Not to nitpick, but that should read '...when users are paying $60/yr per network'.
Ah, but don't forget whom they're targetting. Not the tech-savvy user who understands everything that's going on and can apply his own patches once he gets them. We're talking about the kind of user who understands that the patches must be applied, but doesn't trust himself to always do it properly. He likes to see thtat little blue checkmark next to his clock, or the "0 out of date systems" on the webpage.
RedHat sells peace-of-mind, not productivity, not stability, not performance. The latter three are available for free from numerous sources. Their target audience is the user who feels better having their help.
So, certainly, you or I (and many sysadmins I know) can easily subscribe a single computer and update an entire network based on that subscription. However, we could just as well get our updates from bugtraq or numerous other sources. Right now, RedHat is the most convenient. If we had to pay an absurd amount for their help, RedHat would cease to be the most convenient solution and we'd find another. We're not their target customer.
The sysadmin at my old job who managers 120 windows machines and 5 linux machines is their target customer. He knows how to follow instructions, he an amazingly hard worker, and he knows the importance of security. He doesn't, however, know a lick about programming, and he couldn't even rebuild an RPM without a GUI and specific instructions. He's got a lot of authority and a fat budget because he keeps the trains running on time. He's their target customer.
It costs a lot of money to backport security/bug fixes to old releases for years on end. RedHat can't afford to be doing that for products that people download for free.
Agreed, but they can afford it when people don't download for free. There's a lot of freeloaders out there, but I (for one) pay for my RHN subscription. Extremely little development has to take place supply updated apache RPMs for 7.2 when you're already supporting 8.0. Often, the fixes are done by others anyway, RedHat just has to supply the RPMs and run the update service. Not a bad deal when users are paying $60/yr per system. Sure as hell better ROI than developing a new version and hoping they'll upgrade, like proprietary vendors do. They could have at least phased out demo accounts and watched how it went for a year or so. People freeload because they can. Disable free accounts and more users will start paying. Opening up development to the Fedora project was a great idea, but it doesn't necessitate EOLing the old versions, especially when they can bring in income--possibly quite a lot of it too. RedHat won't know because they've never tried. The reason why people pay for free software is the support--the updates. It adds peace of mind.
Now RedHat has violently ripped that peace of mind out of the heads of its loyal customers. And don't give me that should-have-seen-it-coming crap. The people who really need update support are the ones who haven't been following all the newsletters, website chatter, mailing-list messages, and other forums used to communicate this kind of thing. The people who need this service the most (and are willing to pay for it) are the ones who "just want it to work" without having to spend hours researching. They don't read bugtraq--that's why they've paid RedHat to keep on top of the updates and automatically patch their systems. And now, right as RedHat is starting to build some serious trust within deep-pocketed customers, it is saying: by the way, we're not going to do this anymore in a few months. Sorry for the inconvenience.
They could have handled the change much more tactfully. For example, choose one or more of the following:
"Starting Jan 2004, we will no longer offer free demo update accounts."
"In order to increase our focus on stability and reliability, RedHat has chosen to merge our two product lines, RedHat Linux and RedHat Enterprise Linux in upcoming releases" And RedHat 10.0 Platinum will actually be the new RHEL release with a shiny new silver-and-red package and a slightly heftier pricetag.
"Starting 2Q 2004, RedHat will begin to phase-out support of old and difficult-to-maintain versions of RHL. Users are strongly encouraged to upgrade to the marvelously reliable RHL 10.0 Platinum with it's shiny new packaging"
See, the problem is that the announcement seems much too much like "We're not doing this anymore." You don't want to have the reputation of leaving your customers high and dry. It's bad for future business. A little foresight and some PR help could have saved them from the serious damage this announcement is going to cause to their reputation and future business.
I must say, the points brought up by this article are much more convincing then the "gray goo" argument about "what if the machines take over the world?"
The NYT article actually presents some valid, observable concerns with existing technology and our bodies' abilities to deal with particles on that scale. A surprisingly interesting read.
Cleanly burned pure hydrocarbons produce nothing but carbon dioxide and water--no pollution at all. The C02 and water both get recycled back into system by photosynthesis to make the very fuel and oxygen we used to get the energy out. Hydrocarbons are the form that the earth naturally uses to store the sun's energy, and I think we'd be wise get our energy the way nature intended it.
Fuel cells require more energy to create than they produce--they're just a new form of battery. Any energy source we ultimately decide on will be (at least indirectly) either solar, geothermal, or nuclear. Oil is just another form of solar power. It used to be a limited resource because all we had to use was the earth's surplus buried underground. Now we have a fast and efficient way of completing the cycle.
What we really should be focusing on and lobbying for is cleaner, more efficient ways of getting the energy out of the oil rather than just setting fire to it.
Thanks to rapid advances in thermal depolymerization, oil will likely be the fuel of the future. Only, we won't be getting it out of the ground. Instead, we'll be manufacturing it the same way the earth does: heat and pressure. But instead of taking millions of years, it takes just a few minutes.
And what can you make oil out of? Pretty much anything. Sewage, yard waste, paper, plastic, road-kill...
Recycling at its best. And this isn't theoretically-possible technology. This is currently-profitable-and-expanding technology.
... a corporate entity like Microsoft provides that open source may or may not: - Reasonably consistent consistent APIs across products...
I donno. I think POSIX is pretty standard.
