that are marked as such actually work in WM2002 with no problems that I can discern in testing, and for the few that don't that you require, just use GetProcAddress.

I believe I only am using GetProcAddress() on one function for SE support.

[snayar]

In my case I selected "other" because I'm running a version of Windows Mobile 5.0 (1.50f.blWWE and no, it's not officially released yet) but my previous version was Windows Mobile 2003 SE (1.72.00WWE). I also have an old but trusty Toshiba e10 running Windows Pocket PC 2002.

But at the end, I would rather ask the developers to spend their time on new and improved OS application development.

[mel]

Whoa whoa whoa here. Why the fuss?

There is no need to drop WM2002 support. None at all. VS2005 supports WM2002...with a tiny bit of editing of the project configs. Nothing any more serious than what was required in eVC 4.


Alex,
I agree with you about being able to create 2002 compatible binaries with VS 2005. But this approach has always made me a bit uncomfortable since it is basically an unsupported hack. But if using this option, do you distribute a single .exe created with the WinCE3 (2002) settings, thus targeting the lowest common denominator, or do you distribute separate .exe's for each platform?

I'm also trying to find the best way to handle the big UI changes between WM5 and all previous Pocket PC versions - the new softkeys. Here's how FlexWallet would like without toolbars on WM2003 and WM5:

http://www.twopeaks.com/temp/wm2003.jpg http://www.twopeaks.com/temp/wm2005.jpg

The UI looks very cool on WM5, but bland (almost unprofessional) on WM2003. How are you and other developers handling these UI changes while minimizing extra-code and development effort?

-Mel

[alex_kac]

How is it a hack? Think about it a bit. You're just telling it to link against different libs. Nothing else. The compiler is the same. Nothing changes there. Only the linking to the libs - which are the same libs you link to in eVC3.

If you can link against third party DLLs (like tGetFile) which are build using different tools or heck - even the OS DLLs (which are built using custom tools at MS - not Visual Studio or eVC) then what harm is there in linking against earlier libs?

Plus, I asked the guys at MS when I was there directly about this, and they said go for it :)

Great question on WM5. Well first, I have two binaries - one for WM2002/2003 and one for WM5. We have to anyway because of the POOM internal changes. So since we're doing that, we just #ifdef the changes.

That works great for everything BUT the menu system. So for the menus, we've got a soft_key base menu (two menu buttons, but no menuitems) which we load on WM5 devices (if the user wants it - its user configurable to use Soft Keys on WM5 or not). Then if the user has chosen soft keys, I basically have a little function that loads the text menus from the graphical menu bar that we use for Wm2003 or on WM5 if the user prefers not to have soft keys, and add them dynamically to the soft keys.

This gives us the flexibility of really having one set of menu resources, but also the flexibility for the user to choose Soft Keys or normal menus.

[Robert Levy]

Whoa whoa whoa here. Why the fuss?

There is no need to drop WM2002 support. None at all. VS2005 supports WM2002...with a tiny bit of editing of the project configs. Nothing any more serious than what was required in eVC 4.

Bingo. There are reasonable (though officially unsupported) solutions for native developers.

And for managed developers, the original decision to not support .NET CF 1.0 in VS 2005 was reversed after we got lots of feedback from developers who wanted to continue being able to reach PPC &amp; SP 2002.

You can't connect to a 2002 emulator or device for debugging purposes but building compatible binaries is possible.

[iant54]

To qualify my answer - I have Windows Mobile 2003 on my i-mate PocketPC, and Windows Mobile 2003 SE on my Loox 720. The former I use as my constantly-in-use device, while the latter is more for entertainment/GPS/reference.

My most recent ex-devices are both owned by my sister, one being an iPaq 3970 running 2002, and the other being an iPaq 5550.

