glimcat

80 ms·

661.
by glimcat 15y ago
Resource scaling is likely to be a problem long before theoretical economic maximum anyway. The entire question is just an exercise to play with the math.
662.
by glimcat 15y ago
For reference: http://ycombinator.com/apply.html "The ideal company would have two or three founders. We'll consider those with four or five. We're reluctant to accept one-person companies, though we have funded a couple." If there's a st
663.
by glimcat 15y ago
Um, did that guy have a point? If so, what was it?
664.
by glimcat 15y ago
Angels are likely to want to see a prototype, market research, and some proof of traction if possible. Particularly since you're abnormally young. Ideas get stolen all the time, but it's statistically unlikely (if you find that reassuring).
665.
by glimcat 15y ago
Should have gone here instead. I'm pretty sure that extortion is against Twitter's TOS. https://support.twitter.com/forums/26257/entries/15789
666.
by glimcat 15y ago
It varies a lot. From what I've seen, two is most common and more than 5-6 is a bit odd. More than about 8-10 is hard for sociological reasons. How well the team integrates and works together and whether or not you have adequate skill cover
667.
by glimcat 15y ago
Users, customers - if you do a free promotion, you're doing a business transaction with them. It's not like they went and hunted up a rogue APK. 15 minute refund is usually good enough to run the "does this work on my device" check, althoug
668.
by glimcat 15y ago
Sort of? It depends a lot on how you frame the question. To first-order, modern economics goes as the number of connections between nodes. There is quite a lot of potential growth between the current status quo and the maximized economic st
669.
by glimcat 15y ago
A large part of the problem is the "race to the bottom" mentality. Users have been taught (by Apple) that software is $0.99 - if not free. You have to get massive user numbers before that becomes profitable, or find another way out of the t
670.
by glimcat 15y ago
FYI, Greg Bear is also on the team. http://www.subutai.mn/team.html
671.
by glimcat 15y ago
Says "here's a slide deck" and links to Slideshare. Direct link: http://www.slideshare.net/perone/c0x-introduction-to-some-am...
672.
by glimcat 15y ago
Yeah, saw that back when it was first posted. Nothing new in the linked article. Better update, if you go to the site. http://www.raspberrypi.org/ It looks like they've just printed some alpha PCBs. Also, they are now talking about it bei
673.
by glimcat 15y ago
I develop in C# on occasion and use Linux for some stuff. It's not like one precludes the other. But that's sort of weird as a requirement sheet if there's not more info forthcoming.
674.
by glimcat 15y ago
I tried a few zips around the country. All it ever did was tell me to go to San Francisco. Is there something more I'm supposed to be seeing here?
675.
by glimcat 15y ago
It talks like it should do a lot more, but all it lets me do is zoom in on a few panels of an infographic about the planets. Am I missing something?
676.
by glimcat 15y ago
"we can't demand that teachers be excellent in conditions that preclude excellence" We can if we keep judging "excellence" by standardized tests which we water down until they give an answer we like and institute policies which push educato
677.
by glimcat 15y ago
Making grad school far less accessible in exchange for moderate increases for undergrads sounds like a bit of a mixed message. Also, call me cynical, but I can't help but think that this is a game of three-card Monte. What's in their other
678.
by glimcat 15y ago
That link works for me (or rather, dies and gives an error). By way of giving multiple data points.
679.
by glimcat 15y ago
Software, no. Hardware, fairly often.
680.
by glimcat 15y ago
It's not supposed to be perfect. But you'd bloody well suck it up and get used to doing them and pitching them, because they're expected.
681.
by glimcat 15y ago
If it's an automated system inducing the error state, it suggests that you could poison Bing by creating a similar error for popular keywords.
682.
by glimcat 15y ago
Generally they have terms of service - designed to cover them - and maybe some dispute resolution stuff.
683.
by glimcat 15y ago
Experience. Also, trying to sell it to them and seeing if people pay for your product.
684.
by glimcat 15y ago
Yes, it's an extremely naive first-order model. It assumes a series of independent "fair coin toss" events and gives the probability of at least one success using the given values. Really, business success is not a random event, although th
685.
by glimcat 15y ago
Output is inconsistent over time and typically in the microwatt to low milliwatt range. This can be quite limiting. Sensors with a low update frequency are a good example of things that can work.
686.
by glimcat 15y ago
FYI, most competent developers are going to see that and decide it's worth roughly what you give them in free coffee. If you want a competent developer to do work for you and they're not a cofounder, you generally need to be paying them in
687.
by glimcat 15y ago
The major things to worry about are URL autocomplete on smartphones and Google.
688.
by glimcat 15y ago
Yeah, there's no deep data. But assuming it's roughly correct, iterating n quickly can be a very strong strategy. This is part of why the usual "early and often" advice is so effective.
689.
by glimcat 15y ago
It's often more important to have a local branch with good service. PayPal is a sometimes handy as a payment option, but not fun to deal with any more than you have to.
690.
by glimcat 15y ago
"First-time entrepreneurs only have an 18% chance of succeeding. Interestingly, those have previously failed have a 20% chance of succeeding." p1 = 0.18 p2 = 0.20 Naively, p1 || p2 = 0.344 Cumulative probability after N attempts assuming p

More ›