Sure, there's quite a bit of choice in APIs (bluetooth comes to mind) while products battle it out for dominance. But once a project establishes itself as the dominant force, everything quiets down quite a bit. Established standards are rarely contested--the noise comes from competition to create the best standard for some new technology.
Arbitrarily picking an API and sticking with it (as it seems MSFT does) isn't necessarily a selling point. You end up with crappy APIs, like Win32.
It's been my unfortunate experience that those whose titles contain the word "Certified" tend to lag far behind the rest of us in the industry. In school, the hackers studied Computer Science or Electrical Engineering and learned how to make computers work. The dropouts from these majors who just couldn't hack it (those who wanted the jobs but didn't have that special gift that lets you think like a computer) switched instead to "Information Systems" in the business school.
Instead of learning how to make a computer work, there they learned how to work a computer. I.e., they learned how to use the programs that we learned how to create. Their programming coursework was graded on effort rather than effectiveness. The program they wrote didn't have to work (or even compile), it just had to look right. And yet, interestingly enough, part of this particular university's I.S. program was to complete the MSCE coursework and obtain certification.
Now that's scary.
You may be thinking that this is nothing special--network administrators don't need to know how to write programs, right? Well, immagine hiring an MSCE who can't even write a working batch file--but who can at least recognize one 80% of the time! This gets even more complicated in the *NIX world, where customization means script writing. Certification requires you to have more or less the same understanding of computers as someone like me has of quantum physics: memorized well enough to pass the test, but not necessarily understood enough to remember it next week.
Those of us with real computer skills never bothered to take those certification tests because they cost money and meant nothing (after all, the CS and EE dropouts are all certified). Do I really want to be placed in the same catagory as THEM? Those of us real admins who are actually certified only did it so they could pass the resume word search.
And yet, it's useful to know who understands system administration and who doesn't. Certification was a stab at making such a devision. It's relied on heavily by HR departments everywhere, despite the fact that it means so little in practical terms, because the HR department doesn't know any better.
But the truth is, IT certification is a flawed system at its very core. Certification is based on book work and tests--neither of which you'll find in the real world. It doesn't have to be that way, and here's why:
IT Certification should work like Pilot Licensing
In order to fly an airplane, you need to know the books, but you also need to know how to fly--that means extensive practical knowledge that can be demonstrated to an examiner in an actual airplane thousands of feet above the earth. You can't cram for that test.
In the same way, certified Linux experts should be able to demonstrate practical skills, not just pass tests. For example, a Linux system administrator should be able to write (in just a few minutes) a simple script to parse 300 log files looking for a specific pattern. He should also be able to modify the boot scripts such that a custom executable runs after (and only if) the network is brought online, but before any remote login programs start. They should be able to customize a firewall, build and install a kernel module, examine a syslog file, and maybe even configure xdm. And the stuff that he demonstrates has to *actually work*, not just look right.
I dare say that any of the existing certification courses will have a well-thought-out curriculum, and probably cover all the necessary aspects of system administration. However, knowing the location and purpose of the rc.sysinit script doesn't mean that you know how to customize it to do your bidding. Certified engineers may know "what", "where", and even "why". But only experience and practice teaches you "how". And knowing how is a mandatory component of being useful.
The TIA concept is not new, it's been explored in the sci-fi genre ad nauseum. However, one of the less examined aspects of implementing such a system in our society is the fact that we live in a representative republic. The people responsible for this system must be re-elected every few years by the very people they're oppressing.
So, the rights of the people must be respected to the extent that the people want them to be respected. If the people, as a whole, don't feel like they're being treated fairly, they'll elect officials who will change the laws to suit the people.
Obviously the people don't make all the decisions. If it were supposed to be that way, we'd live in a democracy (no, we don't) rather than a representative republic. The people only decide on really big issues. The rest is left to the politicians' judgement. Politicans can do things people don't want them to do as long as it doesn't make an impact large enough to warrant electing a different official.
So, the TIA initiative, in order to work, must remain unused and unabused enough that most people won't care. Unlike most people seem to think, nothing like this will ever create an undesired change in our lifestyle as a whole. It will instead be either exactly what most of us want, or a dark secret kept out of public view both in implementation and effect. Well-known and unapproved abuses of power in the eyes of his constituency mean political suicide for any elected official.
Big brother may be watching, but he's not allowed to touch if we wants to keep his power.
I downloaded openssh-3.1p1-10.src.rpm from rhn.redhat.com for RH 7.3. And I found in the SOURCES directory the file openssh-3.1p1-buffer-size.patch.
A quick look at the contents of the file reveals:
if (len > 0x100000) @@ -98,8 +99,13 @@ goto restart; } /* Increase the size of the buffer and retry. */ - buffer->alloc += len + 32768; - buffer->buf = xrealloc(buffer->buf, buffer->alloc); + + newlen = buffer->alloc + len + 32768; + if (newlen > 0xa00000) + fatal("buffer_append_space: alloc %u not supported", + newlen); + buffer->buf = xrealloc(buffer->buf, newlen); + buffer->alloc = newlen; goto restart; /* NOTREACHED */ }
If I'm not terribly mistaken, that patch fixes the hole in question. So if they already have the patched SRPMs posted, why can't they have the patched RPMs there too? (as of this posting, they don't yet.)
Well, looks like it's time to install the -devel packages and get to work with rpmbuild.
