Slashdot Mirror

http://members.optusnet.com.au/pennywyatt/Interests/FlandersSwann/Other/Other02.html

(and these guys don't seem to do ad-generated revenue, so my link shouldn't put them on the "top 50" list :)

Here's a simple issue: Some people drive their street vehicles where US traffic laws do not apply. How do you allow for that behavior while enforcing the need for a governor?

What about when your speedo dies because your car is old and now your governor can't tell how fast you're going? Should the car be unlimited or should it no longer work? If you have no limit now you're incentivizing people to have cars where they can't determine the speed they're going easily. If it's non-functional you now how to get your car towed to the shop at additional cost.

Not to mention the fact that adding one more interlock adds one more point of failure. What if that failure occurs while on the highway?

Also, this does nothing to stop someone from going 70 in a 20 MPH school zone. According to this report: http://members.optusnet.com.au/carsafety/paine_impact_speeds_jan07.pdf half of fatal car accidents (front-impact, with seatbelts) occur at less than 30 MPH. Only 10% occur over 45 MPH.

So maybe just limiting cars to the maximum speed limit won't really help all that much. But it'll surely add in some new points of failure.

Worsening water crisis? Water is a closed loop system. You don't "loose" water.

And in contradiction to yourself, trees are actually responsible for helping create water. Ever seen a desert with trees? Nope...

Trees, and vegetation create part of a water cycle where they will store and release water thus creating a moist climate. When you have no trees or vegetation then water has no cycle. You then get the desert torrential rains that come and go, but don't really help.

http://members.optusnet.com.au/benjamink/Water/TheWaterCycleWebQuest.htm

http://members.optusnet.com.au/~weezil0/douchebag.jpg

I think the reference was simply: white hat==good guy black hat==bad guy. See also the "Six Hats" method for thinking (but I'm not sure it applies in this case): http://members.optusnet.com.au/~charles57/Creative/Techniques/sixhats.htm

Yeah, this article is bullshit. When I read the headline, I was thinking more along the lines of a Mospeada style motorbike. This thing is nothing more than a glorified Segway.

in the 1980s. Nice glasses. I used to have a set just like them...

Well, it's like, how can I explain it, kind of like this.

the difference is obvious

I'm guessing that it is the 24-bit rather than the 192khz?

As Flanders and Swann said about much earlier technology:
Flanders: All the highest notes neither sharp nor flat,
Swann: The ear can't hear as high as that.
Flanders: Still, I ought to please any passing bat,
Swann: With my high fidelity.

For this reason, I've been using Con Kolivas' patches to replace the scheduler. http://members.optusnet.com.au/ckolivas/kernel/ - very helpful especially if you don't have the fastest computer around. Also seems to help a bit with I/O - if my hard drive is trashing for whatever reason, interactive stuff still remains reasonably responsive. Or at least it doesn't make my mouse cursor skip...

Even so, I'd prefer to have IO better scheduled - ionice doesn't really seem to work at least for me.

Not strictly related, I just like to plug my fave kernel variant;

Con Kolivas maintains a great kernel patchset containing one of his own patches, swap prefetch.

It basically goes that, on a Linux system, most of you RAM is going to be used up caching stuff or being used for running applications. If you're not doing anything very demanding or you're away from the computer, background processes are going to eat up RAM doing stuff so your running apps get swapped out. As soon as the system drops back to idle again, the previously swapped apps are then copied back to memory so when you get back to your machine it's as if nothing happened.

Not as advanced as Vista's pre-emptive (I think) superfetch stuff, but a start anyway.

P.S. can anyone confirm or deny whether previous versions of NT held stuff in RAM? I've always been a bot confused by the ubiquitous amounts of "free memory" that could be put to better use; is it just that task manager doesn't report that memory as being used?

About Dashboard, try this:
First put a few widgets on your dashboard. Then logout and login. Launch Terminal, extend its window all the way down so you can see a lot of processes, and launch top. Now invoke Dashboard, and move your widgets so you can see the terminal window. See those new 'DashboardC' processes that just popped up, eating your RAM (and sometimes your CPU). They weren't here before.
You can read in the Dashboard technology brief (http://images.apple.com/macosx/pdf/MacOSX_Dashboa rd_TB.pdf), page 5: 'Each widget runs as a separate UNIX process in its own environment, or "sandbox," and is restricted to the same privileges as the user running it'.
However if you really want to deactivate it, try Onyx, it's just a tickbox to mark or not in one of the panes.

As for Spotlight, don't misunderstand me, I also think it could use some beefing, I just objected to using it as a launcher (I'm not even planning to abandon QS when Leopard comes, even if Spotlight improves as a launcher as Apple announced). In fact I even removed the Apps from its index, so as to make it just a file search and nothing else.
May I suggest trying Spotlaser to help with your searches?

