For example we use a Linux cluster for protein simulations to try to figure out how proteins fold (or as you would say - test out our ideas about how proteins fold...).
I would much rather you:
Get the funding you need to buy a real supercomputer, and,
Let the computer figure out "how" the protein folds through exhaustive algorithm analysis.
I understand that there's a practical limit to computer simulation; however, rather than merely testing potential explanations, we should have the computer itself iterate over the possible "simple" algorithms in the problem space, and quit when it finds a match with the observation set.
I must admit that I don't know exactly what you're dealing with exactly, but this is exactly the kind of problem that computer scientists love to solve. If you haven't already partnered with a grad student or CS professor, may I suggest that you do?
A computer scientist will be able to refine your algorithm to be more efficient, or perhaps know of a better algorithm. In my professional work, I have improved runtimes by four orders of magnitude in some cases. The value of algorithm refinement cannot be overstated.
A computer scientist will be able to point out those sections of code that perform redundant work, are inefficient, or non-portable.
A good computer scientist will be able to optimize the most intensive processing sections in assembly language. Typically, this will reduce run times by a factor of two or three, even when using optimizing compilers.
I've seen a lot of naive comments suggesting that supercomputers are being replaced by clusters. The truth is, anyone who can replace their supercomputer with a cluster didn't need a supercomputer in the first place:
(compared to a supercomputer):
The prime advantage of an x86-based server is that it is cheap, and it has a fast processor. It is only fast for applications in which the whole dataset resides in memory - and even then, it is still the slowest of the group.
Clusters are a little better, but suffer from severe scalability problems when driving IO-bound processes. As with the x86 server, if you can't put the full dataset into memory, you might as well forget using a cluster. The node to node throughput is several orders of magnitude slower than the processor bus in multiple CPU systems. (6.4GB/s vs 17MB/s for regular ethernet, or 170MB/s for Gigabit)
Multiple CPU servers do better, but still lack the massive storage capacity of the mainframe. They work better than clusters for parallel algorithms requiring frequent syncronization, but still suffer from a lack of overall data storage capacity and throughput.
Mainframes, OTOH, possess relatively modest processors, but the combined effect of having several of them, and the massive IO capability makes them very good for data processing. However, their processors aren't fast at anything, and often run at 1/2 or 1/3 the speed of their desktop counterparts.
Supercomputers combine the IO throughput of a mainframe with the fast processors typically associated with RISC architectures (if you can still consider anything RISC or CISC nowadays). They have faster processors, more memory, and much greater IO throughput than any other category.
It used to be that the prime reason for faster computers came from the scientific and business communities. But now that the internet has turned computers into glorified televisions, the challenges have gone from that of crunching numbers to serving content:
Clusters are great for serving read-only content, because there's very little active synchronization required between nodes, and the aggregate IO capacity scales well.
Mainframes reign when it comes to IO throughput - companies that formerly had use for a supercomputer are finding that their role is shifting to more of an information-provider role; faster processors are no longer as important as fast IO subsystems.
Scientists aren't being trained to use the computer as a tool; most think of a computer more or less as a means of verifying their hypothesis, rather than a means of discovering possible explanations. Their primary work is done with a calculator and pencil, and only later, when they need something to back up their ideas, do they turn to a computer simulation. The computer is a verification tool, not a means of discovery.
As our economy has shifted away from a technological base to an entertainment one, the need for supercomputers has begun to evaporate. We outsource innovation overseas so that we can lounge around on the couch watching tv and drinking beer (or surfing the net and drinking beer). The primary purpose of technological innovation has shifted from that of discovering the universe to merely bringing us better entertainment.
Why does Microsoft insist on selling Windows? It has more security holes than swiss cheese. It welcomes viruses with open arms.
The operating system has become a commodity. Why would MS develop an OS, when they could build a very good, closed source desktop, on top of Linux? They'd save themselves the headache of constant patching and maintaining an every larger, ever more-unmaintainable code base.
There's nothing in the GPL which states you can't distribute a closed source app with a GPL'd one. Granted, if you improve GPL'd software, you must provide the source. But if your software merely runs on top of Linux, you have no such obligation.
Microsoft has already shown that its strength is not in reliability, nor security. The fundamental selling point for MS software is ease of use - which, in spite of/. protest, is still much better than their FOSS counterparts - at least for the average person. It makes no sense for Microsoft to continue to develop operating systems when they are so much better at office and game applications. Money spent on the OS is simply wasted. No one runs Windows because they like it - they do it because they can't run anything else. Every hour MS spends debugging Windows is an hour that would have been better spent actually creating a revenue-producing product.
Many have weighed in about how Microsoft could crush OS by filing patent lawsuits, and that the individual developer would cave in because they couldn't afford the legal defense.
Suppose MS sued OS developers en masse. What exactly would the developers do then? Consider the alternatives:
Switch to a closed source model, and retain most of the code (absent the patent infringing sections).
Write software which detects, and refuses to interoperate with, Microsoft software. (Where would the internet be if every Linux and UNIX server refused a connection from a Windows box?) Think about it - the internet runs on UNIX, and this is a very real possibility.
Find security holes in Microsoft software.
Write viruses targetting Microsoft Windows.
