That's an interesting issue, and one I didn't know about. Thanks. There's some misinformation in the article (the main posts), partially cleared up in the comments. It looks like a recurrence of the window scaling issue from a couple of years earlier. See http://lwn.net/Articles/92727/ for that one. In both cases this is about non-Linux systems not doing scaling correctly, and some philosophical points about working around other people's problems. Which, unfortunately, can be an issue with any community-developed OS.
OTOH, I've had problems with a TCP/IP stack on a commercial Unix, and Windows XP also had a window scaling issue. But none of this is an excuse for Red Hat not getting on the ball quickly for a client with a service contract, if they failed to do so. It looks as if there were options they could have taken, including adjusting tunables, which worked in some cases.
I'll have to look at the current implementation next week, on the theory that a problem that's recurred once might recur twice. Bummer. I did *not* need more stuff to do. But, seriously, thanks again for bringing this up.
OK, if that was the issue in the original AC post, and RH support couldn't deal with it, I'd say he had a valid complaint. You can deal with most issues (everything I've ever seen, though I haven't certainly seen everything) as either a kernel tunable, or on a per-interface basis. An OS support group should have been able to get that sorted.
Now I'm curious. Hopefully the AC, or someone else that's seen it, will post back.
In general terms, brevity in markup is desirable. Higher transmission speed, simpler (less bug-prone, resource-intensive etc.) creation and rendering tools, etc.
MathML will be the preferred solution for some things, just as Python is. OTOH, would you consider Python the language of choice for all problem domains? Nah, me either.
I've wanted good browser support for higher math for a *long* time. I'd hoped for something simpler than MathML. Alas, it was not to be. I'm not knocking the MathML folk. I trust them to have made things as simple as possible while dealing with a complex issue. Nonetheless, I'm a bit saddened by the verbosity of MathML.
Parent has a point. There could well be a large reduction in phishing sites--at least temporarily.
From TFA: "Naturally, in this sphere, as in other spheres, we should be thinking about adhering to Russian laws, about making sure that child pornography is not distributed, that financial crimes are not committed," he continued. "But that is a task for the law enforcement agencies. Total control and the work of the law enforcement agencies are two different things."
Right you are. Take semiconductor manufacturing as one example. In the mid-eighties, there was concern on the part of the feds about machines disappearing into China. Ion implanters, etc. were regarded as sensitive technology, as the military, the space program, etc., was recognized as being strongly dependent upon a technological edge.
There had been a minor stir at the company I was working for, as we'd sold an implanter into Asia, and it turned out that the company we'd sold it to didn't exist, and the machine was gone. Probably into China. Later, there was some recognition that Perkin-Elmer (who made advanced photolithography equipment--a vital semiconductor technology) was under threat.
That was also the beginning of the end. In '85, during the first of the semiconductor dumping (selling below cost, to win market share) fiascoes, I was laid off. I was pretty bummed, as auto industry layoffs were getting all the press, and at that time, US semiconductor manufacturing was a larger industry. But Detroit had the lobbiests.
Very little semiconductor manufacturing equipment is made in the US any longer. Perkin-Elmer exited the business a long time ago, and most equipment is made in China. Yet our military is more dependent than ever on chips, for everything from night-vision goggles to the fly by wire control systems that all modern military aircraft depend on.
If China were to decide to nationalize the industries that we've off-shored, we would be screwed on many fronts. Semiconductors, advanced machine tools, and much more.
This may have a good side, in that that countries that are deeply into each others pockets don't tend to go to war. But I would date the rise of the multinationals, to the point where they could overrule governmental policy, from about twenty years ago.
It's a complex subject, and that's just my opinion. Nor is it currently the end of the world, even from a military perspective. But my take is that we are gradually painting ourselves into a corner, and we need to regain control of corporations. Right now, they have no loyalty to governments, only short-term profits and CEO bonuses. There's no longer any conception of a corporation being a good citizen. There's no corporate morality whatsoever--and they have the lobby dollars.
I've actually started to notice when a cashier knows enough to count out my change correctly. Years ago, that was the standard of competence for a cashier, and I was annoyed when it wasn't met.
I've been seeing the same sort of thing in other areas, over a period of years. Others obviously have, too, including media writers and editors. I see small bits of evidence for that every day. An article mentions x MW (thermal) from a reactor, and the story is picked up by other publications, but modified to x MW of heat. That's a vote by the succeeding publications that 'thermal' may not be understood.
