Temporary Solutions to Temporary Problems
Being a geek I'm always aware of the limitations of different technologies. Being a wanna-be entrepeneur I'm always thinking of ways to exploit these limitations for monetary gain. But the question always comes up - what is going to be a temporary limitation and what's going to be around for a while, and thus worth doing something about.
For example, my latest thoughts turn to RSS. RSS and Mobiles *do not* mix well. Why? Well, if you have an offline aggregator that needs to check RSS documents, it will constantly be downloading the same file, or something that is 95% the same, at all times. RSS is an incredibly inefficient transport of information. Unless you have some sort of query system on your server which allows you to say "Give me the latest news since this date/time", you're wasting a ton of bandwidth.
Sitting here at home with my wired up ADSL, that's not a problem. I've got an aggregator right now running in the background hitting 50 sites every 20 minutes, asking for new versions - and since many sites don't bother with 304 messages "not changed" (and hell, I don't even know if my aggregator even supports that message), I'm constantly downloading the same file over and over and over again. But even if the RSS *has* changed, it's most likely only changed a small bit, which means you're still wasting bandwidth.
So my thought as an entrepeneur is simply this: Create a webservice which allows the user to add feeds, and create a custom J2ME RSS Reader which queries this web service for latest changes so that they can be read offline, yet be bandwidth conservative. Pretty straight forward solution to a simple problem. Both sides of the equasion, actually, would be pretty easy to set up from a technical standpoint as well. Set it up, add some text-based advertisements to the mix and you've got yourself a biz-plan.
Now the question is: "Is mobile bandwidth constraints a temporary problem?" It's a real question. Here in Spain, I pay $30 for 20MB of GPRS data, and after that 2.5 eurocents per kilobyte (i.e. $25 per MB). So to me, this type of service is *insanely* useful. Now to Matt in the U.S. who has a spankin' new T-Mobile GPRS unlimited data service, he could *really* care less about spending a few extra k's. Maybe the data would arrive a little quicker and more efficiently in the system above, but for the most part, he'd have no real economic reason to use my service.
There are *tons* of areas like this in the mobile world. The limitations of J2ME are a big one. Will people 1) live with the limitations or 2) move en-masse to native applications and OSes like Symbian? Also, the power of phones themselves is a temporary problem. Is it worth it to spend time working around the limitations of current hardware, optimizing your apps to get every ounce of power it can, or just develop at a normal pace and let Moore's law do its thing. Look at how Microsoft does things. It just *waits* for new hardware to come out to support its bloated code. It's worked for over a decade for them, and I don't seem them changing any time soon.
Anyways, you can go nuts trying to predict how long a limitation will last for. Say you realize there's a bunch of problems with inter-carrier MMS and come up with some solution. Great! But how long is that solution good for? The carriers are working on interop furiously right now, so its a window that's only going be open for a very short time. Is it worth it to pursue that strategy when the whole reason for it existing is quickly going away? Look at all the chat apps out there (and more coming) how long do you think it'll be before phones come with that sort of app as as standard? Not long.
All these opportunities and technolgies out there are like Internet time again, only it's "Mobile Time" and it seems even faster. Every 6 months a whole SLEW of new phones gets launched making everything before them seem useless and drab. GPRS just got rolled out and now many carriers are working on EDGE or 3G. Sure these solutions may be a year or so out. But when you're writing a business plan, you've got to think 5 years out at least. Christ KNOWS what will be needed in 5 years.
Okay, that's my new business plan: "Rapidly developing applications that meet short term needs of new technologies as they appear and switching to other short term opportunities when those windows close."
What do you think? Wanna invest? ;-)