glimcat

39 ms·

331.
by glimcat 15y ago
Bluetooth Low Energy might end up being a big thing once there are phones out which have a radio for it. The spec is intended for devices which send small amounts of data with minimal power consumption, so it's well suited to things like en
332.
by glimcat 15y ago
What's better, a fork or a spoon? It depends on what you're eating...
333.
by glimcat 15y ago
A few hundred. It varies depending on how much I want to do or don't want to do what they're asking for. Tends to be "consulting" and monthly. I think the last one worked out to about $400-500/hr, but I wasn't tracking time closely.
334.
by glimcat 15y ago
Seems a wee bit arbitrary. And convenient.
335.
by glimcat 15y ago
I'm pretty sure Amazon will let you do this. Kindle Singles might be a good place. You can then do whatever to promote them with Amazon as the channel.
336.
by glimcat 15y ago
Because he can close his eyes and wish, I suppose. It's a netbook, there are microcontrollers. The microcontrollers have firmware. Besides which, I somehow doubt that all the relevant microcircuitry is open source.
337.
by glimcat 15y ago
Rather than a comments system, I'd prefer something which polls around for copies of the link to see if it turns up on HN or whatever, then includes those. Free moderation, better validation. Comments can be important if you're trying to bu
338.
by glimcat 15y ago
I'm wondering if running the phones off isn't a little too routine to get attention. How would one go about aggregating mail so as to pull a "Miracle on 34th Street" stunt? Several hundred thousand letters ought to do it.
339.
by glimcat 15y ago
Monthly billing tends to be superior because it reduces the "just one more" barrier by a factor of twelve. You do need to have a mean customer lifetime of at least a year for this to be significant. Big, heavily bureaucratic institutions wi
340.
by glimcat 15y ago
It's been done repeatedly, e.g. by Wilson at ACM 2004. It just hasn't passed the complex multidimensional optimization problem which is the process of actually pushing such technologies to market. http://citeseerx.ist.psu.edu/viewdoc/downl
341.
by glimcat 15y ago
If your algorithm isn't black sorcery and does anything visible to the end user, another programmer can reproduce it.
342.
by glimcat 15y ago
Manage your priorities like you're going to die at 30. Take care of your health despite the cost and inconvenience. You may be able to get away with a lot of self-neglect at 20, but it's much harder to repair later.
343.
by glimcat 15y ago
From about five seconds of Google, it looks like the actual product was made by David Kotkin and is considered to be rather significant for disabled users. http://www.ablegamers.com/hardware-news/interview-with-aveng... Paul Christoforo i
344.
by glimcat 15y ago
Build something cool. Try to get people to give you money for it. Evaluate the results. Repeat as necessary.
345.
by glimcat 15y ago
If your business is growing too fast for you to have time for anything else, that might be a good reason to take a break from college. If you're doing badly in your classes, that might be a good reason to revisit your study habits.
346.
by glimcat 15y ago
Unless your problem is "we solve epicycles" - neither is going to insulate you from obsolescence. The benefit to targeting a "problem" is that you have a preexisting need. An "idea" with no existing "problem" may require several years of ex
347.
by glimcat 15y ago
2011 is 404, here's 2010. All prior to 2010 and the "all time" are also 404. http://www.bingocardcreator.com/expenses/expenses-pie-chart?...
348.
by glimcat 15y ago
LLCs are fine for many things - but not generally for dealing with investments or equity. Whether they're worthwhile as an intermediate step depends on what you're doing and when. They're rarely appropriate for "build to sell" companies whe
349.
by glimcat 15y ago
Loops are to software development as spelling is to writing a novel.
350.
by glimcat 15y ago
37signals posted a good series on this, including quantitative results. http://37signals.com/svn/posts/2977-behind-the-scenes-highri... http://37signals.com/svn/posts/2983-behind-the-scenes-ab-tes... http://37signals.com/svn/posts/2991
351.
by glimcat 15y ago
I've explored this idea before. There are a few problems, although not necessarily fatal ones. Major categories are engineering and sales. Engineering, you have to build the thing. The best solution I found was bistable mechanical relays. A
352.
by glimcat 15y ago
Because scaling limits. The stuff we're already getting uses very highly doped silicon. You could go smaller or faster if you could dope it more to keep the field characteristics viable, but more doping would screw up the silicon lattice. "
353.
by glimcat 15y ago
I'd worry about fixing your site first. Also, conversion rates and usability are two completely different things. Usability can affect conversion rates - only relationship. Usability is also not something you're going to diagnose adequately
354.
by glimcat 15y ago
Yeah...certain CSS tricks are "use with extreme caution" e.g. the ability to hide all overflow with no scroll. In general, if you don't know how to do it right, you should start by doing it simple. Unfortunately, this conflicts with the imp
355.
by glimcat 15y ago
Hacking the user agent string often helps, if your browser supports it.
356.
by glimcat 15y ago
I don't think the problem is so much contextual logic as it is failing to make errors easy to recover from.
357.
by glimcat 15y ago
Naive Bayes: the "good enough" classifier.
358.
by glimcat 15y ago
Also, here's a guy who's looking for collaborators to help fight this by making Facebook apps: http://www.reddit.com/r/SOPA/comments/ng84h/have_you_or_anyo...
359.
The SOPA secret: why aren't major Internet companies doing more?
22 points by glimcat 15y ago|6 comments
360.
by glimcat 15y ago
Try thousands of years. Here's one example from about AD 100, Roman, on a wood-and-wax tablet. http://www.abc.net.au/science/articles/2001/03/05/253805.htm

More ›