A 2001 NSF survey showed that about half of Americans didn't know that the earth revolved around the sun, and took a year to do so. But this stuff changes, and all hope is not lost. Though there is both good and bad news in NSF's Science and Engineering Indicators 2006 (the most recent available) Chapter 7: Science and Technology: Public Attitudes and Understanding at: http://www.nsf.gov/statistics/seind06/c7/c7s2.htm which is a very interesting read. There are spreadsheets in the appendices, etc. This is the best reference I know of.
OK, unless I get compelling reasons to opt for other choices, I'm leaning toward the Apple hardware. Though I have destroyed one Toughbook...
I'm rather a Unix fan (vastly too old to pass myself off as a boy of any sort) myself. I have to take a minor exception to claiming that any Unixy OS is the best available. It all depends upon what you need to *do* with it. I've called at least four variants 'best'. But that was about 'best for very specific purposes'. Sometimes it's come down to whether you can do non-blocking operations against a PRNG. So that experience may not apply for anyone needing a general purpose machine, though it has for me.
I haven't run into a case where I 'need Windows for high performance work'. For what I do, I've *almost* always found Windows at the low end of the performance scale. But that's just me, and I am a definite corner case. For instance, I rarely use an office suite, whether that be MS Office, OO, etc. Maybe once a day.
KDE apps meet my communications and calendar needs, as does KDE Kile (LATEX) for writing, as most of my writing involves math. KDE Kate can suck for text editing, as it supports only CVS, and I'm sensitive about sources, but there are easy workarounds. Most things I do that involve number-crunching and databases, I've written my own code to automate much of the drudgery away, where things are repetitive. I don't have to put up with spreadsheet charting limitations, in the sense that there are things you can't do with spreadsheet charting that are easy with http://www.graphviz.org/ and other specialized tools.
It will be interesting to see how my old-fart, laptops-are-lame attitude might change with nice hardware. Though some things that seem important to some people, I won't care about at all. DVD playback comes to mind. I have living-room electronics for that. If I get on a plane, I always have a couple of dozen papers to read. I never seem to get caught up on the journals.
Thanks. Of the three Wintel-type systems, I was leaning toward HP, if I had to go with one of The Big Three. It's hard to judge by my previous experience. For desktops and budget departmental (whitebox) servers, I build. When I've spec'ed rackable systems, I've had far better reliability results with HP than Dell. As a couple of people I occasionally contract with, and speak to (beers at the local) on a regular basis have had similar results, and they tend to think that Lenovo (whom I have no personal experience with) bites, I tend to listen to them. At least as far as ruling Lenovo out.
So far, and based partially upon your advice (thanks again) I'm leaning toward HP or Apple hardware. Possibly more toward Apple. It would be interesting to kick Leopard's tires. Perhaps a fat hard drive, and possibly quad booting CentOS, Fedora, Leopard, and still possibly OpenBSD. I usually opt for spending a bit more up front for an extra year or two of useful life. But somehow I've never taken that approach with laptops--perhaps because I'm so hard on them that I've come to regard them as disposable.
It will be interesting to see what results I get from spending a bit extra for nice hardware. One possibility is that if I know I've just spent top-dollar, I might actually take care of the bloody thing, vice tossing it onto the couch when I come home, etc.
"Apple's 'flashy 3d interface' brings style to an area of computing that desperately needed a breath of fresh air."
On one hand, that gives me the shudders. I don't want to even know that a backup is going on, save when I must unavoidably play media-monkey, or reevaluate the scheme because something has changed. Far better to design a reasonable strategy, test it thoroughly, and then have it be as unobtrusive as possible.
'Unobtrusive' and 'robust' are Good Things, as I am occasionally my own worst enemy. I once needed pretty much needed all my workstation resources while pulling an all-nighter, a daily rsync script launched, and I killed the job. A couple of hours later I shut down to swap some hardware, and the local drive didn't come back. I ended up losing several hours of work. Not good, particularly during an all-nighter. Now my daily rsync scripts look at machine loads, and are willing to sleep for a bit. At a certain point, though, they will run anyway, though bandwidth is throttled, proportional to load, as long as loads are above a threshold. The scripts are also sensitive to how I define the machine. DMZ hosts act somewhat differently than protected-LAN workstations.
