I'd like to provide public access, but I don't want trolls and other idiots getting my IP banned everywhere or criminally investigated. What I'd like to see is some kind of VPN-only / proxy-only access to the Internet. The idea is that I'm giving you access but not identity.

You'd be required to proxy through either your own server (ssh/openvpn), the Tor network, or some kind of commercial VPN/proxy service. I mean, you ought to be doing that anyway. All common ports, *especially* http/https, would be blocked.

That doesn't stop someone from ssh'ing into their hijacked zombie computer in Russia and using that to launch an attack, which could still lead to a criminal investigation if they didn't cover their tracks properly, but at least it'll hopefully stop the sysadmins and bots who assume "IP address == person responsible" from reflexively laying down the banhammer on my IP or suing me for allegedly sharing The_Hobbit_An_Unexpected_Journey_4K_xvid_LEAKED_plus_soundtrack.rar

Inexact matches. In the puzzle described in the paper, the pieces are all square (no notches). So the algorithm has to decide which edge matches best based on the similarity of the pixels, but it could be wrong or there may be multiple pieces that look like they match equally well (e.g. sky pieces which look very similar).

In the cases where it's wrong, it may have to throw out some of the fittings -- e.g. if you have a bunch of smaller groups of tiles that seem to fit together, but when you put it all together the puzzle isn't rectangular, then you have to break up the groups and try again. You can't just match one tile and be done with it.

That's war, not terrorism. Two different things, both horrific. WWII killed a lot more than a couple hundred thousand people.

It's a nice idea in principle but I suspect it'd have some nasty side effects.

For one, people aren't entirely rational (they want lower taxes but more government services). As much as I distrust Congress they at least have to discuss it before doing something monumentally stupid with the budget, and there's a certain amount of inertia against screwing with funding critical services.

California, where every other year there's some spending initiative on the ballot, is a good example of why letting people directly allocate parts of the budget is a bad idea.

Secondly, it'd just encourage every agency to spend a bunch of money on advertising. NASA already does it to encourage interest in space, and thus, their own funding.

Google TV currently runs on Atom (x86) hardware. I'm guessing the main issues for broader x86 support are drivers and performance.

IMHO, sooner or later we're going to need a high speed rail, but it might be better to wait until "later". Right now it's just not cost effective because not enough people would ride it.

I don't think it would be so bad if we waited 15-20 years to start building it. By then, fuel costs and congestion should be bad enough that people will be begging for it and investors will be lining up to finance it.

Also, a *significant* part of the cost of building it right now is that we don't have enough money to finish it quickly. If the state and residents were behind this project 100%, it could have been completed in 10-15 years (or less) for closer to the original projected budget. It's not rocket science. The biggest problems are political.

As much as I'd like to be able to take the high-speed rail in California in the next 20 years (I first heard about putting HSR on the ballot back in 2002), it's just too expensive if we drag it out without being able to fund it right now. California should just buy up all the necessary land and reserve it for future rail development so that it'll be ready when the time comes.

In the meantime, I'm crossing my fingers for the BART San Jose extension within the decade :)

95% of county/city decisions are over boring administrative stuff that usually only applies to one neighborhood (or street ... or sign) and gets zero media coverage. You have to either attend a town hall meeting or skim through the minutes to see if anything interesting in your area is going on, which most people don't bother to do.

In practice, there's rarely anything at the city/county level that really bothers me. The controversial decisions usually happen at the state/federal level, especially because the federal government has so much power compared to state and local government. Maybe if the power was inverted more people would pay attention to local government.

Due to an overall lack of information, and having to know about an entire county board instead of just one or two federal candidates with established reputations, it's also hard to make an informed vote. I mean, WTF do I know about the local county judges? They could be the worst judge in county history (or the best) and I wouldn't know.

There's a 30% federal tax credit until 2016. California used to have a bigger rebate (originally $2.50/kW, now $0.35/kW) but solar panels are also cheaper now.

I would agree. It might be that a very poorly considered choice of course content was made, and the intro course replaced between years (so the previous intro course made sense as a second course), but really not something that should happen.

Looking back at the old course webpages, it looks like the professor normally taught that lesson plan for his ethics class. He taught the intro course exactly *once* (maybe due to a staff shortage or last minute switchup) but didn't come up with new material.

Off-topic, but regarding self-plagiarism and "duplicate credit":

In college I took an Intro to Philosophy 1 class for my humanities general ed requirement. The next year, I signed up for the second class in the series, on ethics, which had TBD listed for the instructor. On the first day of class, I found out it was taught by the same professor -- and the syllabus was exactly the same as Philosophy 1!

The professor had basically plagiarized his own material for what was supposed to be a different course. For a class on ethics. Seriously. I confirmed with the TA (same TA) that the material was exactly the same and dropped the class (I took a different one later). In retrospect, I probably should have complained to the Dean.

Modern pirates usually go around in speed boats with small arms. I'm pretty sure an island run by a bunch of libertarians would have more than enough firepower to take them out, and no qualms about killing in "self-defense". I suppose they could try to kidnap people off fishing vessels further from the island, but wouldn't put my money on the pirates.

Also it sounds like they're going to build these things near US waters, far away from the typical pirate-infested seas.

A long time ago I wrote a GAIM (Pidgin) plugin as a prank based on the observation that people usually try to avoid responding at the same time by watching the typing icon; if one person is typing, the other person stops typing to give the first person a chance to finish.

The plugin, instead of reporting my typing status, would *mirror* the other person's typing status. So as soon as they started typing, it would indicate that *I* was typing as well. Most of my friends didn't realize what was going on; they would start typing and then stop typing as soon as they saw me typing, at which point it would look like I stopped typing as well; rinse and repeat. Eventually they would just ignore my typing status (which I normally have turned off, anyway).

I think maybe if some kind of financial liability was introduced, companies would take notice. Say, $50 for lost personal details (name, address), $100 per lost cc number, $5000 per lost SSN.

Smaller companies would have to use payment processor companies with better security. Larger companies and payment processors would have an incentive to not just follow best practices and minimum compliance, but actively conduct audits to reduce risk. Insurance companies would also insist on good security, in theory.

Of course, there's certainly downsides. Companies have shown time and time again that short-term profits overrule sensible decision making, even if the costs of risky behavior could bankrupt the company. On the other hand, I'm not a fan of government-mandated compliance if it's not kept to date with the latest technology and practices; time spent on compliance and paperwork could be better spent actually improving security instead of just talking about it.

And insurance companies -- not sure if they're a necessary evil, or just plain evil, and it's also not unheard of for insurance companies to go bankrupt failing to consider the risks.

But in principle, adding some kind of responsibility for losses would probably improve security overall.

Jumping the shark has jumped the shark.

Qt is actually LGPL now. Furthermore, if Nokia decides to stop developing Qt, the KDE Free Qt Foundation can vote to release Qt under a BSD license.