1,458 Posts served
6,169 Conversations started
When I explain my job to people who ask me what I do, one of the most common phrases I use is: "People don't buy hardware for hardware's sake. They buy the hardware because of what software can make it do." Now I recognize that techno-geeks (like me) enjoy having the latest gadget just because it is 'cool'; but we're a niche market that isn't going drive serious ROI.
One of the main justifications that I hear for the UMPC is: "It's better than a hand-held PC or smart phone because it has my favorite client OS! All of my desktop and laptop apps will work out-of-the-box on the UMPC." Nice. I think editing my digital photos on my UMPC sounds exhilarating, but not as nice as on a dual-head desktop with two 22-inch wide-screen monitors. The rebuttal: "But it's mobile! You can't take your desktop with you on an airplane." Uh, that is what laptops are for. "But it's MORE mobile than a laptop! It fits in a purse; you don't have to carry any extra cases."
More mobile. Now we're getting somewhere. This à¼ber-mobility comes at a price: smaller screen, reduced processing and storage resources, limited input modalities. Michael Molin proposes a cell PC here, but that is à¼berà¼ber-mobility and it seems as usable as, well, a cell-phone. This price for mobility may be worth the paying for many. But the main question then is this: how many full-client-OS applications are REALLY going to be portable to a form factor with these limits?
While many of these categories certainly make sense on a UMPC, most if not all will require some kind of enabling (UI modification, alternative inputs, etc). But just because you CAN enable for UMPC does it mean you SHOULD? I could certainly enable my favorite photo editing software for UMPC, but would I ever use it on that platform?
And finally, the real zinger: If only a subset of these applications really make sense on the UMPC form factor, do you need your favorite (and expensive), general purpose, client OS? Or does it make more sense to stack specific applications onto a cheaper, customizable OS that does just what the software needs it to do?
Aaron Tersteeg (Intel)
Status Points: 4927
Brown Belt