[Menneisyys]

I look at a bit differently. There is a ton of software out there for all devices, including the older ones. What we are talking about now is new software. I would expect that developing new software and making it backward compatible would limit you to only doing things the old stuff can support. You wouldn't be able to take advantage of new features the new OS allows. That wouldn't make it very competative with stuff that does offer the new features.
Not true. Software compiled for newer versions could take advantage of newer features, while software targeted for older versions might be missing some features, but would still offer value to customers. Look at Resco, for example -- they still support Handheld PCs. :!:

Also, as my previous post mentioned, I disagree with your implication that new software has to use such new features. I bet there are plenty of classes of software that people would buy that don't require features only available in Windows Mobile 2003 or later.

Steve

I agree with Steve, adding that WM2003 itself doesn't know much more than PPC2k2. It's very hard to come up with real improvements except for the speed/Xscale optimization (which is "only" a performance issue), the WZC, some additional system color settings and some minor other additions (for example, the ability to define exceptions in Work URL's). PPC2k2 was much larger a leap compared to PPC2k (for example, the entire Connectivity schema has been reimplemented) than WM2003 to PPC2k2 (again, if you don't take performance issues into consideration).

Therefore, there is very little WM2003 system-level functionality that is missing from PPC2k2.

BTW, sometimes I'd prefer PPC2k MIPS support, especially with games that rely on the D-pad :) Nothing can beat the old Cassiopeias, gaming-wise.

[ctitanic]

Actually, I'd argue that that isn't really where the problem is. The problem is implicit in this discussion -- Microsoft keeps dropping support for older versions of the operating system with each new release of the developer tools.

If Microsoft didn't do that, third-party developers would have a much easier time supporting older devices. Yes, they wouldn't be able to support some newer features perhaps, but I don't think all classes of software need those features.

Steve

that's the main point. For those using C# or VB .NET the Visual Studio 2005 upgrade means only one thing: not support of 2000 and 2002. Unless they keep a second machine with the old copy of Visual Studio. In my case as a freelance I keep only one machine loaded, my laptop and I'm planning to upgrade VS this fall so unless We find a way to keep supporting 2000 and 2002 within VS down on the road I don't see me supporting those OS any more. And it's not because I want not to support them.

[lapchinj]

...I keep only one machine loaded, my laptop and I'm planning to upgrade VS this fall so unless We find a way to keep supporting 2000 and 2002 within VS down on the road I don't see me supporting those OS any more. And it's not because I want not to support them.But you become also stuck. You have to support your customer base. If you can upgrade the people then it's a no brainer and your very lucky to have those types of clients. I'm stuck with about 75% of my clients that will not upgrade unless I can prove a real ROI. I had a customer that was still using a DG-MV20000 until 1999. :roll:

Jeff-

[alex_kac]

Actually, I'd argue that that isn't really where the problem is. The problem is implicit in this discussion -- Microsoft keeps dropping support for older versions of the operating system with each new release of the developer tools.

If Microsoft didn't do that, third-party developers would have a much easier time supporting older devices. Yes, they wouldn't be able to support some newer features perhaps, but I don't think all classes of software need those features.

Steve

that's the main point. For those using C# or VB .NET the Visual Studio 2005 upgrade means only one thing: not support of 2000 and 2002. Unless they keep a second machine with the old copy of Visual Studio. In my case as a freelance I keep only one machine loaded, my laptop and I'm planning to upgrade VS this fall so unless We find a way to keep supporting 2000 and 2002 within VS down on the road I don't see me supporting those OS any more. And it's not because I want not to support them.

Virtual PC my friend :) VS2005 doesn't run well under it at beta 2, but VS2003 should. And for just making a new build - it should work.

[ctitanic]

Actually, I'd argue that that isn't really where the problem is. The problem is implicit in this discussion -- Microsoft keeps dropping support for older versions of the operating system with each new release of the developer tools.

If Microsoft didn't do that, third-party developers would have a much easier time supporting older devices. Yes, they wouldn't be able to support some newer features perhaps, but I don't think all classes of software need those features.