My point is that I just want reliable (defined as knowing you *absolutely* can do a restore, given functional local hardware) backups to Magically Happen. I'm not familiar with Apple's product, but I've yet to see a solution that's as flexible as a set of well-crafted scripts, for networks of up to, say, a couple of dozen machines. I've used enterprise software that doesn't do load-sensing, though it did offer very useful features for dealing with networks of at least a few hundred machines, and a reasonable (not great) GUI.
I suspect, but cannot prove, that the issue for a large class of users (home and business) is an edu thing. If they aren't aware of the need, and the factors involved, those users probably aren't going to be helped much by a GUI. The solution, to my mind, is something involving a 'first boot' setup, such as what runs when you first launch a GUI after a fresh install of a Linux system. Establish a non-root account, etc. Note that at least some Linux distros don't do this if you're not installing a GUI, which I have a problem with.
This wouldn't be a trivial thing to do, as it would need to attempt some network discovery, to prompt for a possible network backup, etc. Detection of a local tape drive should definitely kick off a 'first-boot' backup configuration application app, or at least that should be selectable during OS install. Apple derives powerful advantages from controlling hardware, BIOS, and OS. In the case of backups, I very much doubt they're doing all they could. If they were, it would have been in the news as a major advance. Certainly nobody else is doing it.
"As it turns out, the MacBook is the Mac most people are buying. It is a competitive laptop to all but the bargain-basement craptops that Dell, Lenovo and HP sell. Get beyond the loss-leader "hacked by Chinese" craptops and you will find that MacBook is pretty damn competitive with the competition's lappies."
Query: IYO, are you suggesting that mid- and high-range laptops from Dell, Lenovo and HP are (or at least may be) reasonable values, or are you saying that their entire range is junk, and that you'd recommend another vendor for non-Apple laptops?
Reason I'm asking is that I'll probably be buying a new laptop (dual-booting CentOS and Fedora, possibly OpenBSD as well, though that wouldn't need a GUI) around Christmas, I've at least some reason for avoiding all of the above vendors, and it's not too soon to start asking around about others.
"...100% of 8 PS3s indefinitely is preferable, from what he says, to the costly little slices of "real" supercomputers he tried to rent before."
Actually, it's preferable to me as well. He was getting $5k in NSF grant money for those runs. Which I helped pay for. I've no problem with that, but this is obviously a better deal from my tax-paying perspective.
One of the things he's researching is whether or not gravity waves should actually be detectable. We've already spent large amounts of money searching for them, without success. Then there's LISA, http://lisa.jpl.nasa.gov/WHATIS/intro.html which should fly in 2015.
If he should find that they are undetectable, or that their detection would require technologies we don't yet have, and his results are corroborated, *serious* savings would ensue. Achieving those savings from a corporate equipment donation valued at $3200, and a single researchers' time, would be sweet.
"Marriage exists for one reason, and one reason only - Succession of property rights. Allowing humans and robots to marry would mean allowing robots to own land. No more, no less."
No. Succession of property rights has been entirely subverted. Google for 'marriage strike'. The odds of a marriage ending in divorce are greater than even. No fault divorce means that either party can obtain a divorce on dissatisfaction alone. The odds of a male gaining physical custody of children are something like 1:40. The odds of a male retaining a home are something like 1:3. If you even attempt to get along, doing things that women's magazines promote, such as as buying random gifts, you are acting against your best interests. She now has an 'accustomed lifestyle' argument if the marriage doesn't work out, and it's in her financial interest to divorce.
Bottom line is that on her whim alone, you can lose *all* rights. Property, control of your financial future, even the right to see your kids without being messed with. The percentage of women admitting to messing with guys on that score is 40%, and the reasons are about the usual primate suckage. Revenge for perceived slights, etc. There's a large skew in surveys, involving people not admitting to having done things that demonstrably suck, even when they're assured of anonymity. That 40% is probably very conservative.
Give me a Cherry 2000. I really don't need a potential (the odds favor this outcome) gold-digger to somehow 'make me complete'. I don't read Cosmo. I'll take teh mad robot sex over incomprehensible female issues any time.
Was cooking dinner, kicking back, cleaning up a bit, and just generally having a nice Sunday. Read some news--boom! Now I have to bookmark a bunch of things, do a lot of reading, make moral decisions, etc.
I resent that. Yeah, eternal vigilance is the price of freedom, and all that. But eternal is beginning to get very damned eternal. A few years ago, I was wondering why I might have to explain what a blow job was when the kids did what I encouraged them to do--watch the news.