Re:What's with all of the bellyaching about speed?
on
Does C# Measure Up?
·
· Score: 1
If the application is high in CPU burn (lets call it X)
It's not absolutely damning that MS paid for the study.
Not damning, telling. Companies all over (e.g. IBM) have been doing their own studies in order determine whether or not switching to Linux is a cost-effective solution. And their results have shown that Linux is worth the investment. The article hints at this indirectly, noting that Linux is gaining ground rapidly.
On a recent trip to Alaska, a man in our group asked a park ranger about the polar ice caps melting. The ranger responded, "Of course they're melting. The ice age is over."
...and most of the time the browser catches infinite loops...
Give it up for the Halting Problem Solution. Whoo whoo!
Remember, the halting problem refers to a general algorithm for detecting whether a program finishes. No one ever said that you couldn't write a program to detect infinite loops with some specific archecture.
The WinXP core will actually detect potential infinite loops in device driver code and alert the user. An older version of the GeForce MX driver had such a problem. It used to frustrate me to no end with Win2K--I had no idea what was making the computer freeze.
Then I put XP on and the next time it happened, the computer froze for a second or two and then popped up a BSOD alerting me that it detected an infinite loop in the nVidia driver. Sure, I still had to reboot, but at least this time I knew what to fix.
The most common explanation for early "memories" (like pre-3 years old) is that you had heard the story when you were old enough to remember, then incorporated that into your long term memory.
If you can't trust your own memory, then what can you trust? Kinda makes you wonder how much of what you know really is real. I remember a few glimpses of the house I lived in when I was three, but I wonder how much of that I actually remember and how much I reconstructed from stuff I heard from what other people said. Especially since I don't remember hearing anyone else talk about it.
Just as worrysome is when it goes the other way and real memories turn into immaterial dreams. I once found (when I was 6) a diamond in an alley amid some broken glass. I took it home and showed my mom. She didn't belive me for a week. But she took it to a jeweler and it was, in fact, a real diamond.
Ten years later I was absolutely convinced that it never happened until I had proof otherwise. I had been trained not to trust my memories, because memories that old can't be real.
I think there's a certain degree of imposing intrusiveness in telling a person that his memories are not real, with nothing to back it up with but the latest phychological studies and heresay.
Does a man who knows nothing but books, who has studied psychology for years but whose entire body of relevant knowledge comes from the (sometimes questionable) studies and (often contradictory) conclusions of others, have any right or excuse to impose his own beliefs on others? Particularly when those beliefs contradict the other's own reason? I say I remember something happening. A psychologist says it never actually happened, and if it did happen, I don't really remember it; I just think I do. I have my own memory to back up my claim. The psychologist? Statistics and probability. He says that for other people they don't really remember those things, so therefore I don't either.
I had a girlfriend about 8 years ago who talked in her sleep (not mindless babble; very intelligent conversation) and slept A LOT because she had mononucleosis.
When she was asleep, she behaved a lot like people do when they're hypnotized. When asleep (and only when asleep) her hearing was amazing: she could hear a whisper 80 feet away when we were specifically trying to not let her hear. She also had an absolutely perfect memory of everything. And I do mean everything. She could quote to me word-for-word lengthy conversations I had had with her weeks, even months, earlier.
It might be worth mentioning that she, though absolutely alert, would refuse to open her eyes when she was asleep. She said it made her dizzy. She did just fine without them, though. She could move around, interact with her environment, walk, and I even saw her jog a few steps on a hill outside. Eyes closed the whole time.
Even more frightening still, when she was asleep, she mentioned quite casually that she had complete access to all her prior memories, and furthermore had absolute control over which of those her awake self could remember. She had to pick and choose which ones to give access to "other" awake self because when awake, she way too distracted by life and everything to be able to remember it all. It's as if the pathway to the memories was there, but she couldn't get to them because her mind was so busy doing what it has to do to stay awake.
Looking back, I think that her increased hearing ability and amazing memory were somehow tied to the fact that she refused to use her eyes. Just think of how much computing power it takes to process video, particularly if your primary task is recognizing what the objects you see are. Immagine having a computer that had the power to process images in real time with the power, speed and accuracy our own minds have. Now immagine shutting off that facility and using that processing power elsewhere. I think shutting down image processing takes a tremendous strain off your mind and could, in theory, free it to do more deep introspection than otherwise possible.
I once asked her when she was alseep what her earliest memory was. She said she was very small, laying on her stomach, looking down at her blanket but wanting to look up. She said she felt frustrated because she didn't know how to move. I guess she still hadn't figured how to move her limbs. I don't know how old that would put her at, but certainly not much. She estimated she was about two (days, not years).
She had no reason to lie about it either (and, it seemed, was in fact incapable of intentional deception when she was asleep) so at least she believed what she said. Whether it's true or not I don't know, but I have no reason to disbelieve her. She did things asleep that were far more amazing than remembering her infancy.
These people were stealing a VERY valuable commodity.. bandwidth. For those of you who don't work near the ISP industry, bandwidth is --VERY EXPENSIVE--. $200 per megabit per month is an absolute STEAL (to get that rate, you need to be buying it on the DS3 level). $400 per meg is more realistic on lower levels....
I'm sorry, but I have no sympathy for these people. What they did not only violated their agreement, but it cost someone else a LOT of money.