Have fun spotlighting!

That can be applied to sightings of many other things.

The problem is, how does one determine which patterns indicate something and which patterns are just convincing illusions?

Personally, I think they shouldn't have tested on a laptop with a 5400 RPM HDD.

I'd go with a system that spouts pretty fast dual-channel memory and a fast HDD and focus heavily on I/O tests.

Wasn't http://members.optusnet.com.au/ckolivas/interbench / designed with this in mind?

Or http://ck.kolivas.org/kernbench/

I'd rather see interbench scores than FPS.

Nice try, you cynical fucking asshole. That's not Hitler any more than this, this, or this is Hitler.

Nice try, you cynical fucking asshole. That's not Hitler any more than this, this, or this is Hitler.

Nice try, you cynical fucking asshole. That's not Hitler any more than this, this, or this is Hitler.

Awww, look at cute wittle Adolph. His mother and father loved him dearly.

Slashdot renders ok on mine ... http://members.optusnet.com.au/~defragged/images/I E7-Vista-Beta2.png [screen shot of this page taken in Vista Beta 2 with IE 7+ ]

I'm in total agreement with you on this. My desktop has 256MB swap and 1GB of RAM. The only time Linux has ever touched swap was when I was importing 8GB of Eudora psuedo-MBOX files into KMail (I forget the exact command I was using). At the time I had 2GB of swap (rule of thumb, right?), and the import process swallowed it all up. I didn't notice in time and the system became completely unresponsive; not even a Ctrl-Alt-Backspace worked. I couldn't ssh into the machine either. Eventually the process died, but it took a good 30 minutes.

My MythTV backend with 2GB of RAM touches maybe 5-10MB of it's 2GB swap at the worst of times, even with quite a few memory hungry apps (Opera, amarok, myth transcodes) running as long as the system is up and pre-emptive swapping turned on.

When a process goes nuts, I want it to be killed ASAP. Leaving it to fill up xGB of swap and bring the server to its knees for x minutes is silly. I admit that in machines with less RAM than mine swap is a great idea, but on modern machines the overhead for using virtual RAM seems much higher, and I get no obvious performance increase from it, even with 10k RPM drives. Since the 4GB barrier isn't a problem any more for server at work (64bit Debian, yay!), we just plumped for gobs of RAM instead of new servers every x years (since we don't do much that's CPU limited). Heck, I even have MySQL DB's mounted in tmpfs at home. I don't mean to sound harch, but swap just seems to be a way for those unable to stuff more RAM into their boxes to stretch their budgets a little further...

I use the ck patch series for my thinkpad kernel. From the FAQ of the page I have the following information, which seems to come from a kernel developer and was alway enough for me, also on productive servers out in the wild.

http://members.optusnet.com.au/ckolivas/kernel/
How much swapspace should I allocate?
Unfortunately there is not much good documentation on swapspace that is meaningful on today's hardware. As the discrepancy between ram speed and hard drive speed becomes greater each year, the old rule of 2*ramsize is just plain wrong. This was in the old days because the addr space was directly mapped into the first half of the double RAM sized swap, giving an easy translation formula; only the space >> RAM size was usuable as additional space.

My cable modem has some voodoo.
http://members.optusnet.com.au/a1291762/2006/03/li ght-goes-on-light-goes-off.html

http://members.optusnet.com.au/ckolivas/kernel/ I wonder how Con's patch would do in these tests. There are two patches, one for servers, and one for desktops. They make the box uber faster for its purpose, but it sucks ass for the oposite. Get the patch for the server, and the server is snappy. The patch isn't perfect, however. I use it for my desky. It's swift.

Judging from the tone of your post, I couldn't tell: you may have never heard of the cow-orker. It's a good read, if you've never seen it.

setting aisde the fact that your statement is ironically irrelevant, i do understand your question and will answer.

so now I have question, what do you prefer a car that will run from 0 to 60 in 3 sec but this car is only good for one run, after that you have to replace engine, or a car that will do from 0 to 60 in 4 seconds?

neither. the maintenance may be higher, but i much prefer a maglev train engine that can pull 100,000 metric tons to a car that can push maybe 0-60 in 4 seconds. might take longer to get to 60, but it sure pulls its weight.

let me explain: performance isn't my concern. performance under heavy load is. Mac OS X has severe performance problems with multithreading and multiple processes that bog down databases and processor-intensive applications. Linux (and more recent FreeBSD releases) doesn't have that problem, and maintains a high level of performance through a heavy load of intensive IPC and multithreading that would bring a Mac OS X system to its knees. my typical workload involves a number of multithreaded applications (including MySQL) and real-time or low-latency processes (like audio and video playback) and extremely resource-intensive unthreaded applications (games, for example). my Linux system handles that quite well with Con Kolivas' wonderful staircase scheduler, remaining usable and responsive even with system load averages as high as 30.0. audio applications don't suffer under the pressure of processor-intensive games, and MySQL performs quite well under typical usage patterns.