What if, one morning ALL networked Windows machines refused to boot? The people smart enough to do something like this are now working on FOSS. What if someone distracted them?
FOSS keeps many would-be crackers from becoming problems to Microsoft. It is already bad enough that WinXP can be rooted on day zero. What would happen if Longhorn a virus infected Longhorn's Product Activation code? What if it infected UNIX boxes - not as a victim, but merely a carrier? Longhorn would be a boot once, never run OS.
The consequences of taking away FOSS through litigation are far worse than the dent it makes in corporate profits.
Only about as simple as Linux is to Linus. Let's face it, it's useless to build a new OS kernel from scratch, because everyone already has Windows, right? I mean, every computer comes with a copy of Windows, so why would you bother writing another OS?
Right?
Well, Linus showed that argument to be wrong. Granted, it isn't simple, but some FOSS authors aren't idiots, either. Some of us make our living solving problems far more complicated than MTA's and MTU's.
There are two choices, clientside or serverside.
What was I thinking? - if I can't imagine the solution, it _clearly_ can't be done!..... Oh, I see: it couldn't be both client and serverside, like MS did with Outlook and Exchange....
I've yet to see something that FOSS couldn't do successfully - provided they had some idea of what the program was supposed to do. Your post points out what is painfully obvious to the outside world about FOSS - the projects suffer not from lack of skill, but insight. If someone else hasn't done it first, they don't think it can be done.
Now, I realize this doesn't apply to all FOSS - don't get offended if you can think of several innovations actually created by FOSS. But, the most visible parts of the system to end users - Open Office, Mono, KDevelop, etc... seem to be little more than copies of existing proprietary software. And the reason why these projects don't innovate is because of people like the parent poster who religiously believe that any untried improvement to the product would be doomed to fail.
Granted, it does need to run on the server. But why not simply add to the existing infrastructure?
The web has mostly replaced ftp by now. Yes, it's still available, but far more people like to browse the web than ftp'ing files and viewing them later.
It's the same principle. There's email, and then there's securemail. Write the protocols, the clients, and the servers, and let it take off like P2P did. There's no need to completely replace email at first, but after a while, it would go the way of the BBS - yes, they're still around, but no longer the dominant means of communication.
Who pays for bounces? - Who pays now? That's right, the receiving ISP. This changes nothing.
Bandwidth? Right now, the recipient pays for bandwidth when they receive spam. As this system merely bounces a requests for warranties back and forth between servers, it uses less bandwidth than merely accepted all SPAM carte blanche and filtering on the recipient's server. This would reduce, not increase, the bandwidth used by spam - the recipient's server would simply not accept the spammer's email. Nor would it bounce, either - the connection would be refused for unauthorized mail.
Yes, it adds architectural, but not structural points of failure. By design, it adds more points of failure to a system which already has the potential for an almost unlimited number of failure points; as the internet is currently designed, if any server between me and the host goes down, so does the connection. This would have very little impact on the actual reliability of the service.
Reputation doesn't matter as long as the sender is willing to post a bond.
Again, bond requests will use much less bandwidth than spam - everyone outside the system will appreciate that those who use it won't be sending them spam.
So spammers ddos it. As if they don't ddos legitimate mail servers already?
No, you don't have to top off your escrow account unless you're sending spam and people cash in your bonds. And really, even if you're doing that, all you need is a credit card.
Okay, so you faked the FORM field. Still won't work, though, because your FROM field sender will get the bond request. When the spoofed user's server gets the bond request, it will look for the corresponding email, and upon failure to find it, will either send back a failure message, or simply drop it completely. So the intended recipient never gets the bond from your server, and the spoofed mail never gets sent anyway.
Unless, of course, signing up for a list required the signee to post escrow bonds as well.
Why should I care about reputation if the sender is willing to compensate me if I don't care for his message? Reputation is irrelevant when there's a bond involved.
Why shouldn't free software be the first to implement secure email? Imagine how much easier Linux advocacy would be if we could say: "SPAM? - I thought that was a Windows problem?..."
Imagine this conversation:
Tech: What's the problem?
User: I get all this SPAM, and I can't read my real email.
Tech: Let me guess, you're still using Windows, right?
User: How'd you know?
Tech: Because you're still getting SPAM. If you upgrade to Linux, which uses the SPAM-blocking mail protocol, your SPAM problem will go away... I'll send you a CD in the mail.
What really irks me is that rather than invent new solutions to existing problems, the free software community waits for a commercial vendor to implement a solution, and then copies it. What we should really be doing at this point is implementing a SPAM-free mail protocol in free software, which, once it became the standard, would force commercial companies into compliance, rather than trying to play a game of dodge-the-patent-lawsuit by copying someone else's improperly done anti-SPAM protocol.
Let's face the facts here, folks: if we wait for Microsoft to implement an anti-SPAM protocol, they'll do it wrong, and the free software world will be stuck trying to ensure compatibility with an interface that is fundamentally broken in the first place.
Why are you doing someone else's work for them, for free?
I find it hard to believe that the white hats are really doing it out of genuine concern for Corporate America. If you are really that altruistic, why not build a secure system that others could use, rather than try to break someone else's? So you discover an exploit - how does that help anyone if you don't also volunteer your time to help secure their system? Wouldn't it be better to help them migrate to a secure OS (such as *nix) rather than finding holes in their existing systems?
