I'd just like to remind everyone that putting an object into a low-earth orbit requires about 25 times the energy of just raising it vertically to that height and letting it fall back to earth. That's why the commercial rockets that put satellites into orbit will continue to be big expensive beasts, X-prize or no X-Prize.
Hidden away in their financial report, they mention that in the last quarter they sold $20,000 worth of SCOSource licenses. Don't spend it all at once, guys!
I will be the first to applaud Larry Wall and the Perl developers if they can clean up regex. However, I feel tinkering with a notoriously complex thing may make it worse
What Larry has done with regexes for Perl6 is a work of pure genius. At the base level you can use them in much the same way as in the olden days (with a few minor changes in puctuation), ie same old line-noise.
But it now allows you to do something very clever.
You can give names to individual regexes: these are called rules; you can reference rules within other rules with angle brackets; rules can be grouped together into a grammar; grammars and rules are analogous to classes and methods; and yes, you can do inheritance with them. So someone writes a grammar for say parsing apache log lines and sticks it in a file somewhere; you can import that grammar, override the rule that recognises dates, and have a new grammar that recognises french day names say. Here's a simple example of a grammar:
grammar log_entry { rule month {jan|feb|mar|....|dec}; rule year {\d{4}} rule date {\d+-<month>-<year>} rule ip {\d+\.\d+\.\d+\.\d+} rule line { ^<date> <ip>.... } }
Of course theres' a lot more to it than just that.
See Larry's
apocalyse 5 for more details.
IANAL but I do read Groklaw, and from what I understand copyright restricts the act of copying (duplicating). You can study someone's implimentation of something as much as you like, then go impliment something similiar yourself. As long as you do not copy the code verbatim you are not in violation of copyright law.
Yes, but there are laws other than copyright laws,
such a trade secret and patent stuff. Also, you will have broken copyright laws just by downloading and peeking at the MS code. If MS can later claim that the peeking helped you develop some OS code, then that can help a big effect on the severity of the prosecution of you.
IANAL yada yada.
It gives further credibility to the perception that the "GPL is viral, contaminating everything it touches". Although this view is moot when one realizes that they shouldn't be copying and distributing other people's copyrighted works in the first place without permission (which in the case of the GPL is fairly easy to obtain), it's pretty much a given that SCO's own rendering of this code as GPL was inadvertent will be brought up as grounds for why the GPL should be considered legally invalid.
The whole point about the groklaw article is that is shows that SCO willingly released errno.h et al under the GPL. They were actively involved in a project to allow SCO binaries to run under Linux, and specifically released some "lite" versions of some of their shared libraries to facilitate this. This release was done under the GPL, and the RPMs were signed by SCO. The release includes errno.h etc, and did not contain copyright attributions.
Groklaw has just posted an email from Linus where he shows how he wrote errno.h and ctype.h for the original 0.01 release of Linux.
So it's not from SCO and it's not even from BSD.
Why? Glue records. You are _meant_ to receive certain As from the parent servers of a domain delegated to nameservers which live within its own namespace.
But glue records are very specific, and can be easily checked for. Only if an A record matches one of the names in the NS records need it be kept.
I assume the patch will filter requests, which resolve to the site-finder IP, so what's to stop VeriSign simply changing IPs every so often?
No, the patch doesn't do filtering in that sense.
It just allows you to mark some zones in your BIND config file (such as.com and.net), that should only contain delegation information. So basically if your BIND server recieves back A record(s) rather than NS delegation records from a server authoritative for.com , BIND simply ignores it.
Simple and elegant, and nothing Verislime can do about it. (I hope.)
Suppose I worked for a local council.
Suppose I suspected a council officer of corruption.
Suppose I tipped off a journalist from my home phone or email account.
That council officer can now obtain a complete record of everyone I've phoned or emailed in the last year, plus the fact that I recently visited www.howtoreportcorruptcouncilofficials.co.uk.
