Well, I would go to Mexico and protest this law, but I can't because it's against their constitution for a foreigner to do anything of the sort (Article 33).
Human beings are naturally able to learn more than one language. I think every adult human being should be able to carry on a conversation in English, Spanish, Arabic and Chinese on any subject, and know at least one other language besides. Of course, I can't do this myself yet...
While it's true that the W3C policy could be abused so as to limit actual use of an algorithm in GPL code, it's still a big step in the right direction, and I doubt any organizations will be able to inject such pathological standards successfully.
The FSF's example of a Koqueror version that parses URLs and a server-side script that do the same thing seems a bit contrived. Wouldn't the server-side script be considered an implementation of the standard as well, even though it eventually acts as a client for another web browser?
This is on the Internet, not in the days of the Pony Express. It's understandable that some companies might sync their data-centers with their business-partners on a daily basis to take advantage of periods of light traffic, but anything less in negligent. I say 24 hours and 59 minutes out to be the time between when I opt out of a list and I get the last message related to it.
The other problem with a Biggest Possible Fixed Time is performance. Encryption already causes overhead; maximizing overhead on remote network connections is not generally viewed as a Good Thing (TM).
Look at Figure 4 in the paper carefully; even after compiling with -O0, the RSA library takes no more than 2x10^7 CPU cycles, which works out to 8.3 ms wall-clock time at the processor-speed they mentioned. Some random no-name drive has an 8.9 ms seek-time. Besides, this isn't necessarily a performance-hit; the system could just do a usleep(100000-usecs_taken) and gain the same security-benefit as log as the attacker can't do a ps against the system.
I like writing screenscrapers and running 'ping' with the rest of them, but I'd probably say AA is in the right on this one. I get the idea reading the brief that they played fair for a while and had a fun old time trying to outsmart the outsmarter, but finally gave up and called their lawyer when real harm was done. Hey, I've bought tickets from AA and had to struggle with their slow servers; could I sue Farechaser for unjust enrichment against me, an AA customer?
Incidentally, I did a search on Lexis-Nexis for "screen AND scraper OR automatic AND database". Notes from one interesting hit I came up with are on my website, but I didn't find any case law on the subject at hand.
Slashdot Mirror
Well, I would go to Mexico and protest this law, but I can't because it's against their constitution for a foreigner to do anything of the sort (Article 33).
Human beings are naturally able to learn more than one language. I think every adult human being should be able to carry on a conversation in English, Spanish, Arabic and Chinese on any subject, and know at least one other language besides. Of course, I can't do this myself yet...
The FSF's example of a Koqueror version that parses URLs and a server-side script that do the same thing seems a bit contrived. Wouldn't the server-side script be considered an implementation of the standard as well, even though it eventually acts as a client for another web browser?
MySQL 3.something+ also supports row-level locks in MySQL tables.
I hope the MD5, SHA, etc. functions are default now... they seem to be absent from the 3.23.55 build that comes with the Debian distribution.
This is on the Internet, not in the days of the Pony Express. It's understandable that some companies might sync their data-centers with their business-partners on a daily basis to take advantage of periods of light traffic, but anything less in negligent. I say 24 hours and 59 minutes out to be the time between when I opt out of a list and I get the last message related to it.
Look at Figure 4 in the paper carefully; even after compiling with -O0, the RSA library takes no more than 2x10^7 CPU cycles, which works out to 8.3 ms wall-clock time at the processor-speed they mentioned. Some random no-name drive has an 8.9 ms seek-time. Besides, this isn't necessarily a performance-hit; the system could just do a usleep(100000-usecs_taken) and gain the same security-benefit as log as the attacker can't do a ps against the system.
Incidentally, I did a search on Lexis-Nexis for "screen AND scraper OR automatic AND database". Notes from one interesting hit I came up with are on my website, but I didn't find any case law on the subject at hand.