I mean, who really cares if you can root a Windows box anymore - it's got more holes than swiss cheese. The fact that you can find one of these holes doesn't make you smart, just annoying. We know Windows will never be secure; we'd appreciate it if you didn't break it any faster than normal.
I think a lot of white hats justify their cracking by calling it a community service, figuring that if they don't do any damage, then everything is ok. Well, it isn't. There's a certain degree of privacy that we like to have, and even if someone hacks a box with the best of intentions, the fact that the intrusion occurred is going to cost the victim. Consider how you'd react if you found a "friendly" reminder that you'd been hacked:
If you had any source code, you'd have to immediately file applications for any patentable algorithm contained therein - or risk someone else beating you to it and suing you for using your own invention. You'd also have to do a complete audit to make sure that the hacker didn't introduce any back doors into your code.
If there was any financial data, (such as CC numbers, etc...), you would have to close your accounts and re-open them.
If you had any "intimate" pictures of your girlfriend (okay, this is slashdot, but try to imagine it...), you would then have to explain to her that she might someday find these pictures on the internet somewhere. She'd probably leave you, too...
Really, what it comes down to is that hackers hack for the power they feel when they root someone else's machine - not because they're genuinely concerned about the welfare of others. Even those who don't damage the system are still breaking the law, and the mere fact that the breakin occurred does damage - even if it isn't apparent to the hacker.
HTML, no.... But I've never heard of anyone trying to build an HTML database. However, our vendor is pushing XML and XHTML as a replacement for existing database systems. Seems kind of stupid to replace a database with a markup language. For presentation, yes, it is fine, but it doesn't separate presentation from content any more than HTML did - unless you consider separation on the client to be something of interest. When people speak of separating presentation from content, it usually means one of two things:
For enterprise systems, it means that you don't need an IBM terminal to connect to your mainframe database; your data is reformatted according to the particular needs of the client by an intermediate server.
For web shops, it means that the client is free to reformat the data to their liking once the page has been returned to the client. They let the client do the rendering, rather than the server.
The first definition is the more common one, because it affects the bottom line of the firm. The first is favored in some cases because it optimizes bandwidth; the second in cases where the client is unknown. XHTML only helps in the second case, and it hinders the first by increasing the cost of data access - both in bandwidth and development time.
Actually, the comparison between urine and swamp water is inappropriate. Urine is relatively sterile compared to swamp water - yes, it contains waste chemicals (nasty, smelly chemicals), but very little bacteria. The reason it causes kidney damage (urine) is because you are reintroducing the same chemicals into your body that your kidneys are filtering out, albeit at a higher concentration.
And water is far more important than food. A soldier in the field without water will die in a matter of hours, but they can easily go for a day or more without food.
XHTML 2.0 almost forces you to seperate "content" from "presentation". Which is a good thing.
In theory, but not in practice. What happens is that more often than not, you end up with encoded XML in a database, which means that your data is no longer separate from presentation; access to your data now requires an XML parser as well.
Furthermore, you are forced to hardcode the relationships between elements. If your data is XML based, the only possible view is heirarchical; IMS was heirarchical, and look how long it lasted.
XML (and XHTML by extension) is the new COBOL. It isn't a smart idea by any standard:
It is far more verbose than delimited text files.
It isn't interoperable - if one company lists book authors as 'authors', and another as 'writers', you still can't do a meaningful comparison between the two documents. You'd do just as well with flat files and field descriptions.
It does not separate presentation from content, but rather enforces the presentation model on content. You can't arbitrarily structure your content - it must conform to the DTD. You're stuck if one department wants your books listed by author and another by title - you can do one or the other, but not both. A relational database where HTML is built on the fly would be a far better option - for you could at least use different queries.
You can't build a database with XML. A validating parser must read the entire file in order to extract even a single element. And searching and XML document for a particular record is horrendously slow - you must use a sequential text search.
Even if one did use XHTML for content storage, it would be read-only. Because it isn't a database, no two users could modify a record in the same document at the same time - it need not even be the same record; any record would lock out all others.
I'd like to hear from someone who is actually using XHTML to store content. My guess is that XHTML is being used more as a presentation kludge than as a content storage system.
What makes Linux so extraordinary is that it is NOT a commodity OS. It has a certain level of hackability and customizability that would be dangerous in the hands of Joe Sixpack.
As the saying goes, "Make something that even a fool can use, and only a fool will use it." Linux can't be used by fools, but Windows can.
Microsoft speaks against the GPL for this very reason - now the developers must reveal their source code, because it was based on GPL'ed code. But what they conveniently neglect to mention is that according to the EULA, a Windows developer cannot distribute, or even build, a derivative of Windows, under any terms. The license for GPL code covers only distribution of derivative products, whereas the MS EULA covers merely using the product. In fact, to even view the source code for an MS product requires that a developer agree to never develop a competing product!
Merely posting the source will allow these guys to continue to ship their product, but if they'd chosen the Microsoft development model, they'd owe royalties for every single product shipped!
Even though these guys might not like divulging their source code, they are still in a much better position than had they used Microsoft's code as a basis for their product.
