cperciva

26 ms·

451.
by cperciva 2y ago
Huh, interesting. For many of the users I heard from, the end of December is the ideal time for doing upgrades, since they don't need to worry about system availability while people are away on their Christmas/New Years vacatio
452.
by cperciva 2y ago
I'm not sure that I see the issue with a December EoL; can you elaborate? There's always at least 3 months for upgrades -- it's not like we're doing a release on December 15 and saying "you need to upgrade by the e
453.
by cperciva 2y ago
Heh. The builds we do are called X.0-ALPHA#, X.Y-BETA#, X.Y-RC#, and X.Y-RELEASE. And I got used to capitalizing "BETA", I guess.
454.
by cperciva 2y ago
Fair point, but I'm optimistic we can make this work. Half of the "December" releases will be .0 releases and we have 6 months allocated for those so I'll make sure I allow plenty of slack in the schedule; some of them
455.
by cperciva 2y ago
The hope is that with more frequent releases the upgrade path between minor releases will be smoother -- often in the past we've had breakage due to developers saying "I really need to get this into the tree now because otherwise
456.
Change to FreeBSD release scheduling and support period(lists.freebsd.org)
31 points by cperciva 2y ago|15 comments
457.
by cperciva 2y ago
Post acquisition. Part of the agreement was that Quebecor was going to put a lot of money into building new infrastructure.
458.
by cperciva 2y ago
Right, that would have been my preferred solution. But for unfortunately that's not something the current government is inclined to consider.
459.
by cperciva 2y ago
Ok, that's unfortunate, but clearly a technical problem rather than a policy problem. What would you have them do, keep mailing new SIM cards when they have no idea why they're not working?
460.
by cperciva 2y ago
$10B isn't enough to set up a 5G network. And... I guess it's a good thing they didn't assume ZIRP would be around forever?
461.
by cperciva 2y ago
Shaw didn't have enough capital to continue operating, and foreign ownership restrictions prevented them from raising more. The government decided that Rogers buying Shaw was a better outcome than Shaw going out of business, which is
462.
by cperciva 2y ago
I don't see the problem. The R-squared is 0.01 for blue/red predicting individual votes, because both of the states in question are really just different shades of purple. The R-squared is 1.00 for predicting the total vote shar
463.
by cperciva 2y ago
Oops, I meant the smallest root greater than 1, not the smallest root greater than 0. (Off-by-one errors, they're not just for programmers!)
464.
by cperciva 2y ago
No, we measure angles in radians so that e^(ix) = cos x + i sin x.
465.
by cperciva 2y ago
e is natural because it's the smallest positive root of the equation e^(i pi) + 1 = 0 .
466.
by cperciva 2y ago
Correction: I was the security officer for FreeBSD... about a dozen years ago. I'm now the release engineering lead.
467.
by cperciva 2y ago
This is the sort of bug which pre-announcement coordination is designed for. Anyone who doesn't have patches ready was either forgotten (I've seen a few instances of "I thought you were going to tell them!") or isn&#x
468.
by cperciva 2y ago
Patch out for FreeBSD. Not clear if affected (it has only known to be exploitable with glibc, which we don't use) but best to be safe. https://www.freebsd.org/security/advisories/FreeBSD-SA-24:04...
469.
by cperciva 2y ago
Yes. I believe proving the strict bound is one of the exercises now.
470.
by cperciva 2y ago
<pedant>You never see "erratas", since "errata" is already the plural (of "erratum").</pedant>
471.
by cperciva 2y ago
Knuth judged that it wasn't an erratum, since the bound he included was correct and he never claimed it was optimal. :-/
472.
by cperciva 2y ago
Short answer is that FFTs are about as well behaved as anything can possibly be, because they're rotations in C^n. Explicit bound is in https://www.daemonology.net/papers/fft.pdf
473.
by cperciva 2y ago
The error bound cited in that article is wildly pessimistic. The latest edition of Knuth has the correct bound (because I gave it to him).
474.
by cperciva 2y ago
In a sense, red-black and 2-3 trees are just a special case of B-trees with small nodes. Smaller B-tree nodes have (by a constant factor) more expensive searching than larger nodes; but smaller nodes are more efficient for mutating operati
475.
by cperciva 2y ago
c17 -G to create shared objects Finally! It always seemed very strange to me that posix said that shared objects were a thing and provided a rtld API for using them, but never specified how to create them.
476.
FreeBSD 14.1-Release Announcement(lists.freebsd.org)
7 points by cperciva 2y ago|0 comments
477.
by cperciva 2y ago
How things have changed. 30 years ago we said "please get your pets tattooed" because ear tattoos were how veterinarians reunited lost animals with their humans. Now we use RFID tags and tattooing is frowned upon.
478.
by cperciva 2y ago
Related story for people who haven't heard about it: At one point CERN had problems which seemed to be related to the phase of the moon. After much head scratching, it turned out that the problem -- with an underground particle accele
479.
by cperciva 2y ago
If your similar unknown bug is on FreeBSD/EC2, I want to hear about it!
480.
by cperciva 2y ago
The fix was to teach the ENA driver that "set the MTU to the value it already has" should be a no-op. With that change, the interface didn't bounce.

More ›