I'm hoping some day that he goes into more detail about the system he uses to catalog conversations. I'm actually trying to work on a project similar to that.
An idea that I've been dreaming of (and kicking around some prototypes of) for a while now is a integrated archive of all the communications services/mediums I use these days. I have close contacts who I may interact with across 5 or 6 mediums in a given week - e.g. email, IM, IRC, SMS, twitter, comments on their blog posts. I'd like to be able to go into a contact list, click a buddy, and get a timeline of my interactions with them across these mediums. Google's gotten part way there in the last few years with email, Google Talk logs and Google Voice voicemails and SMS all being searchable from the Gmail interface but it's not as extensible as I'd like and the results quality (ironically) leaves a lot to be desired.
Yeah, I'd love a Xobni Desktop, maybe they can make it happen. The only problem I have with Xobni now is I find myself forcing important things I want to find later through email so I can easily locate it.
If a "Dropbox" style product would also index everything I send it so I could have a search experience like I do with Xobni I'd certainly pay extra for it.
If a "Dropbox" style product would also index everything I send it so I could have a search experience like I do with Xobni I'd certainly pay extra for it.
That's a funny dream, considering he charges a few thousand for a copy of Mathematica. I'm sure he'd have a larger user base if it were less expensive (or free…).
"I've also had systems built for me personally. One of the more important is my archiving and searching system, which includes 25 years of email (and 20 years of keystroke data), as well as searchable scans of all my archived paper documents."
What do you think he means by "20 years of keystroke data"?
Wolfram seems to keep a searchable index of all his keystrokes.
From the 12 December 2009 issue of New Scientist:
"I'm an information pack rat," he [Stephen Wolfram] confesses. Recording our interview is just the tip of his peculiar obsession with documenting every moment of his life. "I have a keystroke logger that has collected my every keystroke for the last 22 years," he says. "Every day I get an email that tells me how many keystrokes I typed the previous day into each application. I find it slightly interesting." He shrugs off my suggestion that it's a way of securing his immortality; he believes that soon everyone will be doing it.
I don't get that from this article. The one time I met him (at StartupSchool), he seemed like a really pleasant person I'd like to invite over for dinner. He was not faux-humble, but he was also not boastful.
Now, from other publications like NKS, I do get the vibe you mentioned.
My main job is being CEO of Wolfram Research. I spend most of my time figuring out technical and strategic things, sometimes short-term and sometime very long-term.
Compare with:
"I am the founder and CEO of Wolfram Research."
Which contains exactly the same amount of information (because strategic planning is the job of a CEO, isn't it?). No subtle "Well everything I do is pretty heavy and it doesn't fit in this little paragraph, you see?"
Nitpicking, I know, but that's how I feel when I read the §...
I loved the "interview" and also the website. I have always wondered what the great people use on a daily basis.
But it could use some sane URL organization. Sub domains are horrible for quick "select and erase/replace" actions, though I understand the final result is quite beautiful.
Trevor must be half insane. Those people sitting around laughing with the Anybot is one of the most bizarre things I've seen. Can you imagine working in a factory where your boss roamed around the floors as an Anybot?!
I'm not going to say this isn't a great and sensible product, but it is very weird.
Trevor must be half insane, but that's not the reason why.
I've gathered that long term goal is telepresence robots that can enable transcontinental labor arbitrage. That has the potential to make Trevor, at least in the developed world, one of the most despised people alive. Talk about disruptive.
Just imagine the immense satisfaction of delivering a well executed spinning kick to one of these things when it drives up to you on a Monday morning squeaking "Hey, what's happening?" - ka-zonk!
Needs a laser drawing tool (think laser light show) to be even remotely useful, not just a pointer, else how do you describe something they haven't seen?
"No, the $circle_7 goes below the $resistor_3_12, inscribed in $heptagon_3, like in a fronkit. What? Never seen one? Oh hell..."
Until then, good luck designing anything collaboratively. I like the hat-like look to the top, though. Personality without uncanny valley, and still highly functional in design; perfect for something like this.
The solution to this is digital white boards. If you are going as far as Anybot to enable virtual participation in design meetings, I think it would be reasonable to go all the way.
You can just draw a doodle and hold it up to the webcam on your end. It doesn't have to be anything as awesome as a laser light show, though to be fair that would be pretty cool.
And to tell someone to connect lines A through G in this shape, when there are, say, 30 on the board? A doodle + screen works slightly, but certainly not optimally. And even if the laser-drawer can only manage a fairly small amount of detail, it lets you do at least part of an idea, let them see it / trace it, and then a later part, all lined up however you please. The screen's even pretty small, so it's not like you'd be able to show your masterpiece on it all at once either.
Anything non-projected break down any time you've got a large scale - what's an efficient way to point out a sparse mesh between 20 objects on a whiteboard? Point to one, and then the next which it connects to, lather-rinse-repeat to 40 points (more if it's a directional graph), or draw 20 lines and leave them up for people to peruse simultaneously? People just can't remember that many things at once, so you'd rely on someone to marker in your lines... to be erased if you come up with something better? Even changing a single line means several dots + waiting for someone to walk over and do it for you, instead of just changing what's displayed.
And how about completing a circuit diagram? Have fun pointing out a resistor, inductor, and XOR gate[1], dot-by-dot, or showing any complex replacement in a number of sketches + several points showing where each connects. Even if you're only able to draw a couple at a time, being able to draw them in place is a huge efficiency boost.
edit:
Or designing in groups: crowd around the 'bot to look at the screen, while you-or-they are unable to see the board, every time you have input that can't be elaborated by a single dot + words. The time it takes you to sketch something and hold it up to a webcam is the same as or longer than it would take to make those lines displayed, but then everyone can see it at once, as you intended it.
in response to the last "the setup" interview posted here, i created a "setups" subreddit for people to post links to their own articles describing their workspaces
I like a man who spouts off about how great it is to use software builds that are (gasp!) an hour old! That's so 1337 dude! Welcome to the world of open source, douche bag.
I appreciate that I probably shouldn't be nagging on the guy who built Mathematica. But seriously? webMathematica? yawn.
Click the down button: it'll knock them down another karma point and fade out their comment more, but it won't display lower than -4 to prevent pile-on downvoting to -50.
Wil Shipley talked about it upon meeting him at Ted in this blog entry: http://wilshipley.com/blog/2010/03/ted-2010.html (towards the bottom of the entry)