I agree - but without the tech sector, the media sector is dead. Yes, you could incorporate DRM into hardware, etc... but it would only stifle the market.
I mean, its 2004 and I still don't have a DVD player for my desktop. It's not that I don't watch movies - just that I don't see the point in buying an already crippled piece of hardware. Why would I buy DVD's if I couldn't at least make a backup copy?*
* - yes, I know there are ways around this, but quite frankly, the hassle would be more effort than it's worth. I shouldn't have to be a reverse-engineer to use my own hardware.
It occurs to me, that as radical as this bill sounds, it is exactly what we need.
Suppose that it was indeed illegal to merely manufacture a device that could be used for copyright infringement. Consider how society would react:
Cameras would be illegal because they could take pictures of copyrighted works.
Scanners and printers would be illegal because they could be used to reproduce photographs and copyrighted texts.
PC's would be illegal because they could be used for copyright infringement.
VCRs and tape recorders would likewise be illegal. So would video or audio recording equipment of any kind.
CD players would likewise be illegal - because the earphone jack could be used to pick up clear signal.
DVD players would be illegal because the output jack could be redirected to illegal copying equipment.
DVD burners would be illegal.
You could still buy a tv, except that it you would have to pay for the built-in DVD player as well. And you couldn't buy a model with a cable hookup unless you also signed a cable contract at the same time. And an antenna hookup wouldn't be allowed under any circumstances.
To make a long story short, people would simply stop consuming digital media. Instead, they'd get out and socialize, or play sports, etc... instead of sitting at home in front of a computer or television. This bill would radically alter the spending habits of the American consumer, destroying the entertainment industry in the process:
About 30% of the MPAA member company profits are from DVD rentals. They could kiss that 30% goodbye.
The RIAA would likewise be affected - even moreso; who would buy a CD when there are no CD players? Instead, we'd just listen to the radio - the RIAA would suffer about a 85 - 95% decrease in revenue (Only the songwriters get paid for airtime).
I gaurantee this wouldn't last five years before it would be overturned.
What the RIAA and MPAA don't realize is that a substantial part of the value of a CD or DVD is the ability to 1.) make a backup copy, and 2.) mix MP3 tracks for mix CD's. Without this, there's not much point - I can simply listen to the songs on the radio and never bother to buy the CD. If this law actually passed, the dearth of technology improvements would actually drive the MPAA and RIAA out of business - who will buy a CD when electronics makers no longer make CD players for fear of legal liability? Who will make a DVD player when you could be sued for doing so?
recent fad in belief-systems (3000 years, granted, is a long fad).
And let's not forget that for about 2600 years of this fad, the most respected thinkers believed the Earth to be the center of the Universe.
They weren't off by much, were they?
Anyone who looks for absolute truth from science is going to be sadly disappointed. Science explains observations; it does not, nor can it, determine what is truth. Science only ponders what is already accepted as true - it does not determine it. The determination of truth belongs to the realm of theologians and philosophers.
If I wanted to know the distance to the moon, I'd ask a scientist. But I wouldn't be so naive to dismiss notions of the spiritual simply because modern science hadn't "proven" them to exist; after all, very few (if any) consider metaphysical questions. Most recognize the limits of the scientific method preclude useful study of such things.
biggest problem with self-aware computers is that we can't even agree what it means...
Step 1 of software engineering: Define the problem....
So, it would seem that the barrier to AI is one of human understanding, not of computational ability? If so, we'll never be able to make a computer smarter than us. The fantasies of a computerized future in which computers possess better judgement and intellect than a human is just that - a fantasy. In short, mankind cannot create what it can't understand.
But Garcia insisted the Democrats have the computer security situation well in hand, with the help of security specialists from Cisco Systems Inc. and Microsoft Corp. ''People can rest assured that we are aware of the need for a strong security system for our technology infrastructure," said Garcia, reading from a prepared statement, ''and we are working with our partners, Cisco and Microsoft, to ensure that our systems remain secure." [emphasis mine]
And this, my friends, is why Kerry won't be moving into 1600 Pennsylvania Ave next year....
Security and stability have never been Microsoft's defining features. The familiar joke about Democrats being stupid but passionate and Republicans being cold but smart seems to fit this situation rather well, unfortunately.
These guys would have been better off to call IBM or Red Hat, who believe that security is more than just a marketing buzzword.
Hopefully, the Cisco folks will be able to minimize the damage that the Microsoft does.
Self-aware computers
Even after 60 years of development, computers are still basically machines that can only crunch an endless stream of ones and zeros. Although several research projects are focusing on imbuing computers with reasoning and decision-making cognition - one has been under way for 20 years - that remains a holy grail for computer science. [emphasis mine]
I've long dismissed computers ever being self-aware. As I've heard before, "The subject of whether or not a computer can think is about as interesting as whether or not a submarine can swim". Computers weren't designed to think; they were designed to follow instructions. Hence, their supposed intelligence is necessarily limited to the intelligence of their programmers.