Steve

that's the main point. For those using C# or VB .NET the Visual Studio 2005 upgrade means only one thing: not support of 2000 and 2002. Unless they keep a second machine with the old copy of Visual Studio. In my case as a freelance I keep only one machine loaded, my laptop and I'm planning to upgrade VS this fall so unless We find a way to keep supporting 2000 and 2002 within VS down on the road I don't see me supporting those OS any more. And it's not because I want not to support them.

Virtual PC my friend :) VS2005 doesn't run well under it at beta 2, but VS2003 should. And for just making a new build - it should work.

if VS2005 does not work well may be VS2003 does work either. ;) But, that could be a solution, let me keep it in mind. Thanks Alex.

[alex_kac]

I'm told that VS2003 does work in VPC, but VS2005 hasn't been well tested in it yet. I tried it a pre-beta 2 in VPC and it worked, but locked up in network access.

[Robert Levy]

For those using C# or VB .NET the Visual Studio 2005 upgrade means only one thing: not support of 2000 and 2002.

Actually, as I mentioned earlier in this thread, VS 2005 *will* allow you to build applictions that target v1 of .NET CF. CF v1 is supported on Pocket PC 2000, Pocket PC 2002, Windows Mobile 2003, Windows Mobile 2003 Second Edition, and Windows Mobile 5.0.

[Flynn Arrowstarr]

if VS2005 does not work well may be VS2003 does work either. ;) But, that could be a solution, let me keep it in mind. Thanks Alex.

I've used VS 2003 in Virtual PC 2004 SP1 with no problems. Connects to the emulator and debugs properly. It can be slow (depending on physical and virtual memory, processor speeds, etc.). The only caveat I can think of is you can't debug on the device natively unless you can fake a network connection on the emulated system to the device. There's no direct USB support in the emulated system, unfortunately.

There may be a way to connect through the network bridge, but I haven't discovered a way to do that yet. I sense a new project coming on, heh.

As for the poll, I voted other. I have three devices -- a Hitachi ePlate running Windows CE 2.11 (H/PC Professional), a Dell Axim x5 running PPC 2k2 (Mrs. Flynn's device), and a Dell Axim x3a running WM2003. While the idea of "hacking" the project file to make a compatable exe for PPC 2k2 is fine in practice, to me it becomes the same situation as we had with eVB and WM 2003 devices. No emulation and no direct device connection makes it difficult to write software for a device.

In eVB, I had a little easier time because I could just target PPC 2k2 devices and things would work as far as emulation. But it became clear that there were some bugs in the eVB runtime for WM 2003 that made debugging on the device very difficult. One bug revolves around the way Combo Box events are handled, which effectively destroyed part of my program. I still keep eVB around though, as it is the only thing I have to write for my H/PC.

Flynn

[stingraze]

Indeed it's true. I still use a Pocket PC 2000 device. I don't think developers should cut Pocket PC 2002 out of their software just yet..

[Steve Jordan]

Although I have a WM2003 device, I'd rather see Two Peaks continue to support the 2002 apps. Supporting customers, even past ones, should be the cost of doing business, and I've never liked companies that cut off customers because their purchases are not up-to-date and contributing to their latest profit margins. That's how customers get alienated and buy the other guy's apps, when it is time for them to finally buy again.

All of this is related to the bigger problem of planned obsolescence inherent in the evolving M$/PPC platforms anyway. We as customers have allowed ourselves to fall into a trap set by Microsoft and its software cohorts, getting locked into a platform that will constantly change and render its past versions useless over time. To an extent, we shouldn't complain... we were fool enough to adopt this tech, knowing M$ as we do and knowing what would happen.

But now that PPC HW can potentially hold an entire OS, and OSs like Linux are building in popularity and apps, we should be looking for alternatives that break the planned obsolescence mold.

[Zman]

Handango would be a much better source for this information, I don't know why they don't include this type of information in their Yardstick.

[Don't Panic!]

I've got one of each but I use the WM2003SE device the most.