Slow down there, turbo. What they did was remove the cap on their cable modem. Removing the locks on the bank's outer doors is not the same as emptying the vault. You're basing your numbers on how much they could have stolen.
Furthermore, you really have no idea what you're talking about in the first place. Bandwidth is kind of a fuzzy term--we sometimes use it to mean throughput (MB/s), and sometimes to mean latency (ping time). Sometimes we actually mean bandwidth, but that's in Hz not MB, and I guarantee that you did not actually mean bandwidth.
Let's just pretend you meant throughput--most people do. My cable modem generally gets about 250 Kb/s, and I pay $40 per month. You just told me that 1 Mb/mo (not 1 Mb/s) should cost $200. I can get better throughput than that with a station wagon filled with magnetic tapes; I've got a month to get it there.
But for the benefit of the doubt, let's pretend you actually meant 1 Mb/s over the course of a month. What are the chances, do you think, that they could have sustained that kind of load for an entire month. That's 2.5 Tb of data they'd have to have transferred. These are home users, not an ISP. And even then, they'd only owe $200 ($400 with the expensive price).
Anyway, when an ISP pays high prices for bandwidth, what he's really paying for is the gurarantee that the bandwidth will be available when you need it--whether you use it or not. These people had no such guarantee.
The lawyers would argue that since they took the cap off their modem, they effectively had access to that kind of bandwidth for a period of time, though they probably never did anything with it (and logically really couldn't have, either). But they should have to pay for the unused ability to consume that bandwidth for that period of time.
What they stole wasn't the bandwidth itself but rather the possiblity to use that bandwidth. If you argue that they abused this ability and degraded the service of the company's other customers, then still the only damages the people would be liable for--what they actually "stole"--could be accounted as follows:
Take the bandwidth (in execess of what they were alotted) they consumed (probably in Mb/s) times the amount of time they spent actually consuming it (probably in seconds, maybe minutes) and divide it by the length of the billing cycle (likely 1 month). Then take that very, very, very small number and multiply it by the actual cost of that bandwidth over the billing cycle. The resulting price--the value of what they stole--could probably be taken out of their pocket change and still leave enough for a burger and fries.
Bottom Line: this definately looks like a civil case, not a criminal one: they received services that they didn't pay for, and they broke the rules of their contract. That's bad, but certainly not news, and certainly doesn't warrant the treatment they got. The value of the services they actually received realistically amounts to so little that settling out of court for $4 and a slurpee would be a good idea if the court understood the technology involved.
Linux is great for many projects like this because it posesses some qualities you won't find most other places. In particular:
No royalties. They can use it, hack it, sell it. Whatever they want, and never have to cut a check to anyone.
The resources. The Linux development community is unlike any other. Using Linux means you have access to all sorts of development and product resources for absolutely free. The newsgroups are friendly, the documentation is deep. And if you're doing something weird, do it with Linux and chances are someone will help you.
The name. If you need to impress the suits and get funding, Linux is a name you want to include. For a lot of people, Linux=cutting-edge technology. They don't understand it, but they know it's powerful, and they know it's gaining ground fast.
The power. There's no two ways around it. Linux is a powerful and flexible system. You can push it, pull it, tune it and tweak it to do just about anything. Unlike some other OSes, the kernel was written to stand on its own, not necessarily part of any prefab package. There's no GUI code in the TCP/IP stack, and it's just as happy in a PDA as it is in a supercomputer.
Could you honestly immagine LLNL buying a Windows-based clustered supercomputer? Yeah. Me neither.
Using Linux helps companies keep from having to re-invent the wheel while at the same time keeping their options open and their money in their own pocket. It works so well it's a wonder more companies don't use it.
For those afraid of the GPL, BSD presents a tempting alternative. But again, you lose a bit of the development resources and don't have the name to use to get your funding. For most people, though, GPL isn't a problem.
Yes, I saw this computer before it was delivered to LLNL (hardly off-the-shelf parts, BTW). Very sleek looking, though.
The interesting thing about this setup is that it doesn't work like the traditional supercomputer. It's more like a community of totally independant computers all willing to work on the same problem.
The system employs a whole lotta control nodes that spend their whole time trying to assign work out to the worker nodes. The problem then becomes not just parallelizing the work but coordinating the workers. Apparently with this cluster design, it's not all as cut-and-dried as with a "real" supercomputer. They have been able to do some really cool stuff, though. Like, for example, any computer in the cluster can address the memory on any other computer.
The admins I talked to said they weren't really sure just how fast the system could go, because they could never get it to operate at full capacity. They said the fastest they'd gotten it to go was 4T-Flops, but they figured they were only at %40 theoretical capacity.
I Am Not a Physicist, but I do remember some of the basics about interference patterns. Correct me if I'm wrong:
Light reflects off two surfaces, one just beneath the other. If the distance between the surfaces is such that the reflected light waves are perfectly out of phase, the waves will cancel eachother out, making it look like the surface actually absorbs that frequency range, producing color. That means that the distance the light travels between the plates is absolutely crucial in producing the right color. That's why butterfly wings shimmer. Your eyes are each viewing the wing at a different angle, each seeing a different color.
When light hits the plates striaght on, the light travels a certain distace between the plates. But when light hits at an angle, it travels slightly farther, depending on the angle. So, for example, instead of being out of phase at 600nm, light at 620nm will be out of phase, making a different color appear if you look at a different angle.
