Terry Jones

"Hi Alex Thanks for taking the time to have a look at FluidDB. It's ok if it's not completely clear to begin with - lots of people have that experience. It's a bit like a wiki in a couple of ways. Two are that you never have to ask permission before adding to an object, and that you can ask for an object about anything at all and if it doesn't already exist FluidDB will just make it for you. It's unlike a wiki in that the data is typed, there's a strong permissions model within objects, and there's a query language. There is the general idea though that FluidDB might be able to do for databases what Wikipedia did for encyclopedias. I'm happy to answer any questions. There's a fluiddb-discuss mailing list you're welcome to join: http://groups.google.com/group... Terry" - Terry Jones

"BTW, for your amusement and to illustrate the level at which I tend to think about even tiny (in)efficiencies in how people work with information, here's a mail I sent to some research folks back when I was at SFI. --- From terry Thu Nov 24 05:15:57 -0700 1994 To: researchers-here@santafe.edu,researchers-away@s..., staff@@santafe.edu, rawlins@cs.indiana.edu, mosterin@hydra.unm.edu, adaptive@cs.unm.edu, joe@cs.ualberta.ca, 100237.3565@CompuServe.COM, rob@research.att.com,cjcolbourn@math.uwaterloo.ca, andrewt@basser.cs.su.oz.au Subject: Minimizing hardware wear and tear... Suppose you need to type n contiguous 1's into a file and that you wish to minimize the wear and tear on the mouse and the keyboard. If you only use the keyboard, you'll clearly have to hit the 1 key n times (unless you have an auto-repeating keyboard and can count fast). But if you're using X windows, you can double-click with the first mouse button to select an entire block of ones and then single click with the..." - Terry Jones

"I'm not sure if I'm overall agreeing or disagreeing. I guess I would argue that people who are extraordinarily talented (I do not include myself btw) are not going to be affected that much by the relatively small impact of (most) tool choices. Virtually all software ever written before (say) 1985 was done using what we'd today consider extremely primitive tools. Yet that stuff forms the cornerstones of our modern computing experience - it includes compilers, operating systems, networking, bitmapped graphics, any number of device drivers, etc. Tons and tons of intricate low-level fiddly code. Those efforts with those tools remind me of ancient engineering undertakings with the primitive tools of those eras. That's interesting as an historical fact. More dynamic and pertinent in some way is what you're also interested in: what do talented people *choose* to use today? Unfortunately it's hard to get good data as, surprisingly, many people are just using what they've always used or what..." - Terry Jones

"Yes, I tend to agree. Consider that things like UNIX and any amount of extraordinarily complex software was written by people banging away on 24x80 VT100s in many cases without even a visual text editor, and compare that to today's need for huge screens, $1K chairs, massive amounts of RAM/disk/whatever, and it all seems a bit ridiculous. I'd love to have more hardware, but have written the whole of FluidDB with one other person with just 2 15" ThinkPads, no external monitors, etc. You're right on software tools too. It's all over the map, but most of the hackers I know prefer simplicity - not to have an IDE, to use a simple text editor, the command line, etc. All the other stuff feels like eye candy and the fact is that in many cases you have to *wait* for it. I love the way apple stuff looks, but I can't stand waiting for my terminal or editor window to pop up, or for a new one to launch - it drives me batty. It happens in the blink of an eye under linux, yet i have to wait 0.5..." - Terry Jones

"Hi David I don't know either. I tend to think it's unknowable. If there was an answer, the entrepreneurial and investor worlds would (probably) look very different. You can only think you know. But, just for example, how could we factor out the distorting impact that receiving an investment has on the successful outcome probability distribution? As I said in one of my infinite posts, I think entrepreneurs are much more up front about the influence of chance. Unlike (some) VCs, they're not in the business of convincing people they have a long term knack of being able to distinguish madness from genius. Perhaps that gets us back to the herd theme, not that I would claim that it's any easier to accurately finger value amongst the larger pool of entrepreneurs who are doing things significantly less likely to be wildly misguided." - Terry Jones