Well put. I would like to add that I think that much of _why_ Jon Katz is here is a twitching newsman's nose. Something is going on, and it is not as simplistic as geeks discovering Linux. Something of a second renaissance, I suspect. But what do I know? Even the historians a few hundred hears hence will get it mostly wrong.
Very good. I would add that, in a world of trees (whatever they are), Katz is looking for the forest. The whole/. thing seems more like a soap opera. Chinese curse? -- May you live in interesting times.
Well put. First, a free and independent/. has value, reputation if nothing else. Try to cage it or manage it and that value evaporates. Second, the impression I have of the/. community is that it seeks TRUTH./. comes from a pro-open-source, pro-Linux, anti-Microsoft bias, but I have found it (/.) to be probably the best source of information about Microsoft software:) (ducks and dons asbestos long-johns). I would expect to see a few more VA stories, not from bias, but from familiarity and the desire to post a good story. I didn't read the article, and judging from your post, I didn't miss a thing.
OK, starting random. Mutation A is "random" evolution. Mutation B is "directed" evolution. Iterate over a few billion years and observe. If there is a survival advantage to "directed" evolution, it will be there.
Also there is a human tendency to see patterns, faces in clouds,...
Agreed. The reportage was incredible. Seems like any random change which would allow a cell to have some slight tendency to behave nonrandomly with respect to environmental signals would have some increased degree of survivability. The differences could be small and subtle enough for quantum effects to matter. After a few billion years of evolution, why should mutations occur randomly with respect to the direction of evolutionary change? Why do most living things use the dextro form of optical isomers? Looks like he may have a handle on the mechanisms of how mutations could behave nonrandomly.
I was not one of the moderators, but I can assure you that the first two sentences were why. You very succinctly summed up the entire article, accidentally perhaps. The moderators saw meaning you never intended. This is the way evolution happens. On a large scale it can look as if it were designed, perhaps it is.
You are very much on track. Entropy only makes sense in a closed system. Closed means there are no inputs, there are no outputs, everything is inside the system. IIRC from thermodynamics, the formula for Enthalpy (Energy, good) is nearly the same as the formula for Entropy (Disorder, bad). Life exists by taking a process that goes from a high energy state to a low energy state and complicating it. To vastly oversimplify, animals turn carbohydrates and oxygen into carbon dioxide and water via a long chain of reactions and side reactions. As to the plausibility of life starting, look at the abundant life surrounding "smokers", undersea vents of very hot water laden with H2S. Also remember that the first life was not in an oxygen environment. The fresh air we breathe is the result of primal life poisoning its environment.
I cannot imagine VA or Andover being dumb enough to try to turn/. into a publicity rag. Any hint of censorship would raise a big stink. Even the possibility has raised a fair amount of/. commentary. I expect that any VA vs Penguin computing bias will be much less than the Linux vs Microsoft bias. Disclaimer. I am reading and posting this with IE5/NT4. (Haven't taken the plunge yet.) Besides,/. seems to be one of the few sources of useful info about Microsoft products (and yes I've got Technet).
Off Topic. I just saw the QOTD, and the mouthful of coffe went flying. "Trust me" (translation of the Latin: "caveat emptor." How do they do these things?
seems to be the norm here. Not that I object. Quite the contrary. Anything that serves as a watchdog _needs_ to be suspicious. Check the replies and the moderations and always take with copious quanties of NaCl.
It's not done with the license, but I imagine the effects would be much worse. Red Hat is positioned between the suits and the hackers, and to survive must have the trust and cooperation of both, ie reputation. Trying to kill the goose (ok, penguin) that lays the golden eggs is bad strategy. In this case it wouldn't be the goose, (er, penguin) that dies.
To see that Red Hat can have an extreme advantage, imagine an IS manager (closet technophobe, promoted from keypuch supervisor, doesn't grok the new-fangled stuff) with a problem, and he goes to Ask Slashdot, and the kindest words are RTFM. The threat may not be that severe, but the _perceived_ threat can be enough to cause panic. A phone number with a comforting voice on the other end is essentially a life-or-death matter.
