I've been thinking lately that (1) I spend a lot of time "browsing" the web (it has gotten more efficient and at the same time, more intense since weblogs and RSS have taken off; at Outhink, we call this "gathering") and (2) I really like listening to and occasionally watching podcasts, screencasts, etc. Also, (3) I miss good radio (they just don't seem to blend the content and personalities like they used to - or at least I have a harder time finding it now, compared to when I lived in LA with a radio, back when I was a skinny kid).
But I think we can recreate some of that with all our new found web toys. How about we aggregate news and information, according to user preferences (ok, done) and then "read" the information using text-to-speech (I have played with the Java TTS API, see also Say-It-Now) and assigning specific voices (and personalities) to the various information sources (AP, Reuters, Blogs, etc.). It might go something like this:
This just in Bob, Microsoft has gone open-source today with their wiki software...
That's right Sue and not the first time either (more data from another source)...
Let's talk to Mitch Ratcliffe and Marc Cantor about this briefly (audio and blog scrapes)...
Here's some video of the big announcement (scrape)...
Well, that was interesting. Now here's a new little item from OddioKatya's playlist on Webjay...
The time is now 2:15pm...
Don't forget you have a conference call planned for 2:30...
Dave is stopping by the lab for a demo at 4...
Here's a good joke about webloggers(scrape)...
The latest West Wing espisode is now available for viewing(feed/torrent)...
Eric has published a new requirements spec for SpinXpress 3.0 (upload)...
Throw in a little voice recognition (Jon Udell discusses Dragon, which I used several years ago on a NASA-AMES funded simulation project for UI and Human Factors research). Let me say things like "what?", "again?", "bullshit!", "elaborate", "more", "save", "send", etc and have the news app respond appropriately.
I imagine that others have all thought this out already, but it seems like it might be something one could put together (might be fun to do it with existing open-source tools). Jon Udell put together a nice post and screencast recently that I liked that has the right flavor. Let's see if we can remove a lot of "individual" effort that Jon had to go through (Thanks! btw)
See also this great video from the museum of Media History that probably leads to something similar (and possibly more frightening). Does anyone know what motivated making this video? Was it a class
project at Georgia Tech? Does anyone know what the CC rights on this are? I do not see an easy
way to find out at the CreativeCommons.org site. I have to watch again, there are some credits at the end.
Addendium (01/31/2005): 
Recent Comments