2 ms·

I think you're trying to solve the generic problem -- how can I speak and have things magically delivered to me? That, of course, is a long, long way from here

I think you're trying to solve the generic problem -- how can I speak and have things magically delivered to me? That, of course, is a long, long way from here. But I don't think we need to boil the ocean. Fact is, you could make 10 or 20 categories of things folks do on the internet and cover a huge chunk of activity. As one commenter pointed out, lots of folks want pictures of folks from the party last night, not "facebook". There's no reason you couldn't have a handheld device (or program) that you pushed a button and it showed you recent pictures from your friends -- no matter where they posted them online. If you wanted to get ambitious, you might could come up with 100 use-cases that would cover an enormous amount of online activity: checking my account balances, hearing from friends, sharing pictures or videos, providing feedback on what I thought after consuming some kind of material, etc. I'd be willing to be such a list would be so complete that's it's not important about the remaining items. You'd simply use a browser and an URL for those. I bet you could create a device with 20 or so buttons that would cover so much common activity on the net as to make the device useful as a stand-alone appliance. The beauty of this, of course, is that since people's behavior is more or less unchanged (people have been sharing pictures with each other ever since there was photography) the same buttons could be easily learned and would work no matter what the underlying technology or transport mechanism was.

swombat 16y ago
And people have been building this sort of stuff... as online services. Why would you build it as anything else than an online service (as the person building it)?
- DanielBMarkham 16y ago
  There are two reasons that come to mind immediately. First, legally you can't use one online service to aggregate from other services -- every walled garden has clear TOS that prevent this. You can, however do the gathering and processing locally. After all, you own your own computer equipment and are solely responsible for how data is collected and presented there. Secondly, and more importantly because of the history, you're trying to solve the problem using the constraints and assumptions that created the problem to begin with. You can't fix seven walled gardens -- by creating yet another walled garden. Stand-alone targeted web apps are great for finding needs and servicing markets, but the nature of the net is monopolistic. That's why many of these stand-alones end up getting flipped. We need to break the technical backbone that keeps creating monopolies if we really want to get rid of them.

swombat 16y ago

And people have been building this sort of stuff... as online services. Why would you build it as anything else than an online service (as the person building it)?

DanielBMarkham 16y ago
There are two reasons that come to mind immediately. First, legally you can't use one online service to aggregate from other services -- every walled garden has clear TOS that prevent this. You can, however do the gathering and processing locally. After all, you own your own computer equipment and are solely responsible for how data is collected and presented there. Secondly, and more importantly because of the history, you're trying to solve the problem using the constraints and assumptions that created the problem to begin with. You can't fix seven walled gardens -- by creating yet another walled garden. Stand-alone targeted web apps are great for finding needs and servicing markets, but the nature of the net is monopolistic. That's why many of these stand-alones end up getting flipped. We need to break the technical backbone that keeps creating monopolies if we really want to get rid of them.