What we are looking at is the start of some kind of symbiosis (read that as mutual parasitism) between the suits and the hackers. For an ecological analog, consider the warfare between grasslands and forests. Grass raises gnus (couldn't resist) to trample and destroy tree seedlings, and probably does something to help out lions which keep the gnu population in shape.
Even if the intentions are honest, it is extremely difficult to set up a fair benchmark. Mindcraft, with the 4 cpu, 4 network card setup and a carefully selected scsi was intentionally unfair from the beginning. It would be interesting to see the results with 5 cpus and 3 network cards.
Slightly off topic, regarding the "Crack this Server" contest (by ZDnet?), after the publication and public reception of the Linux hack (which wasn't really an OS problem anyway), if I _did_ have a crack of the MS box, I would shut my mouth, stifle it, and keep it closed. Something like "a hard act to follow." After the Linux crack, beautifully explained, an MS crack would come off as lame enough to be worse than useless. What happens with all this is that slowly Linux gains the type of credibility that takes years to acquire.
Somehow, reading your post started me off on a rant, a lot of somehow related images in my mind. If I were a wordsmith, this would read much better.
>>I'm not saying that I'm a good programmer or a very experienced one - its just common sense.
I like and trust your "inexperience" more than Microsoft's "experience".
Nothing is more uncommon than common sense. The road to disaster is to think you know what you are doing when you do not. It's not what we don't know, it's what we know that ain't so. -- Will Rogers? There is no "silver bullet".
The advisory looks more like old news than anything just discovered, but with the rapid rise in e-whatever and XML aimed at tying business to business, it's about time to post signs on the minefields (plural) that are waiting to blow up a lot of unsuspecting victims. With Microsoft promulgating the idiot's guide to e-commerce (behind some slick facade), there _will_ be plenty of victims.
Off topic. After some 30 years, unix is still around and going strong. Why? I think somehow the answer is related to what is right about your post. From Webster's second edition, hack 1. To cut irregularly, as if by repeated strokes of a cutting instument. The term hacker is very much associated with unix. A hacker must be one who hacks. Or creates hacks. The term has to be somewhat derogatory, but is a mark of esteem, as in kernel hacker. If you are facing a problem that is bigger than you are, all you _can_ do is hack. From Linus's keynote, "We've learned computers are just too damned hard to use." If Bourbaki (sp?) can have 100 pages on the difficulties of the number 1, imagine the diffulties in something vastly more complicated. Bourbaki is/was a them. Some number of top French mathematicians. Directly responsible for introducing the "new math". Teaching third-graders concepts that I first encountered as a math grad student. You can, and should, make a few things straight forward and easy, but mostly, and where it matters most -- not a chance.
What did you expect? Microsoft is essentially single-user, who can do anything. NT is a little better, but not by much. Unix security is primitive, but adequate for normally trustworthy users. Look for something (Capabilities?) that can control who can do what to whom and does not assume that users are benign. Good luck. As I remember it, MTS (Michigan Terminal System) had at least _some_ desirable properties. I think Multics (think of unix as castrated multics) has done some good work in that area.
It applies to US Congressmen and Senators who can keep in touch with their constituents without paying postage. There was, maybe still is, reduced postage rates for sending newspapers and such.
This could be good, but watch the implementation very closely. The Postal Service has a mandate to provide "universal service" to Americans, regardless of where they live. If the ability to receive e-mail is to be a "right", then the US Postal Service seems to be the best choice. This seems to be a case where a natural monopoly has desirable characteristics. "Universal service" means they have to serve the boondocks and can't just pick the highly profitable routes. With the problems with domain registries, would the Post Office be a better choice?
Slashdot Mirror
Well put. I would like to add that I think that much of _why_ Jon Katz is here is a twitching newsman's nose. Something is going on, and it is not as simplistic as geeks discovering Linux. Something of a second renaissance, I suspect. But what do I know? Even the historians a few hundred hears hence will get it mostly wrong.
Thanks. That made my day. /. would be poorer without him.
Personally, I think Katz adds a useful "outside but interested" perspective.