AI is indeed interesting when it comes to machine "learning" and adaptive problem solving, but things such as self-awareness and good judgement are far beyond the realm of computers. It isn't a problem of processing power or finding the perfect algorithm, but rather of the substance of self-awareness. Self-awareness is a property that the soul impinges on the mind, not an inherent property of neurons. The pursuit of self-aware machines could be compared to building an ever more complicated car and wondering why we still need drivers. The barrier isn't a matter of complexity or understanding per se, but rather the fact that good judgement and self-awareness are the result of a spiritual, rather than mechanical or chemical, process. You won't ever find these traits in an entirely mechanical process.
but unfortunately a little too expensive for my taste right now. Yes, I could design my own motherboard, but I'd pay more for the proto-fab than I would had just went out and bought one retail.
Price aside, this is an electronics hobbyist's dream. After the 80's, it became increasingly more difficult for the electronics enthusiast to build hardware of moderate complexity - soldering SMT components to a board is _NOT_ feasible for the average hobbyist. And I won't miss messing around with touching up broken traces and etchant (permanent marker isn't as necessarily permanent when etchant is involved...)
The day I can get a 20cm by 20cm prefabbed for about $10, I'll be sold. With all the threats of DRM this and Secure Hardware(TM) that, it won't be long before one might have to build their own PC if they still want to listen to MP3's....
It just doesn't make sense to me that in a time when website operators are charged with terrorism that we should only further enable terrorists by making it even easier for them to get a license - especially in light of the manner in which 9/11 was carried out.
Now you might reasonably argue that sport planes don't represent a terrorist threat, and you'd be mostly right. It would take a real stretch to repeat anything even remotely close to 9/11 with a sport plane. However, this adminstration is so paranoid that they're charging website operators with terrorism, and after that whole WMD flap, can you really expect them to know the difference between an ultralight and a 747?
Kudos for the FAA - they still know what freedom is. But this is surprising, coming from a government that considers encryption a weapon, and computers a terrorist tool.
That's all I wanted to point out - the apparent disconnect between government agencies is still alive and well, and the only thing different now is that we have fewer freedoms and more government.
That in the wake of 9/11 I can't get a driver's license without a mountain of paperwork, but I'm now allowed to fly a plane with only minimal qualifications?
We will invade a foreign country because they *might* have WMD's. Yet, now we're easing licensing restrictions for light aircraft, so that terrorists will have an even easier time loading up an aircraft with explosives and flying it into a public place?
While Michael Moore is being interrogated by the FBI, Achmed the suicide bomber is picking up his pilot license from the FAA...
Slashdot Mirror
For example we use a Linux cluster for protein simulations to try to figure out how proteins fold (or as you would say - test out our ideas about how proteins fold...).
I would much rather you:
- Get the funding you need to buy a real supercomputer, and,
- Let the computer figure out "how" the protein folds through exhaustive algorithm analysis.
I understand that there's a practical limit to computer simulation; however, rather than merely testing potential explanations, we should have the computer itself iterate over the possible "simple" algorithms in the problem space, and quit when it finds a match with the observation set.I must admit that I don't know exactly what you're dealing with exactly, but this is exactly the kind of problem that computer scientists love to solve. If you haven't already partnered with a grad student or CS professor, may I suggest that you do?
I've seen a lot of naive comments suggesting that supercomputers are being replaced by clusters. The truth is, anyone who can replace their supercomputer with a cluster didn't need a supercomputer in the first place:
- (compared to a supercomputer):
- The prime advantage of an x86-based server is that it is cheap, and it has a fast processor. It is only fast for applications in which the whole dataset resides in memory - and even then, it is still the slowest of the group.
- Clusters are a little better, but suffer from severe scalability problems when driving IO-bound processes. As with the x86 server, if you can't put the full dataset into memory, you might as well forget using a cluster. The node to node throughput is several orders of magnitude slower than the processor bus in multiple CPU systems. (6.4GB/s vs 17MB/s for regular ethernet, or 170MB/s for Gigabit)
- Multiple CPU servers do better, but still lack the massive storage capacity of the mainframe. They work better than clusters for parallel algorithms requiring frequent syncronization, but still suffer from a lack of overall data storage capacity and throughput.
- Mainframes, OTOH, possess relatively modest processors, but the combined effect of having several of them, and the massive IO capability makes them very good for data processing. However, their processors aren't fast at anything, and often run at 1/2 or 1/3 the speed of their desktop counterparts.
- Supercomputers combine the IO throughput of a mainframe with the fast processors typically associated with RISC architectures (if you can still consider anything RISC or CISC nowadays). They have faster processors, more memory, and much greater IO throughput than any other category.
It used to be that the prime reason for faster computers came from the scientific and business communities. But now that the internet has turned computers into glorified televisions, the challenges have gone from that of crunching numbers to serving content:As our economy has shifted away from a technological base to an entertainment one, the need for supercomputers has begun to evaporate. We outsource innovation overseas so that we can lounge around on the couch watching tv and drinking beer (or surfing the net and drinking beer). The primary purpose of technological innovation has shifted from that of discovering the universe to merely bringing us better entertainment.
Why does Microsoft insist on selling Windows? It has more security holes than swiss cheese. It welcomes viruses with open arms.
The operating system has become a commodity. Why would MS develop an OS, when they could build a very good, closed source desktop, on top of Linux? They'd save themselves the headache of constant patching and maintaining an every larger, ever more-unmaintainable code base.
