jbert

49 ms·

361.
by jbert 17y ago
> perhaps you could put it into some non-volatile heat sink, but unless it's a really big heatsink, you wouldn't want to be anywhere near that thing when it went into resistor mode Specific heat capacity of water ~= 4(KJ per Kg per Kelv
362.
by jbert 17y ago
I've thought this for a while. Ideally, you want to have drives of different ages, so they're at different pts of the 'bathtub curve'. If you're mirroring, you should only ever mirror a "fresh, unproven" disk with an "old stalwart" disk. Do
363.
by jbert 17y ago
Presumably this is to some extent experimentally testable, if we can construct lab analogues of the conditions?
364.
by jbert 17y ago
Because the US internet won out over the UK JANET? http://en.wikipedia.org/wiki/JANET_NRS Which caused no end of problems in mail relay at the time.
365.
by jbert 17y ago
I wonder if there is value to be had in migrating users to increase social graph locality. The probability of (B fr C) is greater than the mean if (A fr B) and (A fr C). That's useful information. Off the top of my head, I wonder how an alg
366.
by jbert 17y ago
OK, you need 1-3 keypresses for English. Ah, of course , it's tree navigation. The number of keypresses required to enter the character is O(log(number-of-chars-in-alphabet)). So having lots of characters is a win for input speed. Thanks.
367.
by jbert 17y ago
> This is one of the reasons that texting is so darn popular in China, and email is not. Texting is fast I don't see why texting in Chinese should be faster than in English. Don't you need to push a similar number of bits through the nu
368.
by jbert 17y ago
When I used KDE, my ssh-to-production script used dcop to change the terminal background colour. (KDE + dcop is great technology. External access to APIs running in other processes. API discovery. Great stuff.) Whilst gnome-terminal claims
369.
by jbert 17y ago
What am I missing? all mysql calls have to go over a system call, so strace will see them (as write() or send() I guess). Why use ltrace for this job?
370.
by jbert 17y ago
Forking predates threads. It's also makes more sense as a system call, since a library writer can implement spawn() in terms of fork() and exec() but not vice versa. It's also still a good model when you want to run N copies of the same cod
371.
by jbert 17y ago
Thanks for the pointers. Looks like the linux/ubuntu equivalent may be 'pfmon', also not enabled in the stock kernel. I think this is the relevant project page: http://perfmon2.sourceforge.net/man/pfm_get_cycle_event.html
372.
by jbert 17y ago
Interesting stuff. Does anyone know easy way to see the instructions-per-cycle on a running (linux) server? Or is it necessary to instrument each application with hires timers and count instructions between the timer points?
373.
by jbert 17y ago
I imagine that it's useful as part of your thermoregulation if your climate is cold. Cold muscles don't work as well, so your reaction time from "sleep to fight" would be improved I guess. As calories have become more plentiful, the selecti
374.
by jbert 17y ago
I'm currently looking at something which falls under the 'fat node' category (with a few wrinkles/complications) at the moment, so this is of great interest, thanks.
375.
by jbert 17y ago
See the 'tux3' linux "phased tree" filesystem which was implemented (and then quashed by NetApp) some time ago (2001): http://www.mail-archive.com/tux3@tux3.org/msg00035.html I assume that all "snapshotting" filesystems (NetApp, ZFS) have
376.
by jbert 17y ago
The main thrust of the presentation is to lower the bar in people's minds to writing test code. Giving people 'permission' to write sloppy test code makes them more likely to write it. In the context of selling people on writing tests, it i
377.
by jbert 17y ago
Houses with power generation already have an indefinitely-large capacity battery on hand. It's called the grid. If you produce too much power, you push your extra into the grid. If you don't produce enough, you pull what you need from the g
378.
by jbert 17y ago
No causation is being suggested in the headline. Also, reciting this as a mantra ignores the fact that correlation correlates well with causation. (Since causation implies correlation). Basically, correlations need investigating and explana
379.
The Ham Sandwich Theorem(en.wikipedia.org)
6 points by jbert 17y ago|4 comments
380.
by jbert 17y ago
A price worth "sacrificing your good name" for? If people had higher standards, the cost of being unethical would be higher - hence fewer people would be unethical. A virtuous circle.
381.
by jbert 17y ago
> Why not take the short term ethical hit? Because it's unethical? [Edit: add a little more of my thinking. Ethics should count towards your reputation. A reputation for honesty or dishonesty should be important professionally and publi
382.
by jbert 17y ago
Presumably it's the difference between a text-mode filehandle and a binary-mode filehandle. So (untested, because I'm baking a cake) printing to fdopen(1, "wb") should just send bytes, allowing printf etc to work?
383.
by jbert 17y ago
> The only problem is political. And economic? Perhaps low-grade waste is the bigger issue?
384.
by jbert 17y ago
Seconded. That apache process is chewing cpu, is it spinning in an inf loop internally to userspace? No, strace shows it repeatedly doing backend conn/op/disconnect - aha! Another trick in this vein is/was to use 'pstack' to pull a C backtr
385.
by jbert 17y ago
> This is at least partially a reaction to the Linux community's tricks to get fonts. Some people in the Linux community wrote a program to automatically download and install Microsoft's TrueType Font Collection fonts after installation
386.
by jbert 17y ago
But just tracking gains them nothing, surely? They need to hold a seed themselves to get the bandwidth benefits? Or have I misunderstood what they're doing?
387.
by jbert 17y ago
Why do they need to modify the .torrent? If they simply read passing .torrent files and acted as a peer-then-seed, wouldn't they get the same benefit? Perhaps the problem with that is they want to firewall off non-customer IPs. Would this
388.
by jbert 18y ago
I think that comes under the heading of "nice problem to have". I guess a related question is "how/when do you decide to iterate - i.e. change your direction/customer focus/etc"
389.
by jbert 18y ago
If I read that search correctly, it's saying that the number of threads about mod_perl has declined. That symptom is also affected by: - increasing maturity of mod_perl (fewer problems) - a decrease in apache/apache2 modperl/mod_perl2 discu
390.
by jbert 18y ago
> You don't usually want to give out your email address as a form of general ID, But why is that? Is it because people will abuse it and send you data you don't want (spam)? The same problem applies to twitter ids, except that not many

More ›