I love DD-WRT and have used it for years, but I get the impression it's a fragile project. The bulk of the work seems to rest on the shoulders of one or two people who only have so much time. I have always preferred Netgear's hardware with DD-WRT on top of it, but Netgear's latest product line (which has a TON of different router models ... way too many, IMO) has only partial support from the DD-WRT project. Netgear's fanciest two routers, the R7500 and R8000, aren't yet supported. All we can do is sit and beg Brainslayer or Kong to spend time on them, but they've got a lot of irons in the fire.

I really wish Netgear would just give up on Genie and pay DD-WRT to support development and license it as their official firmware. Rebrand it or something if you want, but give us the power of a real firmware. I've used Genie lately on the R6100 and found quite frustrating for anything fancier than a typical home wifi router use case. Security bugs like this only prove that they're failing to get it right on their own.

It makes sense that Cisco doesn't want their Linksys-branded routers to be too powerful, since it might hurt sales of fancier Cisco stuff, but what's Netgear's excuse?

To solve this latency problem, most well-designed websites use a single large GIF or PNG for all their tiny CSS images, then slice the image to indicate each independent icon, border, etc. This not only reduces the total image overhead but also greatly reduces the total number of 304s to receive.

Example: one of Facebook's icon resource files

A similar rule applies to amateur rally races in Finland, covered on the BBC show Top Gear. In folk racing, every car is given a nominal and equal value, such as €1000. At the end of a race, if anyone asks to buy your car then you have you sell it to them, which keeps anyone from putting too much into a car.

Android Market apps are mostly super cheap. Who can't afford $1 on a game they'll play for a few days non-stop? Or a few bucks on a ROM management app? Prices for most paid apps are so low that I imagine that the largest barrier to entry is not price, but the effort required to set up one or more credit cards. My hypothesis, for that reason, is that a large portion of the piracy comes from the age 15-20 crowd who have fancy phones and lots of free time to figure out piracy options, but no credit card(s).

Google can greatly reduce this kind of piracy by working out pricing deals with the carriers to allow charges to appear on phone bills. How else would the ringtone industry thrive as it has? Verizon certainly doesn't offer a direct-bill Android Market option. Maybe this is already the case on other carriers? How does piracy compare in those cases?

Another annoyance of the Market is currency conversion. I've bought apps for sale in both Yen and Euros, and for those purchases I had to set up a Visa card since my AMEX didn't support foreign purchases (on the Market, at least). Most users don't want to deal with that kind of crap ... again, piracy is easier. Can't Google Checkout handle currency conversion on the developer's end without hassling end-users?

This is true until they release a first-party game with the update included as a requirement. For example, Super Mario Galaxy 2 includes the 4.2 system update and requires you to update your system before you can play the game ... unless you start the game with a homebrew tool that blocks the update. Funny how the very feature they're trying to remove is capable of blocking the removal.

Homebrew users know to avoid system updates at all costs, so the only people affected by them are people who have not yet hacked their Wii. Once updated, though, those people will have a harder time installing homebrew should they choose to try it.

I'll keep the suggestions flowing: I love last.fm for its ability to help me find (and listen to) new music. It even offers a dedicated client that doesn't require an open browser window all day.

MIT's strategy is very interesting. Several groups (like our team) have been forming their teams for weeks, but MIT appeared on the scene just today, and it's fascinating that they got a front-page Slashdot plug. I give them lots of credit for flooding the scene with mentions in such a short time. Whereas some teams give their winnings to charity (like ours), others entice balloon spotters with cash portions of the earnings, and MIT has decided to do a little of both.

DARPA is the sole decider of how difficult this competition will be. Will they place the balloons in dense urban areas, or will the launch them in small rural communities?

Best of luck to all the teams tomorrow, MIT included. I hope that the contest winner will write a paper describing their strategy, both in network-building and in launch-day data collection.

This story is all well and interesting, but I imagine it's impossible to prove that you're the first business to make money exclusively over the internet. You might be "one of the first", but to go all the way to the birth of my business represents the birth of dot-com is a bit vain, no?

These are the same people that probably sugar coat the truth and then eat it.

Just make sure it's not that peppermint-flavored dog poop.

This comment appears regularly on /. articles that use the term "brick." May I suggest that the term "brick" is slang and has no official definition? If I plug a [poorly patched] HD into a computer and get no sign of life, I'd consider that a "brick" until it's been flashed back into proper function.

"If you can reflash it" is also subjective: does that mean via a normal IDE/SATA interface, or does it extend to a direct JTAG connection, or do you have to desolder the ROM to flash it? There's a broad spectrum of functionality, but it seems most useful to use the term "brick" to refer to any device that seems to have no useful function under normal circumstances. My point is that it's open to interpretation, so don't be so picky.