Very good. /. thing seems more like a soap opera. Chinese curse? -- May you live in interesting times.
I would add that, in a world of trees (whatever they are), Katz is looking for the forest.
The whole
It's a play on words from vaxen as a clutch of vaxes. Boxes is too sterile a term.
Well put. /. has value, reputation if nothing else. Try to cage it or manage it and that value evaporates. /. community is that it seeks TRUTH. /. comes from a pro-open-source, pro-Linux, anti-Microsoft bias, but I have found it (/.) to be probably the best source of information about Microsoft software :) (ducks and dons asbestos long-johns).
First, a free and independent
Second, the impression I have of the
I would expect to see a few more VA stories, not from bias, but from familiarity and the desire to post a good story. I didn't read the article, and judging from your post, I didn't miss a thing.
Moderate this up, please.
Sure. Montana, Yukon Territories, Polar Ice sheet, Siberia, Italy, Spain, Paris, London. Evolution is not direct and simple.
OK, starting random.
...
Mutation A is "random" evolution.
Mutation B is "directed" evolution.
Iterate over a few billion years and observe.
If there is a survival advantage to "directed" evolution, it will be there.
Also there is a human tendency to see patterns, faces in clouds,
Agreed. The reportage was incredible.
Seems like any random change which would allow a cell to have some slight tendency to behave nonrandomly with respect to environmental signals would have some increased degree of survivability. The differences could be small and subtle enough for quantum effects to matter.
After a few billion years of evolution, why should mutations occur randomly with respect to the direction of evolutionary change? Why do most living things use the dextro form of optical isomers? Looks like he may have a handle on the mechanisms of how mutations could behave nonrandomly.
I was not one of the moderators, but I can assure you that the first two sentences were why. You very succinctly summed up the entire article, accidentally perhaps. The moderators saw meaning you never intended. This is the way evolution happens. On a large scale it can look as if it were designed, perhaps it is.
You are very much on track. Entropy only makes sense in a closed system. Closed means there are no inputs, there are no outputs, everything is inside the system. IIRC from thermodynamics, the formula for Enthalpy (Energy, good) is nearly the same as the formula for Entropy (Disorder, bad). Life exists by taking a process that goes from a high energy state to a low energy state and complicating it. To vastly oversimplify, animals turn carbohydrates and oxygen into carbon dioxide and water via a long chain of reactions and side reactions. As to the plausibility of life starting, look at the abundant life surrounding "smokers", undersea vents of very hot water laden with H2S. Also remember that the first life was not in an oxygen environment. The fresh air we breathe is the result of primal life poisoning its environment.
I cannot imagine VA or Andover being dumb enough to try to turn /. into a publicity rag. Any hint of censorship would raise a big stink. Even the possibility has raised a fair amount of /. commentary. I expect that any VA vs Penguin computing bias will be much less than the Linux vs Microsoft bias. /. seems to be one of the few sources of useful info about Microsoft products (and yes I've got Technet).
Disclaimer. I am reading and posting this with IE5/NT4. (Haven't taken the plunge yet.) Besides,
Off Topic. I just saw the QOTD, and the mouthful of coffe went flying. "Trust me" (translation of the Latin: "caveat emptor." How do they do these things?
seems to be the norm here.
Not that I object. Quite the contrary. Anything that serves as a watchdog _needs_ to be suspicious. Check the replies and the moderations and always take with copious quanties of NaCl.
Considering the moderated post, looks more like a feature. (browsing at -6 threshold).
It's not done with the license, but I imagine the effects would be much worse. Red Hat is positioned between the suits and the hackers, and to survive must have the trust and cooperation of both, ie reputation. Trying to kill the goose (ok, penguin) that lays the golden eggs is bad strategy. In this case it wouldn't be the goose, (er, penguin) that dies.
To see that Red Hat can have an extreme advantage, imagine an IS manager (closet technophobe, promoted from keypuch supervisor, doesn't grok the new-fangled stuff) with a problem, and he goes to Ask Slashdot, and the kindest words are RTFM. The threat may not be that severe, but the _perceived_ threat can be enough to cause panic. A phone number with a comforting voice on the other end is essentially a life-or-death matter.
