GeekForce

I’ve been interested in Ubiquitous Computing for some time.
There is a really interesting analysis over at skilfulminds.com

For those that dont know, ubiquitous computing is the notion that all devices - fridges, toasters, the ring on your finger, all of that stuff should have at least some smarts in it, and it should convey, acquire, and compute information for and from the other devices in this “computing fabric”. See wikipedia:Ubiquitous Computing for more details.

There is real promise in it, but it’s main challenges right now, in my opinion are not so much findability, they are in two places:
1) scale of the systems necessary to maintain a Ubiquitous infrastructure.
In part I think this way, because until recently has only been done in small research labs. Noone has really DONE this for say a neighborhood, or even an ISP - if you want to take the network as a first step toward deployment…
2) My biggest issue here is that common end users still have trouble remembering where they saved a file, or that they saved it in the first place.

Lets take a step back here…

When we save a document, more than half the time we ahve to remember where we saved it. THINK ABOUT THIS. Especially where software development is concerned, its not like you can just make your .h file, and put it anywhere. You have to be able to communicate its whereabouts to the other files that make up your source code.

As developeers, we are used to having to know this, and frown on users that arent aware of where they save things, or that dont actively try to organize files in their home directory.

But end users, neither know, nor care about where files are they just expect to see them the next time they go looking for them. While this may seem like a simple problem to solve, there are at this time many great minds trying to make this easier to do.

A while back Xerox extended their desktop systems with a different metaphor for working with the files that we create. it was a basically a glorified timeline. It never really took off - it had great promise, just like the alto, and star…

My point here is that a big part of the low level things that Ubiquitous Computing has to do is each device, or node has to be able to discover each other, but more than that, an end user has to be able to deal with the onslaught of data, and they need ot be able to filter and effectively search this giant diaphanous body of data.

Right now, its not going to happen soon. Users still dont really understand how to work wiith files, and until either that changes, or - Think about this: we arent working with files anymore, this will be a big problem.

With this kind of obstacle - the difficulty that people ahve remembering, or even KNOWING where things are, this technology isn’t going to get outside of the elite geek-o-sphere for some time, and we (us geeks) all know how to find things. So, there’s not really an issue there, other than a really good research project.

There are many efforts out there right now that are attempting to make this file saving, finding, searching problem a non issue.

My favorites are the files-arent-files-anymore projects. Documents are streams of bits, at least to the computer. Why cant the computer store and index the content, and let me focus on work. essentially this would sort of model would do away with us having to save things, and remember where we put them, those details are left to the computer, what the machine can synthesize for us then, is all the documents that we ahve ever worked on, we can search the corpus with keywords, we can see a timeline of our actions on some or all of the documents we’ve worked on, and so on.

This changes the metaphor, and the workflow for humans sitting in front of a machine in a dramatic way. we’re now concerned about the content of the document, and its logical characteristics. It doesn’t really have a useful ‘location on disk’ anymore, as this was abstracted away by the file system.

This is, in my opinion one of the changes that we will need to get converged into current systems, in order for other work to bear fruit. Yes, I’m saying that I think that humans need to change their behavior, and the way to do that is to deploy some more modern technology. Mostly because people dont bother to remember where they stored something, they just expect to be able to work on it when they to work the next day.

If people were used to being confronted with the deluge that is a google results page, then they’ll respond when they are confronted by a larger deluge of information from a Ubiquitous environment.

What do you think?