So unless I missed something, what we'll end up with is a display that "shimmers" like a butterfly wing. The hue of the display will shift when the screen is angled. That means that the effective viewable angle will suck a lot more than it does for LCDs, and it will be almost impossible to be perfectly sure what color you're looking at (particularly important for desktop publishing).
Perhaps someone who knows more about physics can explain how they intend to make this actually work. For now, though, I'm going to wait till I see a working prototype before I sell the farm to invest in their product.
This comparison brings up an interesting point. Darwin is open-source, and Linux is more mature and more quickly progressing.
Why did Apple choose to go out and start a new kernel project when they could have just based OS X on the Linux kernel instead? They could have gained so much ground and lost so little. It's worked for so many other companies--why not Apple?
Only reason I can see for running Darwin is for Mac hackers who want to enhance the OS -- but that poses another question: does Apple accept patches?
This is a serious question -- what are the benefits of Darwin being open?
A little research reveals that Apple does in fact accepts patches and hopes to see real real help and real results from the open source community with their kernel.
So the real question about their open-source philosophy is, Does it actually work? In other words, are they actually seeing results, and are we really trying to contribute.
I know for a fact that I don't ever plan to contribute to Apple's open-source projects because:
Darwin is relatively useless on x86
I can't afford to waste that much money to buy a mac just to play with it. I can build a quality x86 box for just a few hundred bones.
All of the fun Apple projects (Aqua et. al.) that I would be interested in tweaking are closed-source.
There's already another high-quality open-source UNIX-compatible kernel that's much more widely supported and understood. If I want to do any hacking, I do it with Linux.
I realize that Apple has reasons for not opening their other projects, and I don't expect them to change their minds any time soon. But how much help can they really expect when they don't give us any incentive to work with them?
Did Apple decide to take this road because "open source" was just one of those buzzwords that translated to "free labor" in the minds of management? Do they really have any intention of listening to what hackers want, or do they just expect us to work on anything that calls itself "open source"?
Slashdot Mirror
Ah, but don't forget whom they're targetting. Not the tech-savvy user who understands everything that's going on and can apply his own patches once he gets them. We're talking about the kind of user who understands that the patches must be applied, but doesn't trust himself to always do it properly. He likes to see thtat little blue checkmark next to his clock, or the "0 out of date systems" on the webpage.
RedHat sells peace-of-mind, not productivity, not stability, not performance. The latter three are available for free from numerous sources. Their target audience is the user who feels better having their help.
So, certainly, you or I (and many sysadmins I know) can easily subscribe a single computer and update an entire network based on that subscription. However, we could just as well get our updates from bugtraq or numerous other sources. Right now, RedHat is the most convenient. If we had to pay an absurd amount for their help, RedHat would cease to be the most convenient solution and we'd find another. We're not their target customer.
The sysadmin at my old job who managers 120 windows machines and 5 linux machines is their target customer. He knows how to follow instructions, he an amazingly hard worker, and he knows the importance of security. He doesn't, however, know a lick about programming, and he couldn't even rebuild an RPM without a GUI and specific instructions. He's got a lot of authority and a fat budget because he keeps the trains running on time. He's their target customer.
Agreed, but they can afford it when people don't download for free. There's a lot of freeloaders out there, but I (for one) pay for my RHN subscription. Extremely little development has to take place supply updated apache RPMs for 7.2 when you're already supporting 8.0. Often, the fixes are done by others anyway, RedHat just has to supply the RPMs and run the update service. Not a bad deal when users are paying $60/yr per system. Sure as hell better ROI than developing a new version and hoping they'll upgrade, like proprietary vendors do. They could have at least phased out demo accounts and watched how it went for a year or so. People freeload because they can. Disable free accounts and more users will start paying. Opening up development to the Fedora project was a great idea, but it doesn't necessitate EOLing the old versions, especially when they can bring in income--possibly quite a lot of it too. RedHat won't know because they've never tried. The reason why people pay for free software is the support--the updates. It adds peace of mind.
Now RedHat has violently ripped that peace of mind out of the heads of its loyal customers. And don't give me that should-have-seen-it-coming crap. The people who really need update support are the ones who haven't been following all the newsletters, website chatter, mailing-list messages, and other forums used to communicate this kind of thing. The people who need this service the most (and are willing to pay for it) are the ones who "just want it to work" without having to spend hours researching. They don't read bugtraq--that's why they've paid RedHat to keep on top of the updates and automatically patch their systems. And now, right as RedHat is starting to build some serious trust within deep-pocketed customers, it is saying: by the way, we're not going to do this anymore in a few months. Sorry for the inconvenience.
They could have handled the change much more tactfully. For example, choose one or more of the following:
- "Starting Jan 2004, we will no longer offer free demo update accounts."
- "In order to increase our focus on stability and reliability, RedHat has chosen to merge our two product lines, RedHat Linux and RedHat Enterprise Linux in upcoming releases" And RedHat 10.0 Platinum will actually be the new RHEL release with a shiny new silver-and-red package and a slightly heftier pricetag.
- "Starting 2Q 2004, RedHat will begin to phase-out support of old and difficult-to-maintain versions of RHL. Users are strongly encouraged to upgrade to the marvelously reliable RHL 10.0 Platinum with it's shiny new packaging"
See, the problem is that the announcement seems much too much like "We're not doing this anymore." You don't want to have the reputation of leaving your customers high and dry. It's bad for future business. A little foresight and some PR help could have saved them from the serious damage this announcement is going to cause to their reputation and future business.The NYT article actually presents some valid, observable concerns with existing technology and our bodies' abilities to deal with particles on that scale. A surprisingly interesting read.
