The Real Science Gap basically makes the same point - the jobs are horrible for scientists, so lots of smart people avoid the field.

I *like* to help people. Providing password-less wireless access is a nice way to help others. I don't do it at the moment, but only because of time pressures; I hope to do this in the future. It'd be best if there was a common convention that "no password means anyone can use" because there's no other way to make that obvious. In the meantime, I suggest using "public" somewhere in the network name, so that people will know that you're intentionally providing a service to others. Bruce Schneier has similar comments.

Actually, to be patent independent does NOT "require significant differences in their implementation". They just need to avoid or invalidate the patent claims, which are often really narrow. For more information, see Andrew Tridgell on Patent Defence. Which is why the statement that "VP8 is similar to H264" can be both true and a non-problem.

You jest, but it really is hard right now. Not because calculations are hard (they're obviously easy), but because the laws of every state/county/city/etc. categorize stuff differently, and the tax amount depends on the category. You need an across-the-country standard of tax categories, so that for each product you can correctly categorize it (and then figure out the tax). Quoting the article, there are "7,500 different taxing jurisdictions in the United States, each with a set of very precise rules describing what can and can't be taxed and at what rate. That makes it challenging terrain for retailers to navigate. In New Jersey, for instance, bottled water and cookies are exempt from sales tax, but bottled soda and candy are taxable. In Rhode Island, buying a mink handbag is taxed, but a mink fur coat is not". If there was a standard set of categories across the country, then it'd be easy. So if taxing jurisdictions want to collect the dollars, they're going to have to work out (AND AGREE ON) a standard.

Wow, what an inspiring story! Someone who worked to help others, to the very end. My congrats to him, and my condolences to his family.

Can no one look up and confirm well-known facts? Heck, this stuff is still within living memory. The article claims that EDSAC was the "first working stored-program computer" and that is just wrong.

The Manchester Small-Scale Experimental Machine [often known as "Baby"] was the first stored-program computer, not EDSAC. Baby was operational on June 1948; EDSAC didn't run anything until May 1949. Please don't play semantics with the word "working"; Baby worked, and in any case, all of these early computers were wimpy if you measure by storage or speed. EDSAC is important in computer history - don't take anything away from THAT - but let's get the facts right.

... in the air, anyway. As a technique to "engage the public in the science of environmental pollution" this might even help.

Once again, it's clear that fuzzing is really useful for testing security. Not that it's a be-all/end-all, but people developing secure software should be using fuzzers. It's unfortunate that this fuzzer's "design can make it unexpectedly difficult to get clean, deterministic repro"; without deterministic repros, it's often really hard to find and fix the problem.

Um, what? It's hard to estimate profit margins, but Daniel Eran Dilger estimates that Microsoft has a 66% profit margin on Office and 81% on Windows. That's far beyond typical profit margins, so such prices are not "rock bottom".

In theory, you can live without cable/internet/cell/phone, just as you can live without roads. But unless you already have a lot of farmable land (think Amish), you cannot realistically survive. If you wish to have most jobs, or start a business, you need to be able to communicate. Internet is no longer a luxury for many.

In most cases realistically useful Internet access is only provided by monopolies or duopolies. Regulation should be limited, but in the case of monopolies, they are often necessary. In this case, it's necessary.

The "trusting trust" attack is a nasty attack, but there is a counter-measure. Diverse double-compiling can detect compiler executables subverted by the "trusting trust" attack. See my paper for more, if you're curious.

I'm posting this, from a Linux desktop. It doesn't look dead to me.

You're talking about the trusting trust attack, which was made famous by Ken Thompson.

Thankfully, you can counter the "trusting trust" attack using a technique called "Diverse Double-Compiling" (DDC). See the linked PhD dissertation for details.

Actually, the U.S. administration has already admitted that the current export control system is messed up. In April 2010 U.S. Defense Secretary Robert Gates called for a major overhaul of America’s export control regime, saying the current system is outdated, hurts America’s competitiveness, and does not adequately protect national security. Of course, admitting there's a problem is not the same as making a change that solves it (or makes it better), but at least they know there are problems and are trying to find solutions. I particularly like this part: "One major culprit is an overly broad definition of what should be subject to export classification and control. The real-world effect is to make it more difficult to focus on those items and technologies that truly need to stay in this country. Frederick the Great’s famous maxim that “he who defends everything defends nothing” certainly applies to export control."

If they do the real job effectively, and don't cost too much more, they should do it. In fact, I'd like to see these worldwide. If human-shaped ones don't have enough legs, then animal-shaped ones might be good alternative (dinosaurs? dogs? dragons?).

Today's pylons do the job, but let's face it, they're ugly. If we have to dot our landscapes with pylons, we should at least make them interesting.