Those were the simple days. Now, if you can get your kids to watch the news, you might be having discussions about torture.
This is just depressing as hell. I'm tempted to vote for Ron Paul, in hopes of at least rattling some cages. In the end, I'll vote for whoever is toughest on lobbyists. But no candidate can be tough enough, as far as I'm concerned. The Great Experiment is failing.
TFA: According to statistics provided by Market Share by Net Applications, starting in December 2006 and through September 2007, Linux doubled its market share. This detail would sound nothing short of promising, except for the fact that the doubling in market share is equivalent to a jump from 0.37% to 0.81%. In the past month, the open source operating system only increased its footprint on the market by 0.4%, from 0.77% to 0.81%.
YOU: Only increased by 0.4%?
Actually 0.81 - 0.77 = 0.04, not 0.4. Author of article can't do arithmetic, and you missed it.
Winds aloft are far more reliable, which is why wind generators have their blades higher than is required for ground clearance and common sense safety. Even a few meters of elevation is an efficiency win.
But all the press is about wind power generated almost at ground level. I'm wondering why aerial generation via aerostats isn't getting a few R&D dollars.
http://skywindpower.com/ww/index.htm is an interesting site. Only a dozen pages or so in total. It uses frames, and I more or less hate it, but the *content* is extremely interesting. To break out one frames page: http://skywindpower.com/ww/page002.htm "Capacity Factor" breaks out capacity factors for locations, at 15,000ft and 10k. There's also information on tethers, safety, etc.
Have a look. It would seem that wind could be more reliable than many people realize.
Spontaneity, in terms of brain function, probably isn't always a Good Thing. I tend to think of brain function the way I do backups. Reliable, or counterproductive. In my experience of the average meeting, you can burn a lot of cycles trying to sort out whether the idiot in $arbitrary_chair actually made any sense this time around...
I got the joke. But I've just gotten out of a horrible end-of-week meeting, so I was forced to write this by Higher Powers.
I'd agree with running on a different port cutting your log noise down quite a bit. The last time I ran the experiment (about 6 months ago, IIRC), I got better than an order of magnitude reduction in brute force attacks.
One point, though. If anyone is part of a corporate admin team, make sure you talk to the group about it! Hosts running on a non-standard port can disappear as known/allowed ssh hosts, to some management software. If they're picked up later in a network scan, IDS logs, etc., much hilarity may ensue. Murphy being alive and well, it will be the middle of the night before someone connects, the IDS lights up, and your phone starts ringing.
I guess it all depends upon whether you think factoring large numbers is a hard problem, whether special cases might exist, whether huge amounts of investment dollars matter, etc. From there you make your own call about whether or not to go all elliptical (another bag of worms) or not, etc. In the end, you either trust the math, or you don't. Not counting valid points you brought up about whether you can trust your hardware, compiler, or binary blobs.
One point you didn't bring up is rubber-hose cryptanalysis, which has a proven track record dating back through several centuries. It might be a lot easier for an adversary to ignore your opinions on math, the openness of your compiler, etc. and just beat the living hell out of you. Or just toss you in a cell for contempt of court until you either give up a passphrase, or grow old enough to win a sympathy argument.
Nothing is certain. First you evaluate the *perceived* value of the secrets you're trying to protect. Until you've done that, you can't estimate the potential intensity of the attacks that might be brought to bear in order to obtain those secrets. And only then can you think in terms of effective countermeasures. Assuming there are any, which may not be the case where, for example, an individual is squaring off against the resources of a governmental organization.
I don't know about an MIT ion drive test, but Dawn is scheduled for launch tomorrow. First non-experimental ion drive spacecraft, which is supposed to orbit Vesta and Ceres. http://dawn.jpl.nasa.gov/
Excellent link, recommended to anyone who hasn't hit it yet. Espie makes several good points, only one of which I could disagree with, and I could be wrong about that one.
Slashdot Mirror
That's an interesting issue, and one I didn't know about. Thanks. There's some misinformation in the article (the main posts), partially cleared up in the comments. It looks like a recurrence of the window scaling issue from a couple of years earlier. See http://lwn.net/Articles/92727/ for that one. In both cases this is about non-Linux systems not doing scaling correctly, and some philosophical points about working around other people's problems. Which, unfortunately, can be an issue with any community-developed OS.