Fuel cells require more energy to create than they produce--they're just a new form of battery. Any energy source we ultimately decide on will be (at least indirectly) either solar, geothermal, or nuclear. Oil is just another form of solar power. It used to be a limited resource because all we had to use was the earth's surplus buried underground. Now we have a fast and efficient way of completing the cycle.
What we really should be focusing on and lobbying for is cleaner, more efficient ways of getting the energy out of the oil rather than just setting fire to it.
And what can you make oil out of? Pretty much anything. Sewage, yard waste, paper, plastic, road-kill...
Recycling at its best. And this isn't theoretically-possible technology. This is currently-profitable-and-expanding technology.
I donno. I think POSIX is pretty standard.
Sure, there's quite a bit of choice in APIs (bluetooth comes to mind) while products battle it out for dominance. But once a project establishes itself as the dominant force, everything quiets down quite a bit. Established standards are rarely contested--the noise comes from competition to create the best standard for some new technology.
Arbitrarily picking an API and sticking with it (as it seems MSFT does) isn't necessarily a selling point. You end up with crappy APIs, like Win32.
Instead of learning how to make a computer work, there they learned how to work a computer. I.e., they learned how to use the programs that we learned how to create. Their programming coursework was graded on effort rather than effectiveness. The program they wrote didn't have to work (or even compile), it just had to look right. And yet, interestingly enough, part of this particular university's I.S. program was to complete the MSCE coursework and obtain certification.
Now that's scary.
You may be thinking that this is nothing special--network administrators don't need to know how to write programs, right? Well, immagine hiring an MSCE who can't even write a working batch file--but who can at least recognize one 80% of the time! This gets even more complicated in the *NIX world, where customization means script writing. Certification requires you to have more or less the same understanding of computers as someone like me has of quantum physics: memorized well enough to pass the test, but not necessarily understood enough to remember it next week.
Those of us with real computer skills never bothered to take those certification tests because they cost money and meant nothing (after all, the CS and EE dropouts are all certified). Do I really want to be placed in the same catagory as THEM? Those of us real admins who are actually certified only did it so they could pass the resume word search.
And yet, it's useful to know who understands system administration and who doesn't. Certification was a stab at making such a devision. It's relied on heavily by HR departments everywhere, despite the fact that it means so little in practical terms, because the HR department doesn't know any better.
But the truth is, IT certification is a flawed system at its very core. Certification is based on book work and tests--neither of which you'll find in the real world. It doesn't have to be that way, and here's why:
IT Certification should work like Pilot Licensing
In order to fly an airplane, you need to know the books, but you also need to know how to fly--that means extensive practical knowledge that can be demonstrated to an examiner in an actual airplane thousands of feet above the earth. You can't cram for that test.
In the same way, certified Linux experts should be able to demonstrate practical skills, not just pass tests. For example, a Linux system administrator should be able to write (in just a few minutes) a simple script to parse 300 log files looking for a specific pattern. He should also be able to modify the boot scripts such that a custom executable runs after (and only if) the network is brought online, but before any remote login programs start. They should be able to customize a firewall, build and install a kernel module, examine a syslog file, and maybe even configure xdm. And the stuff that he demonstrates has to *actually work*, not just look right.
I dare say that any of the existing certification courses will have a well-thought-out curriculum, and probably cover all the necessary aspects of system administration. However, knowing the location and purpose of the rc.sysinit script doesn't mean that you know how to customize it to do your bidding. Certified engineers may know "what", "where", and even "why". But only experience and practice teaches you "how". And knowing how is a mandatory component of being useful.
Well, I guess he's the one who'd know, right? Who better to tell everyone you're dead than your most serious competitor.
So, the rights of the people must be respected to the extent that the people want them to be respected. If the people, as a whole, don't feel like they're being treated fairly, they'll elect officials who will change the laws to suit the people.
Obviously the people don't make all the decisions. If it were supposed to be that way, we'd live in a democracy (no, we don't) rather than a representative republic. The people only decide on really big issues. The rest is left to the politicians' judgement. Politicans can do things people don't want them to do as long as it doesn't make an impact large enough to warrant electing a different official.
So, the TIA initiative, in order to work, must remain unused and unabused enough that most people won't care. Unlike most people seem to think, nothing like this will ever create an undesired change in our lifestyle as a whole. It will instead be either exactly what most of us want, or a dark secret kept out of public view both in implementation and effect. Well-known and unapproved abuses of power in the eyes of his constituency mean political suicide for any elected official.
Big brother may be watching, but he's not allowed to touch if we wants to keep his power.
If I'm not terribly mistaken, that patch fixes the hole in question. So if they already have the patched SRPMs posted, why can't they have the patched RPMs there too? (as of this posting, they don't yet.)
Well, looks like it's time to install the -devel packages and get to work with rpmbuild.
Indeed. :)
Not damning, telling. Companies all over (e.g. IBM) have been doing their own studies in order determine whether or not switching to Linux is a cost-effective solution. And their results have shown that Linux is worth the investment. The article hints at this indirectly, noting that Linux is gaining ground rapidly.