There's nothing in the GPL which states you can't distribute a closed source app with a GPL'd one. Granted, if you improve GPL'd software, you must provide the source. But if your software merely runs on top of Linux, you have no such obligation.
Microsoft has already shown that its strength is not in reliability, nor security. The fundamental selling point for MS software is ease of use - which, in spite of /. protest, is still much better than their FOSS counterparts - at least for the average person. It makes no sense for Microsoft to continue to develop operating systems when they are so much better at office and game applications. Money spent on the OS is simply wasted. No one runs Windows because they like it - they do it because they can't run anything else. Every hour MS spends debugging Windows is an hour that would have been better spent actually creating a revenue-producing product.
Many have weighed in about how Microsoft could crush OS by filing patent lawsuits, and that the individual developer would cave in because they couldn't afford the legal defense.
Suppose MS sued OS developers en masse. What exactly would the developers do then? Consider the alternatives:
What if, one morning ALL networked Windows machines refused to boot? The people smart enough to do something like this are now working on FOSS. What if someone distracted them?
FOSS keeps many would-be crackers from becoming problems to Microsoft. It is already bad enough that WinXP can be rooted on day zero. What would happen if Longhorn a virus infected Longhorn's Product Activation code? What if it infected UNIX boxes - not as a victim, but merely a carrier? Longhorn would be a boot once, never run OS.
The consequences of taking away FOSS through litigation are far worse than the dent it makes in corporate profits.
Without breaking copyright law? Are they really going to license content from every single site? Or will the Feds bust them like Adam McGaughey?
Only about as simple as Linux is to Linus. Let's face it, it's useless to build a new OS kernel from scratch, because everyone already has Windows, right? I mean, every computer comes with a copy of Windows, so why would you bother writing another OS?
Right?
Well, Linus showed that argument to be wrong. Granted, it isn't simple, but some FOSS authors aren't idiots, either. Some of us make our living solving problems far more complicated than MTA's and MTU's.
There are two choices, clientside or serverside.
What was I thinking? - if I can't imagine the solution, it _clearly_ can't be done! ..... Oh, I see: it couldn't be both client and serverside, like MS did with Outlook and Exchange....
I've yet to see something that FOSS couldn't do successfully - provided they had some idea of what the program was supposed to do. Your post points out what is painfully obvious to the outside world about FOSS - the projects suffer not from lack of skill, but insight. If someone else hasn't done it first, they don't think it can be done.
Now, I realize this doesn't apply to all FOSS - don't get offended if you can think of several innovations actually created by FOSS. But, the most visible parts of the system to end users - Open Office, Mono, KDevelop, etc... seem to be little more than copies of existing proprietary software. And the reason why these projects don't innovate is because of people like the parent poster who religiously believe that any untried improvement to the product would be doomed to fail.
Granted, it does need to run on the server. But why not simply add to the existing infrastructure?
The web has mostly replaced ftp by now. Yes, it's still available, but far more people like to browse the web than ftp'ing files and viewing them later.It's the same principle. There's email, and then there's securemail. Write the protocols, the clients, and the servers, and let it take off like P2P did. There's no need to completely replace email at first, but after a while, it would go the way of the BBS - yes, they're still around, but no longer the dominant means of communication.
Why shouldn't free software be the first to implement secure email? Imagine how much easier Linux advocacy would be if we could say: "SPAM? - I thought that was a Windows problem?..."
Imagine this conversation:
Tech: What's the problem?
User: I get all this SPAM, and I can't read my real email.
Tech: Let me guess, you're still using Windows, right?
User: How'd you know?
Tech: Because you're still getting SPAM. If you upgrade to Linux, which uses the SPAM-blocking mail protocol, your SPAM problem will go away... I'll send you a CD in the mail.
What really irks me is that rather than invent new solutions to existing problems, the free software community waits for a commercial vendor to implement a solution, and then copies it. What we should really be doing at this point is implementing a SPAM-free mail protocol in free software, which, once it became the standard, would force commercial companies into compliance, rather than trying to play a game of dodge-the-patent-lawsuit by copying someone else's improperly done anti-SPAM protocol.
Let's face the facts here, folks: if we wait for Microsoft to implement an anti-SPAM protocol, they'll do it wrong, and the free software world will be stuck trying to ensure compatibility with an interface that is fundamentally broken in the first place.
Why are you doing someone else's work for them, for free?
I find it hard to believe that the white hats are really doing it out of genuine concern for Corporate America. If you are really that altruistic, why not build a secure system that others could use, rather than try to break someone else's? So you discover an exploit - how does that help anyone if you don't also volunteer your time to help secure their system? Wouldn't it be better to help them migrate to a secure OS (such as *nix) rather than finding holes in their existing systems?
I mean, who really cares if you can root a Windows box anymore - it's got more holes than swiss cheese. The fact that you can find one of these holes doesn't make you smart, just annoying. We know Windows will never be secure; we'd appreciate it if you didn't break it any faster than normal.