What we are looking at is the start of some kind of symbiosis (read that as mutual parasitism) between the suits and the hackers. For an ecological analog, consider the warfare between grasslands and forests. Grass raises gnus (couldn't resist) to trample and destroy tree seedlings, and probably does something to help out lions which keep the gnu population in shape.
We live in interesting times.
Even if the intentions are honest, it is extremely difficult to set up a fair benchmark. Mindcraft, with the 4 cpu, 4 network card setup and a carefully selected scsi was intentionally unfair from the beginning. It would be interesting to see the results with 5 cpus and 3 network cards.
Slightly off topic, regarding the "Crack this Server" contest (by ZDnet?), after the publication and public reception of the Linux hack (which wasn't really an OS problem anyway), if I _did_ have a crack of the MS box, I would shut my mouth, stifle it, and keep it closed. Something like "a hard act to follow." After the Linux crack, beautifully explained, an MS crack would come off as lame enough to be worse than useless.
What happens with all this is that slowly Linux gains the type of credibility that takes years to acquire.
Readers Digest used to have them IIRC.
Anyway, there is no reason to belive The Next Big Thing is _better_.
Who is stupid? The scammer or the scamee?
Somehow, reading your post started me off on a rant, a lot of somehow related images in my mind. If I were a wordsmith, this would read much better.
>>I'm not saying that I'm a good programmer or a very experienced one - its just common sense.
I like and trust your "inexperience" more than Microsoft's "experience".
Nothing is more uncommon than common sense.
The road to disaster is to think you know what you are doing when you do not.
It's not what we don't know, it's what we know that ain't so. -- Will Rogers?
There is no "silver bullet".
The advisory looks more like old news than anything just discovered, but with the rapid rise in e-whatever and XML aimed at tying business to business, it's about time to post signs on the minefields (plural) that are waiting to blow up a lot of unsuspecting victims. With Microsoft promulgating the idiot's guide to e-commerce (behind some slick facade), there _will_ be plenty of victims.
Off topic. After some 30 years, unix is still around and going strong. Why? I think somehow the answer is related to what is right about your post. From Webster's second edition, hack 1. To cut irregularly, as if by repeated strokes of a cutting instument. The term hacker is very much associated with unix. A hacker must be one who hacks. Or creates hacks. The term has to be somewhat derogatory, but is a mark of esteem, as in kernel hacker. If you are facing a problem that is bigger than you are, all you _can_ do is hack. From Linus's keynote, "We've learned computers are just too damned hard to use." If Bourbaki (sp?) can have 100 pages on the difficulties of the number 1, imagine the diffulties in something vastly more complicated. Bourbaki is/was a them. Some number of top French mathematicians. Directly responsible for introducing the "new math". Teaching third-graders concepts that I first encountered as a math grad student. You can, and should, make a few things straight forward and easy, but mostly, and where it matters most -- not a chance.
Argggh.
What did you expect?
Microsoft is essentially single-user, who can do anything. NT is a little better, but not by much.
Unix security is primitive, but adequate for normally trustworthy users.
Look for something (Capabilities?) that can control who can do what to whom and does not assume that users are benign. Good luck. As I remember it, MTS (Michigan Terminal System) had at least _some_ desirable properties. I think Multics (think of unix as castrated multics) has done some good work in that area.
Installed Microsoft software.
Right.
Whatever your feelings about vi vs emacs, emacs is surely innovative.
It applies to US Congressmen and Senators who can keep in touch with their constituents without paying postage. There was, maybe still is, reduced postage rates for sending newspapers and such.
This could be good, but watch the implementation very closely.
The Postal Service has a mandate to provide "universal service" to Americans, regardless of where they live. If the ability to receive e-mail is to be a "right", then the US Postal Service seems to be the best choice. This seems to be a case where a natural monopoly has desirable characteristics. "Universal service" means they have to serve the boondocks and can't just pick the highly profitable routes.
With the problems with domain registries, would the Post Office be a better choice?