Yeah, I guess so, huh.
Remember, the halting problem refers to a general algorithm for detecting whether a program finishes. No one ever said that you couldn't write a program to detect infinite loops with some specific archecture.
The WinXP core will actually detect potential infinite loops in device driver code and alert the user. An older version of the GeForce MX driver had such a problem. It used to frustrate me to no end with Win2K--I had no idea what was making the computer freeze.
Then I put XP on and the next time it happened, the computer froze for a second or two and then popped up a BSOD alerting me that it detected an infinite loop in the nVidia driver. Sure, I still had to reboot, but at least this time I knew what to fix.
If you can't trust your own memory, then what can you trust? Kinda makes you wonder how much of what you know really is real. I remember a few glimpses of the house I lived in when I was three, but I wonder how much of that I actually remember and how much I reconstructed from stuff I heard from what other people said. Especially since I don't remember hearing anyone else talk about it.
Just as worrysome is when it goes the other way and real memories turn into immaterial dreams. I once found (when I was 6) a diamond in an alley amid some broken glass. I took it home and showed my mom. She didn't belive me for a week. But she took it to a jeweler and it was, in fact, a real diamond.
Ten years later I was absolutely convinced that it never happened until I had proof otherwise. I had been trained not to trust my memories, because memories that old can't be real.
I think there's a certain degree of imposing intrusiveness in telling a person that his memories are not real, with nothing to back it up with but the latest phychological studies and heresay.
Does a man who knows nothing but books, who has studied psychology for years but whose entire body of relevant knowledge comes from the (sometimes questionable) studies and (often contradictory) conclusions of others, have any right or excuse to impose his own beliefs on others? Particularly when those beliefs contradict the other's own reason? I say I remember something happening. A psychologist says it never actually happened, and if it did happen, I don't really remember it; I just think I do. I have my own memory to back up my claim. The psychologist? Statistics and probability. He says that for other people they don't really remember those things, so therefore I don't either.
It just seems wrong.
When she was asleep, she behaved a lot like people do when they're hypnotized. When asleep (and only when asleep) her hearing was amazing: she could hear a whisper 80 feet away when we were specifically trying to not let her hear. She also had an absolutely perfect memory of everything. And I do mean everything. She could quote to me word-for-word lengthy conversations I had had with her weeks, even months, earlier.
It might be worth mentioning that she, though absolutely alert, would refuse to open her eyes when she was asleep. She said it made her dizzy. She did just fine without them, though. She could move around, interact with her environment, walk, and I even saw her jog a few steps on a hill outside. Eyes closed the whole time.
Even more frightening still, when she was asleep, she mentioned quite casually that she had complete access to all her prior memories, and furthermore had absolute control over which of those her awake self could remember. She had to pick and choose which ones to give access to "other" awake self because when awake, she way too distracted by life and everything to be able to remember it all. It's as if the pathway to the memories was there, but she couldn't get to them because her mind was so busy doing what it has to do to stay awake.
Looking back, I think that her increased hearing ability and amazing memory were somehow tied to the fact that she refused to use her eyes. Just think of how much computing power it takes to process video, particularly if your primary task is recognizing what the objects you see are. Immagine having a computer that had the power to process images in real time with the power, speed and accuracy our own minds have. Now immagine shutting off that facility and using that processing power elsewhere. I think shutting down image processing takes a tremendous strain off your mind and could, in theory, free it to do more deep introspection than otherwise possible.
I once asked her when she was alseep what her earliest memory was. She said she was very small, laying on her stomach, looking down at her blanket but wanting to look up. She said she felt frustrated because she didn't know how to move. I guess she still hadn't figured how to move her limbs. I don't know how old that would put her at, but certainly not much. She estimated she was about two (days, not years).
She had no reason to lie about it either (and, it seemed, was in fact incapable of intentional deception when she was asleep) so at least she believed what she said. Whether it's true or not I don't know, but I have no reason to disbelieve her. She did things asleep that were far more amazing than remembering her infancy.
Somehow I don't think that a 5 fps animated mouth is going to catch on as a major tool for the hearing impared.
"Sure looked good on paper...."
What was wrong with speech to text?
Slow down there, turbo. What they did was remove the cap on their cable modem. Removing the locks on the bank's outer doors is not the same as emptying the vault. You're basing your numbers on how much they could have stolen.
Furthermore, you really have no idea what you're talking about in the first place. Bandwidth is kind of a fuzzy term--we sometimes use it to mean throughput (MB/s), and sometimes to mean latency (ping time). Sometimes we actually mean bandwidth, but that's in Hz not MB, and I guarantee that you did not actually mean bandwidth.
Let's just pretend you meant throughput--most people do. My cable modem generally gets about 250 Kb/s, and I pay $40 per month. You just told me that 1 Mb/mo (not 1 Mb/s) should cost $200. I can get better throughput than that with a station wagon filled with magnetic tapes; I've got a month to get it there.
But for the benefit of the doubt, let's pretend you actually meant 1 Mb/s over the course of a month. What are the chances, do you think, that they could have sustained that kind of load for an entire month. That's 2.5 Tb of data they'd have to have transferred. These are home users, not an ISP. And even then, they'd only owe $200 ($400 with the expensive price).
Anyway, when an ISP pays high prices for bandwidth, what he's really paying for is the gurarantee that the bandwidth will be available when you need it--whether you use it or not. These people had no such guarantee.