would i be able to do this under Mac OS X? doubtfully. is it harder to use? probably. is maintenance a larger problem? definitely - i won't deny that.

would i use Mac OS X on a server? you'd have to be utterly insane to think it's a good idea. running Mac OS X for a high-traffic web server makes about as much sense as running unpatched Windows on an unsecured network. it's stupid.

if you want to use Mac OS X, fine. that's your choice. i certainly won't use it. it doesn't meet my needs, and it certainly doesn't meet my performance expectations.

Is there some method of setting up some sort of qos-like bandwith limiter to the HD for niced processes? (or anything else?)

have you looked at Con Kolivas' patchset? it sets the CFQ I/O scheduler as default and implements a process scheduler tooled specifically toward responsiveness and interactivity.

i've been running it for some months and have found that even under severe I/O traffic, the system remains usable overall. (the only catch is that the CFQ scheduler is still feature-incomplete and doesn't schedule I/O traffic by niceness properly yet.) i've had a Java app (Eclipse and Azureus both) push my load average up past 30.0 because of I/O and the system still remained as usable and responsive as ever (even though anything that did I/O would halt painfully, including starting new apps).

it's worth a try in any case. -ck kernels just "feel" better, IMO, and there are a lot of other people who seem to agree.

(also, you can set your scheduler to CFQ or whatever whenever you please: echo cfq > /sys/block/hda/queue/elevator)

But what about the lyrics of The Gnu Song??

It's going nowhere.

http://www.members.optusnet.com.au/pennywyatt/Inte rests/Tripod/OpenSlather/Open%20Slather07.html

Good comments should explain these areas:

a) What you're doing. b) Why you're doing it. c) How you're doing it.

I wrote a rant on this topic called How (Not) to Write Comments at http://members.optusnet.com.au/clausen/rants/comme nts.html a few months ago for my students. Before giving my recommendations, I made some fun of the comments my students were writing, which I classified as mystic translations, autobiography, military, comic strip and haiku.

You may want to consider using idle CPU power for something that will help, then, such as Stanford's Folding@Home or one of the cancer distributed projects.

I tried running Folding@Home. For whatever reason, running it even in batch priority (with patches from Con Kolivas against Linux 2.6) makes the system feel slow (which, of course, means that there is a bug in the batch patch, since it's supposed to make the process idle-only). Furthermore, if you try to kill the process, it keeps spawning new ones.

For these reasons I've stopped running Folding, and won't take part in another grid computing project anytime soon.

Here's hoping this will hold out:
The Quicktime .torrent

Well, according to Tridge himself:

"I did write a tool that is interoperable with BitKeeper," Tridgell said in an interview. "I did not use BitKeeper at all in writing this tool and thus was never subject to the BitKeeper license. I developed the tool in a completely ethical and legal manner."

However, Con Kolivas maintains a patchset for desktop users which incorporates a fix that allows the nVidia drivers to work at his kernel patch page. If you don't want the other stuff and just the nVidia fix, you can find the patch split out, and instructions on which patches to apply in his announcement of his patchset release. Check out the -ck patch though, it has a lot of cool stuff.

(yay, I actually got a story submission in...hi mom!)

Gentoo has a few dozen different kernels at your disposal. There used to be a gaming-sources kernel that was based on ck-sources. The other popular kernel in portage for gaming is mm-sources by Andrew Morton (the guy Linus lets go hog-wild with the kernel).

Kernels not in Gentoo portage but compatible and designed for speed are:
nitro-sources (ck-sources + reiser4 + framebuffer + other stuff)
love-sources (community maintained kernel to optimize desktop performance and test "unstable" patches)
speedy-sources (love-sources w/ reiser4)

Oh, and looking at the forums nearly everyone uses an nVidia card. So far, nVidia plays nicer with x.org than ATI.

'The statistical risk of humans getting wiped out in the next 100 years due to a super volcano or asteroid or comet impact is 1 in 455. How does that relate? You're 10 times more likely to get wiped out by a civilization-ending event in the next 100 years than you are getting killed in a commercial airline crash.'

Um, as much as I hate to dis a national hero, I'm gonna call "bullshit" on this one...

According to the above quote, statistically speaking, we should be getting wiped out by a huge catastrophy every 45500 years, over the entire history of the planet...

Has the frequency of comets appearing and volcanoes erupting increased?

According to this article http://members.optusnet.com.au/mpaineau/paine_bioa stronomy02.pdf the last major mass extinction was 65M years ago...