OTOH, I've had problems with a TCP/IP stack on a commercial Unix, and Windows XP also had a window scaling issue. But none of this is an excuse for Red Hat not getting on the ball quickly for a client with a service contract, if they failed to do so. It looks as if there were options they could have taken, including adjusting tunables, which worked in some cases.
I'll have to look at the current implementation next week, on the theory that a problem that's recurred once might recur twice. Bummer. I did *not* need more stuff to do. But, seriously, thanks again for bringing this up.
OK, if that was the issue in the original AC post, and RH support couldn't deal with it, I'd say he had a valid complaint. You can deal with most issues (everything I've ever seen, though I haven't certainly seen everything) as either a kernel tunable, or on a per-interface basis. An OS support group should have been able to get that sorted.
Now I'm curious. Hopefully the AC, or someone else that's seen it, will post back.
I'll want to see a lot more detail before I'll believe the RHEL TCP/IP stack was causing "long random networking pauses."
In general terms, brevity in markup is desirable. Higher transmission speed, simpler (less bug-prone, resource-intensive etc.) creation and rendering tools, etc.
MathML will be the preferred solution for some things, just as Python is. OTOH, would you consider Python the language of choice for all problem domains? Nah, me either.
I've wanted good browser support for higher math for a *long* time. I'd hoped for something simpler than MathML. Alas, it was not to be. I'm not knocking the MathML folk. I trust them to have made things as simple as possible while dealing with a complex issue. Nonetheless, I'm a bit saddened by the verbosity of MathML.
Parent has a point. There could well be a large reduction in phishing sites--at least temporarily.
From TFA: "Naturally, in this sphere, as in other spheres, we should be thinking about adhering to Russian laws, about making sure that child pornography is not distributed, that financial crimes are not committed," he continued. "But that is a task for the law enforcement agencies. Total control and the work of the law enforcement agencies are two different things."
But then there's the Russian Business Network http://economist.com/displaystory.cfm?story_id=9723768, which has been an ongoing problem for some time. It's apparently politically protected.
Right you are. Take semiconductor manufacturing as one example. In the mid-eighties, there was concern on the part of the feds about machines disappearing into China. Ion implanters, etc. were regarded as sensitive technology, as the military, the space program, etc., was recognized as being strongly dependent upon a technological edge.
There had been a minor stir at the company I was working for, as we'd sold an implanter into Asia, and it turned out that the company we'd sold it to didn't exist, and the machine was gone. Probably into China. Later, there was some recognition that Perkin-Elmer (who made advanced photolithography equipment--a vital semiconductor technology) was under threat.
That was also the beginning of the end. In '85, during the first of the semiconductor dumping (selling below cost, to win market share) fiascoes, I was laid off. I was pretty bummed, as auto industry layoffs were getting all the press, and at that time, US semiconductor manufacturing was a larger industry. But Detroit had the lobbiests.
Very little semiconductor manufacturing equipment is made in the US any longer. Perkin-Elmer exited the business a long time ago, and most equipment is made in China. Yet our military is more dependent than ever on chips, for everything from night-vision goggles to the fly by wire control systems that all modern military aircraft depend on.
If China were to decide to nationalize the industries that we've off-shored, we would be screwed on many fronts. Semiconductors, advanced machine tools, and much more.
This may have a good side, in that that countries that are deeply into each others pockets don't tend to go to war. But I would date the rise of the multinationals, to the point where they could overrule governmental policy, from about twenty years ago.
It's a complex subject, and that's just my opinion. Nor is it currently the end of the world, even from a military perspective. But my take is that we are gradually painting ourselves into a corner, and we need to regain control of corporations. Right now, they have no loyalty to governments, only short-term profits and CEO bonuses. There's no longer any conception of a corporation being a good citizen. There's no corporate morality whatsoever--and they have the lobby dollars.
This can't be a Good Thing.
"... or tell if they got the right change."
I've actually started to notice when a cashier knows enough to count out my change correctly. Years ago, that was the standard of competence for a cashier, and I was annoyed when it wasn't met.
I've been seeing the same sort of thing in other areas, over a period of years. Others obviously have, too, including media writers and editors. I see small bits of evidence for that every day. An article mentions x MW (thermal) from a reactor, and the story is picked up by other publications, but modified to x MW of heat. That's a vote by the succeeding publications that 'thermal' may not be understood.