I think a lot of white hats justify their cracking by calling it a community service, figuring that if they don't do any damage, then everything is ok. Well, it isn't. There's a certain degree of privacy that we like to have, and even if someone hacks a box with the best of intentions, the fact that the intrusion occurred is going to cost the victim. Consider how you'd react if you found a "friendly" reminder that you'd been hacked:
- If you had any source code, you'd have to immediately file applications for any patentable algorithm contained therein - or risk someone else beating you to it and suing you for using your own invention. You'd also have to do a complete audit to make sure that the hacker didn't introduce any back doors into your code.
- If there was any financial data, (such as CC numbers, etc...), you would have to close your accounts and re-open them.
- If you had any "intimate" pictures of your girlfriend (okay, this is slashdot, but try to imagine it...), you would then have to explain to her that she might someday find these pictures on the internet somewhere. She'd probably leave you, too...
Really, what it comes down to is that hackers hack for the power they feel when they root someone else's machine - not because they're genuinely concerned about the welfare of others. Even those who don't damage the system are still breaking the law, and the mere fact that the breakin occurred does damage - even if it isn't apparent to the hacker.HTML, no.... But I've never heard of anyone trying to build an HTML database. However, our vendor is pushing XML and XHTML as a replacement for existing database systems. Seems kind of stupid to replace a database with a markup language. For presentation, yes, it is fine, but it doesn't separate presentation from content any more than HTML did - unless you consider separation on the client to be something of interest. When people speak of separating presentation from content, it usually means one of two things:
- For enterprise systems, it means that you don't need an IBM terminal to connect to your mainframe database; your data is reformatted according to the particular needs of the client by an intermediate server.
- For web shops, it means that the client is free to reformat the data to their liking once the page has been returned to the client. They let the client do the rendering, rather than the server.
The first definition is the more common one, because it affects the bottom line of the firm. The first is favored in some cases because it optimizes bandwidth; the second in cases where the client is unknown. XHTML only helps in the second case, and it hinders the first by increasing the cost of data access - both in bandwidth and development time.Actually, the comparison between urine and swamp water is inappropriate. Urine is relatively sterile compared to swamp water - yes, it contains waste chemicals (nasty, smelly chemicals), but very little bacteria. The reason it causes kidney damage (urine) is because you are reintroducing the same chemicals into your body that your kidneys are filtering out, albeit at a higher concentration.
And water is far more important than food. A soldier in the field without water will die in a matter of hours, but they can easily go for a day or more without food.
XHTML 2.0 almost forces you to seperate "content" from "presentation". Which is a good thing.
In theory, but not in practice. What happens is that more often than not, you end up with encoded XML in a database, which means that your data is no longer separate from presentation; access to your data now requires an XML parser as well.
Furthermore, you are forced to hardcode the relationships between elements. If your data is XML based, the only possible view is heirarchical; IMS was heirarchical, and look how long it lasted.
XML (and XHTML by extension) is the new COBOL. It isn't a smart idea by any standard:
I'd like to hear from someone who is actually using XHTML to store content. My guess is that XHTML is being used more as a presentation kludge than as a content storage system.
I'll stop using it.
What makes Linux so extraordinary is that it is NOT a commodity OS. It has a certain level of hackability and customizability that would be dangerous in the hands of Joe Sixpack.
As the saying goes, "Make something that even a fool can use, and only a fool will use it." Linux can't be used by fools, but Windows can.
Microsoft speaks against the GPL for this very reason - now the developers must reveal their source code, because it was based on GPL'ed code. But what they conveniently neglect to mention is that according to the EULA, a Windows developer cannot distribute, or even build, a derivative of Windows, under any terms . The license for GPL code covers only distribution of derivative products, whereas the MS EULA covers merely using the product. In fact, to even view the source code for an MS product requires that a developer agree to never develop a competing product!
Merely posting the source will allow these guys to continue to ship their product, but if they'd chosen the Microsoft development model, they'd owe royalties for every single product shipped!
Even though these guys might not like divulging their source code, they are still in a much better position than had they used Microsoft's code as a basis for their product.
I agree - but without the tech sector, the media sector is dead. Yes, you could incorporate DRM into hardware, etc... but it would only stifle the market.
I mean, its 2004 and I still don't have a DVD player for my desktop. It's not that I don't watch movies - just that I don't see the point in buying an already crippled piece of hardware. Why would I buy DVD's if I couldn't at least make a backup copy?*
* - yes, I know there are ways around this, but quite frankly, the hassle would be more effort than it's worth. I shouldn't have to be a reverse-engineer to use my own hardware.
It occurs to me, that as radical as this bill sounds, it is exactly what we need.
Suppose that it was indeed illegal to merely manufacture a device that could be used for copyright infringement. Consider how society would react:
To make a long story short, people would simply stop consuming digital media. Instead, they'd get out and socialize, or play sports, etc... instead of sitting at home in front of a computer or television. This bill would radically alter the spending habits of the American consumer, destroying the entertainment industry in the process:
I gaurantee this wouldn't last five years before it would be overturned.
What the RIAA and MPAA don't realize is that a substantial part of the value of a CD or DVD is the ability to 1.) make a backup copy, and 2.) mix MP3 tracks for mix CD's. Without this, there's not much point - I can simply listen to the songs on the radio and never bother to buy the CD. If this law actually passed, the dearth of technology improvements would actually drive the MPAA and RIAA out of business - who will buy a CD when electronics makers no longer make CD players for fear of legal liability? Who will make a DVD player when you could be sued for doing so?
