westmeadboy - interesting, but mostly related to mobile app front-ends to websites, I think; more of a branding than a sales question in that case.
RobRedbeard - maybe it's just recent history; easily the single worst business decision I've made in the decade or so I've been doing this was deciding to put Pleco 2.0 on Palm OS along with Windows Mobile. We spent roughly a year getting it working only to have Palm discontinue Palm OS a few months later; we did it for the sake of our customers who loved Palm / had recently bought Palms / etc. If it weren't for that, we might very well be working on Android version now simply for lack of anything better to do with desktop / vastly-improved iPhone versions already out the door.
So my feeling now is that accommodating customers' platform preferences has to take a backseat to making great software; there are $10 iPhone apps that do everything our 2003-vintage software did (albeit with not as nice a dictionary as Oxford's), and by the end of the year they'll probably have caught up to where we were in 2005-6. If we don't keep adding features / refining our design, if we don't stop these endless cycles of porting and porting, pretty soon we're going to be Just Another Chinese Software Company, with nothing but a longer history to differentiate us from the pack. (we've never been a leader in platform support - there are companies that have been doing this almost as long as Pleco that support 5 or 6 platforms now)
I appreciate it's annoying to buy into something only to find that it's not available on your new preferred platform, but this isn't like a music or video file, there's a lot more than DRM involved in supporting our databases; we could decrypt all of our database files and even release format documentation and it'd still take you the better part of a year to get to a point where you could use them, there's 10 years' worth of difficult-to-explain optimizations and tweaks in there.
We certainly might be able to do it in less time than that, but there's no realistic way this would stop at releasing a crappy Lite version - however many This Is Not A Full Version or We Don't Really Give A Damn About Android But Are Doing This Just To Be Nice warnings we might put on it, people would still talk about it, blog about it, show it to their friends, review it, download it, complain about how mediocre / slow / feature-lacking it was compared to our Palm / WM / iPhone software, threaten to abandon us for one of the competitors you're all telling me don't exist... if we're on Android then we're on Android, and we risk irreparably damaging our brand by releasing a crappy and poorly-supported Pleco for Android. Better to stick to one platform and disappoint people who don't like that platform than spread ourselves too thin and ruin a reputation we've spent a decade building.
(the "Lite" version I described earlier was considerably more advanced - and would take considerably longer to develop - than what you've been talking about)
And honestly, if we are launching new platform ports out of a sense of obligation to our customers there's a much stronger case to be made for desktops - we've actually been saying we're going to develop a desktop version for quite a few years now, so advocates of that have a much more legitimate gripe with us than Android users. (and FWIW I got a total of about 5x as many emails in response to my last announcement email complaining about the no-offline-desktop-version part than about the no-Android-version part, though if I'm going to make business decisions based solely on numbers of emails received then Android would have to take a backseat to BlackBerry anyway)