A 2001 NSF survey showed that about half of Americans didn't know that the earth revolved around the sun, and took a year to do so. But this stuff changes, and all hope is not lost. Though there is both good and bad news in NSF's Science and Engineering Indicators 2006 (the most recent available) Chapter 7: Science and Technology: Public Attitudes and Understanding at:
http://www.nsf.gov/statistics/seind06/c7/c7s2.htm
which is a very interesting read. There are spreadsheets in the appendices, etc. This is the best reference I know of.
OK, unless I get compelling reasons to opt for other choices, I'm leaning toward the Apple hardware. Though I have destroyed one Toughbook...
I'm rather a Unix fan (vastly too old to pass myself off as a boy of any sort) myself. I have to take a minor exception to claiming that any Unixy OS is the best available. It all depends upon what you need to *do* with it. I've called at least four variants 'best'. But that was about 'best for very specific purposes'. Sometimes it's come down to whether you can do non-blocking operations against a PRNG. So that experience may not apply for anyone needing a general purpose machine, though it has for me.
I haven't run into a case where I 'need Windows for high performance work'. For what I do, I've *almost* always found Windows at the low end of the performance scale. But that's just me, and I am a definite corner case. For instance, I rarely use an office suite, whether that be MS Office, OO, etc. Maybe once a day.
KDE apps meet my communications and calendar needs, as does KDE Kile (LATEX) for writing, as most of my writing involves math. KDE Kate can suck for text editing, as it supports only CVS, and I'm sensitive about sources, but there are easy workarounds. Most things I do that involve number-crunching and databases, I've written my own code to automate much of the drudgery away, where things are repetitive. I don't have to put up with spreadsheet charting limitations, in the sense that there are things you can't do with spreadsheet charting that are easy with http://www.graphviz.org/ and other specialized tools.
It will be interesting to see how my old-fart, laptops-are-lame attitude might change with nice hardware. Though some things that seem important to some people, I won't care about at all. DVD playback comes to mind. I have living-room electronics for that. If I get on a plane, I always have a couple of dozen papers to read. I never seem to get caught up on the journals.
Thanks. Of the three Wintel-type systems, I was leaning toward HP, if I had to go with one of The Big Three. It's hard to judge by my previous experience. For desktops and budget departmental (whitebox) servers, I build. When I've spec'ed rackable systems, I've had far better reliability results with HP than Dell. As a couple of people I occasionally contract with, and speak to (beers at the local) on a regular basis have had similar results, and they tend to think that Lenovo (whom I have no personal experience with) bites, I tend to listen to them. At least as far as ruling Lenovo out.
So far, and based partially upon your advice (thanks again) I'm leaning toward HP or Apple hardware. Possibly more toward Apple. It would be interesting to kick Leopard's tires. Perhaps a fat hard drive, and possibly quad booting CentOS, Fedora, Leopard, and still possibly OpenBSD. I usually opt for spending a bit more up front for an extra year or two of useful life. But somehow I've never taken that approach with laptops--perhaps because I'm so hard on them that I've come to regard them as disposable.
It will be interesting to see what results I get from spending a bit extra for nice hardware. One possibility is that if I know I've just spent top-dollar, I might actually take care of the bloody thing, vice tossing it onto the couch when I come home, etc.
"Apple's 'flashy 3d interface' brings style to an area of computing that desperately needed a breath of fresh air."
On one hand, that gives me the shudders. I don't want to even know that a backup is going on, save when I must unavoidably play media-monkey, or reevaluate the scheme because something has changed. Far better to design a reasonable strategy, test it thoroughly, and then have it be as unobtrusive as possible.
'Unobtrusive' and 'robust' are Good Things, as I am occasionally my own worst enemy. I once needed pretty much needed all my workstation resources while pulling an all-nighter, a daily rsync script launched, and I killed the job. A couple of hours later I shut down to swap some hardware, and the local drive didn't come back. I ended up losing several hours of work. Not good, particularly during an all-nighter. Now my daily rsync scripts look at machine loads, and are willing to sleep for a bit. At a certain point, though, they will run anyway, though bandwidth is throttled, proportional to load, as long as loads are above a threshold. The scripts are also sensitive to how I define the machine. DMZ hosts act somewhat differently than protected-LAN workstations.
My point is that I just want reliable (defined as knowing you *absolutely* can do a restore, given functional local hardware) backups to Magically Happen. I'm not familiar with Apple's product, but I've yet to see a solution that's as flexible as a set of well-crafted scripts, for networks of up to, say, a couple of dozen machines. I've used enterprise software that doesn't do load-sensing, though it did offer very useful features for dealing with networks of at least a few hundred machines, and a reasonable (not great) GUI.