We must be careful here to distinguish between the generic malloc code, and the more recent SMP-specific stuff. The basic malloc code has been around, as many people have pointed out, since the early 70's, and it's fair to assume that that particular piece of code has been cut+pasted to death, since every time any UNIX vendor needed a new pool allocator, there was one sitting there ready for reuse.
However, what is more interesting is the more recent additions, ie the mutex stuff needed for SMP etc. If the SCO red highlighting is to be
believed, then both sets of code share the mutex_spinlock() bits. Since these are much more modern additions, the real question is where did those additions come from, and who copied/stole it from whom?
Slashdot Mirror
Hidden away in their financial report, they mention that in the last quarter they sold $20,000 worth of SCOSource licenses. Don't spend it all at once, guys!
What Larry has done with regexes for Perl6 is a work of pure genius. At the base level you can use them in much the same way as in the olden days (with a few minor changes in puctuation), ie same old line-noise. But it now allows you to do something very clever. You can give names to individual regexes: these are called rules; you can reference rules within other rules with angle brackets; rules can be grouped together into a grammar; grammars and rules are analogous to classes and methods; and yes, you can do inheritance with them. So someone writes a grammar for say parsing apache log lines and sticks it in a file somewhere; you can import that grammar, override the rule that recognises dates, and have a new grammar that recognises french day names say. Here's a simple example of a grammar:
Of course theres' a lot more to it than just that. See Larry's apocalyse 5 for more details.
Yes, but there are laws other than copyright laws, such a trade secret and patent stuff. Also, you will have broken copyright laws just by downloading and peeking at the MS code. If MS can later claim that the peeking helped you develop some OS code, then that can help a big effect on the severity of the prosecution of you.
IANAL yada yada.
The whole point about the groklaw article is that is shows that SCO willingly released errno.h et al under the GPL. They were actively involved in a project to allow SCO binaries to run under Linux, and specifically released some "lite" versions of some of their shared libraries to facilitate this. This release was done under the GPL, and the RPMs were signed by SCO. The release includes errno.h etc, and did not contain copyright attributions.
Can SCO really be that incompetent?
So, after 9 months during which the alleged Weapons of Mass IP Infringment failed to materialize, they finally got him. I hope they string him up.
Just out of curiosity, why *was* Darl in Tikrit?
The Crack Smokers Association of America sues SCO for bringing their name into disrepute.
...the Pope converts to catholicism and a bear performs an easement in the woods.
Why? Glue records. You are _meant_ to receive certain As from the parent servers of a domain delegated to nameservers which live within its own namespace.
But glue records are very specific, and can be easily checked for. Only if an A record matches one of the names in the NS records need it be kept.
No, the patch doesn't do filtering in that sense. It just allows you to mark some zones in your BIND config file (such as .com and .net), that should only contain delegation information. So basically if your BIND server recieves back A record(s) rather than NS delegation records from a server authoritative for .com , BIND simply ignores it.
Simple and elegant, and nothing Verislime can do about it. (I hope.)
Suppose I suspected a council officer of corruption.
Suppose I tipped off a journalist from my home phone or email account.
That council officer can now obtain a complete record of everyone I've phoned or emailed in the last year, plus the fact that I recently visited www.howtoreportcorruptcouncilofficials.co.uk.
This is scary.
We must be careful here to distinguish between the generic malloc code, and the more recent SMP-specific stuff. The basic malloc code has been around, as many people have pointed out, since the early 70's, and it's fair to assume that that particular piece of code has been cut+pasted to death, since every time any UNIX vendor needed a new pool allocator, there was one sitting there ready for reuse. However, what is more interesting is the more recent additions, ie the mutex stuff needed for SMP etc. If the SCO red highlighting is to be believed, then both sets of code share the mutex_spinlock() bits. Since these are much more modern additions, the real question is where did those additions come from, and who copied/stole it from whom?
In the UK, any under-age representation is illegal, even if it's been digitally maniplated from legal sources.
Ie the onus is on you not to have anything dodgy, rather than for the court to prove its dodgy