Well, what I heard went something like this:
And let's not forget that for about 2600 years of this fad, the most respected thinkers believed the Earth to be the center of the Universe.
They weren't off by much, were they?
Anyone who looks for absolute truth from science is going to be sadly disappointed. Science explains observations; it does not, nor can it, determine what is truth. Science only ponders what is already accepted as true - it does not determine it. The determination of truth belongs to the realm of theologians and philosophers.
If I wanted to know the distance to the moon, I'd ask a scientist. But I wouldn't be so naive to dismiss notions of the spiritual simply because modern science hadn't "proven" them to exist; after all, very few (if any) consider metaphysical questions. Most recognize the limits of the scientific method preclude useful study of such things.
biggest problem with self-aware computers is that we can't even agree what it means...
Step 1 of software engineering: Define the problem....
So, it would seem that the barrier to AI is one of human understanding, not of computational ability? If so, we'll never be able to make a computer smarter than us. The fantasies of a computerized future in which computers possess better judgement and intellect than a human is just that - a fantasy. In short, mankind cannot create what it can't understand.
From the article:
But Garcia insisted the Democrats have the computer security situation well in hand, with the help of security specialists from Cisco Systems Inc. and Microsoft Corp. ''People can rest assured that we are aware of the need for a strong security system for our technology infrastructure," said Garcia, reading from a prepared statement, ''and we are working with our partners, Cisco and Microsoft, to ensure that our systems remain secure." [emphasis mine]
And this, my friends, is why Kerry won't be moving into 1600 Pennsylvania Ave next year....
Security and stability have never been Microsoft's defining features. The familiar joke about Democrats being stupid but passionate and Republicans being cold but smart seems to fit this situation rather well, unfortunately.
These guys would have been better off to call IBM or Red Hat, who believe that security is more than just a marketing buzzword.
Hopefully, the Cisco folks will be able to minimize the damage that the Microsoft does.
From the article:
Self-aware computers
Even after 60 years of development, computers are still basically machines that can only crunch an endless stream of ones and zeros. Although several research projects are focusing on imbuing computers with reasoning and decision-making cognition - one has been under way for 20 years - that remains a holy grail for computer science. [emphasis mine]
I've long dismissed computers ever being self-aware. As I've heard before, "The subject of whether or not a computer can think is about as interesting as whether or not a submarine can swim". Computers weren't designed to think; they were designed to follow instructions. Hence, their supposed intelligence is necessarily limited to the intelligence of their programmers.
AI is indeed interesting when it comes to machine "learning" and adaptive problem solving, but things such as self-awareness and good judgement are far beyond the realm of computers. It isn't a problem of processing power or finding the perfect algorithm, but rather of the substance of self-awareness. Self-awareness is a property that the soul impinges on the mind, not an inherent property of neurons. The pursuit of self-aware machines could be compared to building an ever more complicated car and wondering why we still need drivers. The barrier isn't a matter of complexity or understanding per se, but rather the fact that good judgement and self-awareness are the result of a spiritual, rather than mechanical or chemical, process. You won't ever find these traits in an entirely mechanical process.
but unfortunately a little too expensive for my taste right now. Yes, I could design my own motherboard, but I'd pay more for the proto-fab than I would had just went out and bought one retail.
Price aside, this is an electronics hobbyist's dream. After the 80's, it became increasingly more difficult for the electronics enthusiast to build hardware of moderate complexity - soldering SMT components to a board is _NOT_ feasible for the average hobbyist. And I won't miss messing around with touching up broken traces and etchant (permanent marker isn't as necessarily permanent when etchant is involved...)
The day I can get a 20cm by 20cm prefabbed for about $10, I'll be sold. With all the threats of DRM this and Secure Hardware(TM) that, it won't be long before one might have to build their own PC if they still want to listen to MP3's....
Well at least one person is paying attention!
It just doesn't make sense to me that in a time when website operators are charged with terrorism that we should only further enable terrorists by making it even easier for them to get a license - especially in light of the manner in which 9/11 was carried out.
Now you might reasonably argue that sport planes don't represent a terrorist threat, and you'd be mostly right. It would take a real stretch to repeat anything even remotely close to 9/11 with a sport plane. However, this adminstration is so paranoid that they're charging website operators with terrorism, and after that whole WMD flap, can you really expect them to know the difference between an ultralight and a 747?
Kudos for the FAA - they still know what freedom is. But this is surprising, coming from a government that considers encryption a weapon, and computers a terrorist tool.
That's all I wanted to point out - the apparent disconnect between government agencies is still alive and well, and the only thing different now is that we have fewer freedoms and more government.
That in the wake of 9/11 I can't get a driver's license without a mountain of paperwork, but I'm now allowed to fly a plane with only minimal qualifications?
We will invade a foreign country because they *might* have WMD's. Yet, now we're easing licensing restrictions for light aircraft, so that terrorists will have an even easier time loading up an aircraft with explosives and flying it into a public place?
While Michael Moore is being interrogated by the FBI, Achmed the suicide bomber is picking up his pilot license from the FAA...