May 20th, 2010

device

Software old and new, part 1: Dreamwidth

Having noticed this morning that my Dreamwidth account had expired and gone back to Free, I just re-upped. I'm not using it significantly yet, but I still very much approve of the project and would like to see it succeed.

But it does remind me to ask: which of my friends are actually using DW actively? I want to make sure my reading list covers at least all the people I'm actively following on other social networks, and if more of my friends were posting there, I might well pay it more regular attention. I think I only know of two who are posting there, but I may be missing folks.

(For those coming in late: DW is most easily thought of as a good alternative to LiveJournal, currently in beta. They started with LJ's open-source software, and have been enhancing it at a pretty good clip. Frankly, they have both a better attitude towards their customers and towards the software itself than LJ does these days, and care more about doing a really good job. I'm still mainly focused on LJ because that's where the bulk of my friends are, but I do think DW will eventually be the better system, and may already be...)
device

Software old and new 2: Google Prediction API

I hadn't even known about this one when I planned out this morning's posts, but this just dropped into my lap. No shit: Google is helping you predict the future.

The Google Prediction API is one of their new Labs projects, and just makes so much sense it hurts. Google has near-infinite computing power, and loads of experience by now in analytical and predictive techniques. So they're beginning to generalize that into a public API: you feed it data, it learns from that data and susses stuff from it. (Mostly it's about classifying and categorizing information, but they do say themselves that a major usage is prediction. Surely somebody's going to call this thing Crystal Ball by the time it becomes real.)

Their suggested uses on the web page include:
  • Language identification
  • Customer sentiment analysis
  • Product recommendations & upsell opportunities
  • Message routing decisions
  • Diagnostics
  • Document and email classification
  • Suspicious activity identification
  • Churn analysis
There's lots more you could do with it, of course -- basically, they're throwing open some useful basic technology, to see what people so with it.

From Google's POV it's a slam-dunk, and I do expect this one to make it out of the Labs. It's currently free, but of course you are feeding all that yummy data into the great maw of Google. The Terms of Service are reasonable, but I suspect they leave Google some wiggle-room to at least learn from that data, even if they disclaim any rights to the data itself.

Still, this is a neat toy. Ideas of what you might try to use it for?
device

Software old and new 3: Google Wave

On the existing-software front: Google Wave is now officially in public beta, and open to everybody -- I gather you no longer need an invite code.

Yes, I'm a bit of a cheerleader for the project, but seriously: it's worthwhile to sign up and get a feel for it. When it first came out, it was basically an early alpha: slow, buggy and missing a bunch of necessary features. It now deserves the beta label -- it's a good ways from done yet, but it's genuinely useful and a lot more stable. It's still hellishly demanding on both your bandwidth and browser (translation: you still can't use it with Internet Explorer, which is just too slow), but seems to work pretty well and is gradually getting easier to use.

The standard question is: Why would I want to use it? The answer is for collaborating on documents. At the moment, the sweet spot is still working together on text that is content-heavy and format-light: if you have a bunch of people who are trying to collaborate on a text project, Wave is the best tool I'm aware of, combining the best elements of collaborative editing, mind-mapping and conversation. It's particularly excellent for meaty projects where you want to be able to work together in real-time when possible, but where not everybody can necessarily be there at the same time.

There are a bunch of other good applications as well. Brainstorming seems to be another sweet spot: several of the most useful Waves I've participated in have been multi-day brainstorming sessions. And people are slowly hooking in gadgets to make it useful in more ways. But if for nothing else, I've found it enormously useful for cases where we have 3-20 people who need to produce a document over the span of a few days. It's not the only tool for such jobs, but it does seem to be the best one to date.

So I recommend checking it out. Feel free to poke at me there (my usual handle @googlewave.com) -- it's also good for random chat, and conversations in Wave are one of the better ways to learn it. The access-control tools are still a bit weak, and it doesn't have the social networking capabilities I so desperately crave, but I expect it to eventually become a kick-ass general conversation system, and maybe eventually a genuine rival to LJ...
device

Software old and new 4: Newsmap (and magic wands)

Okay, last of this big blort of online tools.

I spent yesterday afternoon down at a presentation at the Fidelity Center for Applied Technology, which was talking about the way they're integrating all the cool current software trends into their business. For the most part, this was a misfire -- the thing was more aimed at CEOs and VCs, and I was kind of at sea in all the hob-nobbing. But while poking around in their demo area, I came across Newsmap, which is a neat, useful little application.

Basically, Newsmap is a big Flash app (and I mean physically big: it really wants to be used full-screen, and the bigger the screen the better), which scrapes the Google news feed and represents it as a giant Wall of News. You can select what kinds of news you want, both by geography and topic; it pulls in everything it can find, and splashes up boxes of headlines, grouped and colored topically so you can at a glance get a gestalt idea of what's going on in the world, with each box's size proportional to how many related articles Google is finding.

The best way to describe is in science fiction terms. You know how, in various stories, the bad guy always has this gigantic wall of televisions, turned to many different news stations, so he can keep track of everything that's going on? This is a simplied, practical version of the same idea. On one screen, you can see what's happening in the world, and can quickly and easily drill down into it. It's easy to use, intuitive, and really quite powerful.

So if you have a big monitor (and a reasonably current version of Flash), give it a look. I lost a surprising amount of time just surfing via this thing yesterday: it's a delightful window into the world, and looks useful enough that I've made a donation to the author. (Who is soliciting PayPal donations to "take some time off from work and bring it to the next level".)


Postscript: there was one other thing of note that came out of the FCAT presentation. I spent much of the time kind of disengaged on my own, but at the end, I wound up talking with the "Senior Project Manager" who was doing some of the demos -- who was really their in-house hardware geek, basically a Media Lab type who had somehow landed at Fidelity. He showed me his latest pride and joy. It looks like a small wizard's wand, a la Harry Potter; the neat part was when he picked it up and started using it to control all the TVs in the room. *Flick* up, and that one's volume turns up. *Flick* sideways, and it changes channels. And so on: he could do a surprising amount with the thing.

The neat bit is how simple is actually was: he had apparently just built a pair of accelerometers and an RF projector into the wand, with some software to make it into a gesture-based learning remote. Even I (the pure software guy) can pretty much see what he did, but the thing was elegant as hell, and just plain magical in an Arthur C. Clarke kind of way. Probably the coolest gadget I've seen in years: I find myself not just wanting one, but wanting to rip it apart and start enhancing it. The potential is scary and delightful...