Even if you accept that we've been really, really lucky, you've gotta see that statement as BS.

It's been proposed and implemented a dozen times or more for linux - Andrew Clausen and I implemented one years ago called 'upstart', Rusty Russell presented one at linuxconf.au-2001 in Sydney, and showed how to preroll the block cache to reduce disk time.

No doubt it was suggested back when sysV boot scripts were first invented.

Oops. Try this.

Cow Orker

Reminds me of some of the books I have at home. The 1981 edition of Tomorrow's Home has a fantastic description of how technology will change your life. Predictably it features dodgy Buck Rogers technology being used by Jason King lookalikes. It also shows someone using a five-button mouse - an interesting item that shows how the modern computer interface was embedded in modern culture at the time.

Does anyone know of a good ogg client for OS X?

Possibly using convertfs, but I have no idea if it works or not.

This page seems to have more info about it.

an alarm system which cracks windows/sunroof slightly in hot weather BUT senses things like rain ... or a criminal's fingers

Or a cat's head

Bob Bemer ASCII Art Tribute

Hats off to a truly great man.

I just thought I would post a brief message about supermount. If anyone wants to upgrade to 2.6.7 and still use supermount, I don't think vanilla kernels have it in there (yet, I'm sure it'll get in there sooner or later). I'm pretty sure the Mandrake and Gentoo kernels have support for it (gentoo-dev-sources do, anyway), but I just looked at gentoo-dev-sources and it is at version 2.6.5, dunno about Mandrake, but I'm sure it will take a few days for all the distros to catch up.

If you want to upgrade for security reasons, but you also want supermount in your kernel (as I do), this guy seems to have a patch for 2.6.7, which might come in handy if you don't want to wait for your distro to catch up. I am going to use this patch myself, but I cannot guarantee that it won't bone your system so to speak. The patch is not just supermount, it looks like it has some other stuff in it too, so decide for yourself!

Seeing as how I'm posting this, I may as well give a little background for those not "in the know". Supermount is a sort of filesystem, you mount your CD-ROM and floppy drives (or even USB sticks) with it, and it will automatically mount and unmount the media when you insert or remove it, kind of like on Windows. Personally, I think it is great, and it is hard to live without it now I have it.

You can learn more about it at the project website. Jeez, if it turns out the vanilla kernel does have supermount after all, I am going to look a right idiot... *presses Submit*

Well since in most cases programmers do not use madvise a solution colud be the "autoregulate swappines" patch by Con Kolivas.

So if that's a goal of ours, perhaps it's a goal of another race's. And perhaps they got a head start on us, and that large percentage of "dark matter" actually consists of Dyson Spheres which capture everything, so are "undetectable" by us. That's pretty scary, to think that we just lost that much playground, and will eventually have to deal with the bully--on his own terms perhaps.

I mentioned this a year ago or so, and someone pointed me in the direction of Matrioshka Brains, so I will include some links for that as well. And an excellent discussion.

I would add to the last part that the larger planets could be taken apart by space elevators as well. They'd just start with the upper atmosphere; then work their way down. All the time the mass is getting smaller, and the elevators are pulling mass out so they can make themselves bigger in order to reach deeper. I think it's workable, and appears to be the most efficient way to do it--get the mass all out into "orbit" first. Actually, when you're about halfway done you can then start shipping what you mine off to other locations, and taking that amount of mass out of the elevators as well since they won't need to counterbalance as the planet's now smaller. (I don't know what the mathematical "middle point" where you start dismantling the elevators actually is--it could be something other than 50%.)

We could have "planet splitter seeds" which we shoot off to other stars, and they start with a tiny, correctly-placed elevator and build more of them as fast as is physically possible; the seed would be smart enough to calculate all the masses and start with the most effective one that would lead to the earliest date at which the entire mass of the star system is being used for computation.

The only problem is if we encounter life. Will our machines just assimilate it? Are the ones out there programmed to preserve us? Have they already done so?

Oh that's funny, I must be imagining this, then.

Please, there's enough disinformation about Blender's interface in this article already.

Earth is closest to the sun (perihelion) when the northern hemisphere is experiencing winter, and furthest (aphelion) when the northern hemisphere is experiencing summer. (In 2004, perihelion was on Jan 4, and aphelion will be July 5. [source])

Earth's perihelion: 147,000,000 km = 8.17 light-minutes
Earth's aphelion: 152,000,000 km = 8.44 light-minutes [source]

Con Koliva's patchset for 2.4.23 contains XFS. GrSecurity has been included for quite a while, but is not yet included for 2.4.23. So it's even easier

That's news to me! I invite the slashdot crowd to see how much Australia's two competing cable providers charge for their respective services. Take note of the download allowances.

Optus
Telstra

How is that for good value?