I suspect, but cannot prove, that the issue for a large class of users (home and business) is an edu thing. If they aren't aware of the need, and the factors involved, those users probably aren't going to be helped much by a GUI. The solution, to my mind, is something involving a 'first boot' setup, such as what runs when you first launch a GUI after a fresh install of a Linux system. Establish a non-root account, etc. Note that at least some Linux distros don't do this if you're not installing a GUI, which I have a problem with.
This wouldn't be a trivial thing to do, as it would need to attempt some network discovery, to prompt for a possible network backup, etc. Detection of a local tape drive should definitely kick off a 'first-boot' backup configuration application app, or at least that should be selectable during OS install. Apple derives powerful advantages from controlling hardware, BIOS, and OS. In the case of backups, I very much doubt they're doing all they could. If they were, it would have been in the news as a major advance. Certainly nobody else is doing it.
"As it turns out, the MacBook is the Mac most people are buying. It is a competitive laptop to all but the bargain-basement craptops that Dell, Lenovo and HP sell. Get beyond the loss-leader "hacked by Chinese" craptops and you will find that MacBook is pretty damn competitive with the competition's lappies."
Query: IYO, are you suggesting that mid- and high-range laptops from Dell, Lenovo and HP are (or at least may be) reasonable values, or are you saying that their entire range is junk, and that you'd recommend another vendor for non-Apple laptops?
Reason I'm asking is that I'll probably be buying a new laptop (dual-booting CentOS and Fedora, possibly OpenBSD as well, though that wouldn't need a GUI) around Christmas, I've at least some reason for avoiding all of the above vendors, and it's not too soon to start asking around about others.
Is there any chance of being able to subscribe to a pizza-bush? I'd sell out in a heartbeat if I could subscribe to a pizza-bush...
"...100% of 8 PS3s indefinitely is preferable, from what he says, to the costly little slices of "real" supercomputers he tried to rent before."
Actually, it's preferable to me as well. He was getting $5k in NSF grant money for those runs. Which I helped pay for. I've no problem with that, but this is obviously a better deal from my tax-paying perspective.
One of the things he's researching is whether or not gravity waves should actually be detectable. We've already spent large amounts of money searching for them, without success. Then there's LISA, http://lisa.jpl.nasa.gov/WHATIS/intro.html which should fly in 2015.
If he should find that they are undetectable, or that their detection would require technologies we don't yet have, and his results are corroborated, *serious* savings would ensue. Achieving those savings from a corporate equipment donation valued at $3200, and a single researchers' time, would be sweet.
"Marriage exists for one reason, and one reason only - Succession of property rights. Allowing humans and robots to marry would mean allowing robots to own land. No more, no less."
No. Succession of property rights has been entirely subverted. Google for 'marriage strike'. The odds of a marriage ending in divorce are greater than even. No fault divorce means that either party can obtain a divorce on dissatisfaction alone. The odds of a male gaining physical custody of children are something like 1:40. The odds of a male retaining a home are something like 1:3. If you even attempt to get along, doing things that women's magazines promote, such as as buying random gifts, you are acting against your best interests. She now has an 'accustomed lifestyle' argument if the marriage doesn't work out, and it's in her financial interest to divorce.
Bottom line is that on her whim alone, you can lose *all* rights. Property, control of your financial future, even the right to see your kids without being messed with. The percentage of women admitting to messing with guys on that score is 40%, and the reasons are about the usual primate suckage. Revenge for perceived slights, etc. There's a large skew in surveys, involving people not admitting to having done things that demonstrably suck, even when they're assured of anonymity. That 40% is probably very conservative.
Give me a Cherry 2000. I really don't need a potential (the odds favor this outcome) gold-digger to somehow 'make me complete'. I don't read Cosmo. I'll take teh mad robot sex over incomprehensible female issues any time.
Was cooking dinner, kicking back, cleaning up a bit, and just generally having a nice Sunday. Read some news--boom! Now I have to bookmark a bunch of things, do a lot of reading, make moral decisions, etc.
I resent that. Yeah, eternal vigilance is the price of freedom, and all that. But eternal is beginning to get very damned eternal. A few years ago, I was wondering why I might have to explain what a blow job was when the kids did what I encouraged them to do--watch the news.