The lawyers would argue that since they took the cap off their modem, they effectively had access to that kind of bandwidth for a period of time, though they probably never did anything with it (and logically really couldn't have, either). But they should have to pay for the unused ability to consume that bandwidth for that period of time.
What they stole wasn't the bandwidth itself but rather the possiblity to use that bandwidth. If you argue that they abused this ability and degraded the service of the company's other customers, then still the only damages the people would be liable for--what they actually "stole"--could be accounted as follows:
Take the bandwidth (in execess of what they were alotted) they consumed (probably in Mb/s) times the amount of time they spent actually consuming it (probably in seconds, maybe minutes) and divide it by the length of the billing cycle (likely 1 month). Then take that very, very, very small number and multiply it by the actual cost of that bandwidth over the billing cycle. The resulting price--the value of what they stole--could probably be taken out of their pocket change and still leave enough for a burger and fries.
Bottom Line: this definately looks like a civil case, not a criminal one: they received services that they didn't pay for, and they broke the rules of their contract. That's bad, but certainly not news, and certainly doesn't warrant the treatment they got. The value of the services they actually received realistically amounts to so little that settling out of court for $4 and a slurpee would be a good idea if the court understood the technology involved.
- No royalties. They can use it, hack it, sell it. Whatever they want, and never have to cut a check to anyone.
- The resources. The Linux development community is unlike any other. Using Linux means you have access to all sorts of development and product resources for absolutely free. The newsgroups are friendly, the documentation is deep. And if you're doing something weird, do it with Linux and chances are someone will help you.
- The name. If you need to impress the suits and get funding, Linux is a name you want to include. For a lot of people, Linux=cutting-edge technology. They don't understand it, but they know it's powerful, and they know it's gaining ground fast.
- The power. There's no two ways around it. Linux is a powerful and flexible system. You can push it, pull it, tune it and tweak it to do just about anything. Unlike some other OSes, the kernel was written to stand on its own, not necessarily part of any prefab package. There's no GUI code in the TCP/IP stack, and it's just as happy in a PDA as it is in a supercomputer.
Could you honestly immagine LLNL buying a Windows-based clustered supercomputer? Yeah. Me neither.
Using Linux helps companies keep from having to re-invent the wheel while at the same time keeping their options open and their money in their own pocket. It works so well it's a wonder more companies don't use it.For those afraid of the GPL, BSD presents a tempting alternative. But again, you lose a bit of the development resources and don't have the name to use to get your funding. For most people, though, GPL isn't a problem.
The interesting thing about this setup is that it doesn't work like the traditional supercomputer. It's more like a community of totally independant computers all willing to work on the same problem.
The system employs a whole lotta control nodes that spend their whole time trying to assign work out to the worker nodes. The problem then becomes not just parallelizing the work but coordinating the workers. Apparently with this cluster design, it's not all as cut-and-dried as with a "real" supercomputer. They have been able to do some really cool stuff, though. Like, for example, any computer in the cluster can address the memory on any other computer.
The admins I talked to said they weren't really sure just how fast the system could go, because they could never get it to operate at full capacity. They said the fastest they'd gotten it to go was 4T-Flops, but they figured they were only at %40 theoretical capacity.
Michael Dell is only paying pennies per license. Msft had to release that information during the trial and it made Gateway very, very mad.
Light reflects off two surfaces, one just beneath the other. If the distance between the surfaces is such that the reflected light waves are perfectly out of phase, the waves will cancel eachother out, making it look like the surface actually absorbs that frequency range, producing color. That means that the distance the light travels between the plates is absolutely crucial in producing the right color. That's why butterfly wings shimmer. Your eyes are each viewing the wing at a different angle, each seeing a different color.
When light hits the plates striaght on, the light travels a certain distace between the plates. But when light hits at an angle, it travels slightly farther, depending on the angle. So, for example, instead of being out of phase at 600nm, light at 620nm will be out of phase, making a different color appear if you look at a different angle.
So unless I missed something, what we'll end up with is a display that "shimmers" like a butterfly wing. The hue of the display will shift when the screen is angled. That means that the effective viewable angle will suck a lot more than it does for LCDs, and it will be almost impossible to be perfectly sure what color you're looking at (particularly important for desktop publishing).
Perhaps someone who knows more about physics can explain how they intend to make this actually work. For now, though, I'm going to wait till I see a working prototype before I sell the farm to invest in their product.
Why did Apple choose to go out and start a new kernel project when they could have just based OS X on the Linux kernel instead? They could have gained so much ground and lost so little. It's worked for so many other companies--why not Apple?
This is a serious question -- what are the benefits of Darwin being open?
A little research reveals that Apple does in fact accepts patches and hopes to see real real help and real results from the open source community with their kernel.
So the real question about their open-source philosophy is, Does it actually work? In other words, are they actually seeing results, and are we really trying to contribute.
I know for a fact that I don't ever plan to contribute to Apple's open-source projects because:
I realize that Apple has reasons for not opening their other projects, and I don't expect them to change their minds any time soon. But how much help can they really expect when they don't give us any incentive to work with them?
Did Apple decide to take this road because "open source" was just one of those buzzwords that translated to "free labor" in the minds of management? Do they really have any intention of listening to what hackers want, or do they just expect us to work on anything that calls itself "open source"?