Slashdot Mirror
Slashback: Eldred, Cruise, SOAP
And then we saw the sharks. a9db0 writes "Part II of Doc Searl's travelogue recounting his experiences on the Geek Cruise has been posted here by the fine folks over at the Linux Journal"
In an earlier report from Geek Cruise, Linus predicted 2.6 by June 2003. If you liked the list of features being considered for 2.6, you can thank puriots0 for "the list of what's been included in time for the feature freeze for Linux 2.6", as found at kernelnewbies.org.
Peel back your eyelids and let these images flood your brain. strredwolf writes "I think we had half the story when Cartoon Network said they were going to remove Zoids and G Gundam in their Toonami block. It was more like remove Zoids, move G Gundam to Midnight Run with GI Joe, put HeMan and Transformers on full weekdays, and double up on DB and DBZ. The website and broadcast prove it now. (This report was done while watching to Toonami live.)"
And Stalke writes "Recently, rumours about Stargate SG1 7th season included it both being renewed as well as speculation that it might be cancelled. MGM and Scifi put those rumours to rest today by officially announcing a 7th season. It will begin filming next year with a full 22 episodes ordered. No word about Daniel Jackson returning though :("
Cracking down on alien fraudsters. yep writes "Administrators of the alien-hunting distributed computing experiment SETI@home have announced they will crack down on cheats who rort statistics on computing power lent to the project. The announcement follows a united protest from the chief contributors. SETI@home director David Anderson announced SETI@home would do its best to investigate users returning suspiciously high amounts of work and delete their accounts if it uncovered solid evidence of cheating."
Sure they're not. tiltowait writes "The Hartford Courant article "The FBI Has Bugged Our Public Libraries" has been retracted (this was mentioned here - but the older article has been removed). Even if the retraction can be trusted, this doesn't change the fact that the FBI can still bug libraries as freely as the CIA can assasinate with impunity, or that more McCarthyism is on the way."
This story retracts the claims of bugging made in the previous one. Since the FBI has little incentive to tell the truth on this count, I don't see what incentive anyone has to believe their denial.
Cleaning up the future for SOAP. Makarand writes "A major hurdle in finalizing the SOAP 1.2 specification has been removed. Both Epicentric, a subsidiary of Vignette, and WebMethods, which makes integration software, had said in earlier statements that they may have patents that cover the technology used in the SOAP 1.2 specification which would have made SOAP 1.2 non royalty-free hindering approval by W3C. Epicentric has now amended its earlier statement saying they no longer believe they hold any such patents, and even if they did, they are interested in making them available on a royalty-free basis. WebMethods has made no comments yet."
If their site DOES get Slashdotted...
The 144KB (*ahem* KiB) PDF is also here.
RudeDude
Perl/Linux/PHP hacker
Because what MANY of the cheats are doing is to take an *almost* finished work unit, then distribute that to a whole bunch of machines to finish and subsequently report. Thus, they flood back a tremendous number of identical (and valid) work units and get credit for many more work units than could possibly be done with the actual hardware that they have.
These cheats aren't diluting the validity of the results, only getting credit for huge quantities of work units. (Though one could argue that they are disrupting things by chewing up bandwidth) Credit is one of the reasons that folks volunteer to do SETI@Home, so this could also cause people to loose interest and drop out if not corrected.
Why don't they use a standard principle of distributed systems: just send out the same work unit to multiple machines and teams, and use some cross-comparison scheme to detect anamolies? Work units that disagree with the majority are flagged as invalid.
That is PRECISELY what that they are said to be abusing. One machine completes 99% of a work unit, that unit is then passed on to a hundred other machines. They each complete the last 1% and all hand in correct units. The cross-check program verifies that they all agree and flags them all as valid, they all get credit.
The good news it that this does not currupt any of the results. The bad news is that the "work done" figures are hosed and that worthless data is burning up bandwidth and processor time on the central servers.
-
- - You can't take something off the Internet! That's like trying to take pee out of a swimming pool.
Please, there is no Supreme Court ruling re MS. Not even close. The case was settled, so if you want to blame someone, blames the Bush Administration Justice Department for the weak terms it sought, or more importantly the portions of the case it simply dropped after winning on them (e.g., tying).
But, anyway, NOT insightful.
Won't they just tap your inbound/outbound connections upstream? I agree running your own mail server does have it advantages, but security from your upstream ISP doesn't seem like one of them.
If you don't want to read the PDF version of Lessig's arguments, I have converted it to HTML and formatted it for your enjoyment. Here is the link.
(If you're like me, you find it highly annoying reading a PDF file of pure text, double-spaced with line numbers.)