Those were the simple days. Now, if you can get your kids to watch the news, you might be having discussions about torture.
This is just depressing as hell. I'm tempted to vote for Ron Paul, in hopes of at least rattling some cages. In the end, I'll vote for whoever is toughest on lobbyists. But no candidate can be tough enough, as far as I'm concerned. The Great Experiment is failing.
Good catch. Certainly better than mine.
TFA: According to statistics provided by Market Share by Net Applications, starting in December 2006 and through September 2007, Linux doubled its market share. This detail would sound nothing short of promising, except for the fact that the doubling in market share is equivalent to a jump from 0.37% to 0.81%. In the past month, the open source operating system only increased its footprint on the market by 0.4%, from 0.77% to 0.81%.
YOU: Only increased by 0.4%?
Actually 0.81 - 0.77 = 0.04, not 0.4. Author of article can't do arithmetic, and you missed it.
Winds aloft are far more reliable, which is why wind generators have their blades higher than is required for ground clearance and common sense safety. Even a few meters of elevation is an efficiency win.
But all the press is about wind power generated almost at ground level. I'm wondering why aerial generation via aerostats isn't getting a few R&D dollars.
http://skywindpower.com/ww/index.htm
is an interesting site. Only a dozen pages or so in total. It uses frames, and I more or less hate it, but the *content* is extremely interesting. To break out one frames page:
http://skywindpower.com/ww/page002.htm "Capacity Factor" breaks out capacity factors for locations, at 15,000ft and 10k. There's also information on tethers, safety, etc.
Have a look. It would seem that wind could be more reliable than many people realize.
Are still on about hippies? I live in one of the most hippie-rich states in the US. And even here, they are *rare*.
Spontaneity, in terms of brain function, probably isn't always a Good Thing. I tend to think of brain function the way I do backups. Reliable, or counterproductive. In my experience of the average meeting, you can burn a lot of cycles trying to sort out whether the idiot in $arbitrary_chair actually made any sense this time around...
I got the joke. But I've just gotten out of a horrible end-of-week meeting, so I was forced to write this by Higher Powers.
I'd agree with running on a different port cutting your log noise down quite a bit. The last time I ran the experiment (about 6 months ago, IIRC), I got better than an order of magnitude reduction in brute force attacks.
One point, though. If anyone is part of a corporate admin team, make sure you talk to the group about it! Hosts running on a non-standard port can disappear as known/allowed ssh hosts, to some management software. If they're picked up later in a network scan, IDS logs, etc., much hilarity may ensue. Murphy being alive and well, it will be the middle of the night before someone connects, the IDS lights up, and your phone starts ringing.
You sending people off to this reference would seem to indicate that you don't think anyone will read more than the first bits.
http://en.wikipedia.org/w/index.php?title=Data_Encryption_Standard&oldid=161828931, so the Wiki article is versioned.
I guess it all depends upon whether you think factoring large numbers is a hard problem, whether special cases might exist, whether huge amounts of investment dollars matter, etc. From there you make your own call about whether or not to go all elliptical (another bag of worms) or not, etc. In the end, you either trust the math, or you don't. Not counting valid points you brought up about whether you can trust your hardware, compiler, or binary blobs.
One point you didn't bring up is rubber-hose cryptanalysis, which has a proven track record dating back through several centuries. It might be a lot easier for an adversary to ignore your opinions on math, the openness of your compiler, etc. and just beat the living hell out of you. Or just toss you in a cell for contempt of court until you either give up a passphrase, or grow old enough to win a sympathy argument.
Nothing is certain. First you evaluate the *perceived* value of the secrets you're trying to protect. Until you've done that, you can't estimate the potential intensity of the attacks that might be brought to bear in order to obtain those secrets. And only then can you think in terms of effective countermeasures. Assuming there are any, which may not be the case where, for example, an individual is squaring off against the resources of a governmental organization.
I don't know about an MIT ion drive test, but Dawn is scheduled for launch tomorrow. First non-experimental ion drive spacecraft, which is supposed to orbit Vesta and Ceres.
http://dawn.jpl.nasa.gov/
Excellent link, recommended to anyone who hasn't hit it yet. Espie makes several good points, only one of which I could disagree with, and I could be wrong about that one.
Without Linux, I doubt very much that we'd have a free Solaris. I'd be wildly surprised at a free Minux. Perhaps a BSD or two.