Re:Linux needs a "centralized clearinghouse"
on
Linus Does Not Scale
·
· Score: 4, Insightful
I think this problem points out the biggest issue in regards to Linux--its almost anarchist style of code development.
It may seems so, to people coming from the corporate culture that uses language as a bludgeon (e.g. "issue" as a euphemism for "problem")--but having worked in both worlds I'm convinced that "independence" and "freedom" are better words for it than "anarchy" and "chaos."
I think the folks at IBM and Oracle ought to seriously have a LONG talk with Linux Torvalds himself and convince him to create a true clearinghouse where every improvement is approved by a committee. That way, Linux improvements happen in an organized fashion, which makes things way easier for developers and IT managers.
Convince him how? Work him over? Impress him with their org charts? Offer him the chance to work on a bigger project? Imply darkly that it won't look good on his annual review? If a variety of very smat people haven't been able to convince him with their best rational arguments, what exactly are the corporations supposed to do? And more to the point, do we really want to "convince" people of things that way?
And do you honestly believe that committees speed things up, or make them more organized, or easier to deal with? (This isn't a rhetorical question; for all I know you could be in full agreement with me, but writing in heavy sarcasm mode here and I'm just not getting it.)
Having managed a number of large projects I certainly find Linus's arguments more convincing; I'd much rather get on a ship run by one good captain than a ship where a variety of gizmos and rubber stamp committees had been put in charge for the express purpose of making the ship go faster than the captain thought prudent.
What I miss most is the old text-based Borland IDE. That was the most productive development environment ever. RHIDE is close, but wasn't stable on Linux when last I checked.
In other words, it is not a closed system he is describing, but an open system where energy is introduced (from the molecular motion of the atmosphere, which in turn is powered by the sun).
Furthermore, heating issues can be handled in the way they are handled in any electrical or mechanical system (in this case decoupling the ratchet, using active cooling, or whatever).
And how to you propose to power this "active cooling" system? If it and your ratchet are both 100% efficient you can break even; otherwise, you'll be operating at a net loss.
Before anyone else (the poster to whom I'm responding seems to understand this point) suggests passive cooling, that won't work either; your device is surrounded on all sides by the heat bath, otherwise you wouldn't be seeing the Brownian motion, remember?
T.E.D.: It doesn't matter who "governs" OpenGL. All Microsoft has to do to kill it now is refuse to license their 3D patents to any hardware vendor who chooses to make OpenGL drivers instead of DirectX.
Consider hooking this thing up to a Brownian Ratchet, such as discribed by Feynman in his lectures
The Brownian Ratchet you describe won't work, because of the second law of thermodynamics. The second law is potent enough that even evoking Feynman's name won't make it go away. Besides, what Feynman described was why this won't work.
See Chapter 46 of the Lectures if you want the details, but in short, it would quickly get hot enough that its own shaking (heat=random motion remember?) would drown out the Brownian motion.
Uh I hate to say this... (Score:1)
by jgerman on 3:32 09 January 2002 (#2808571)
(User #106518 Info)
...but that's not Spam. The difinition on that site is entirely too loose. The mail including the headers was valid, and Niel decides after recieving one unsolicited message that it's spam. Bullshit, if I send this guy an email by mistake am I spamming him? What if I send him an email and he just isn't interested, is it spam? Just sending large numbers of the same email out IS NOT spam. Note the definition listed does not even have any criteria based on whether or not the mail was solicited so even list hosts that you subscribe to qualify.
Quite true. I'd mod you up but I have no mod points save the ones I make up.
Better that than using it for APL I suppose. But I think what you really need for hands-free Perl is a personal (i.e. trainable) facial expression encoding system.
Gracenote is Rambus (Score:2)
by Nailer on 3:01 08 January 2002 (#2802924)
(User #69468 Info | http://www.cyber.com.au/)
Anyone else see the parallels?
How is this redundant? No one else has said anything remotely similar. I know--I just read all the points. I'd give you back a point of karma you lost if I had any mod points, but alas I do not.
What? They made *two* categories for window managers and time-wasters? After spending six hours trying to get Sawfish to work right, I would put them in the same group...
*laugh* I have no points and I must mod.
-- MarkusQ
P.S. As far as I know there is no truth to the rumour that "MQR points" will be phased out in favour of the euro.
Everything should be linked, from the highest level abstraction, all the way down to the bytes generated by the compiler. If you want to tweak the output of the compiler, you should be able to do so, attaching a modification tag into the source code.
Very interesting idea; it raises some obvious questions (e.g., how to code the "modification tags" in a way that doesn't totally break portability). I suspect it would work best at the upper levels (e.g. if you could write "sort X" or "sort [use quick-sort] X" or "sort [if count X lt 20 use bubble-sort else use quick-sort] X", etc.) But then of course the question would be "can't you already do that?"
Perhaps, but it's redundant. "it can work up to 1 kilometre" is sufficient.
Not really. In this case the "up to 1 kilomtre" (of air) that it "works over" is the signal medium. Suppose instead we were going over some type of cable (and fliping the length specification to postfix for clarity):
...it can work over optical cat-hair cables upto 1 kilometre long...
You clearly wouldn't write:
...it can work optical cat-hair cables upto 1 kilometre long...
And by the same logic, I would say the word "over" is needed in the original context.
To think that some twenty-something geek having Linux dreams in his head could convince someone in finance to implement an open source accounting package is pure fantasy (if you have anyone of a rational mindset in finance).
This sounds very much like what I was told in the 70's when I was in a start up trying to sell microcomputer turnkey accounting software to small businesses: no one will ever trust a micro, and you're delusional if you think otherwise.
Our main problem turned out to be keeping up with demand and managing explosive growth.
There are about 10^120 positions in chess, if I remember correctly. Even if you could solve it, there'd be no possible way to store the answer
Not quite true; it is likely that the solution could be expressed as an algorithm that would take up much less space than an exhaustive list of state transitions.
For example, if you had a game where the goal was to take turns naming positive integers until someone was forced to name an integer larger than the one their opponent just named (and thus lost), there are an infinite number of states, but the "solution" can be expressed quite simply:
Say "one."
So, at least in this case, the solution is much smaller than the list of all posible cases. I suspect the solution to chess, while much larger than this, is still smaller that you propose.
-- MarkusQ
Re:Please don't take it personally...
on
CPU Wars
·
· Score: 2
by Anonymous Coward on 10:57 03 December 2001 (#2648773)
but I'm sick of this "C for kernel, bloatlang for everything else" BS. I'm writing a ALife sim of language evolution program (for my thesis) in C and I'm thinking about writing agent ai code in *asm*, because compiler generated code won't be fast enough.
I assume you registered at the university instead of attending as an "Anonymous Coward"--otherwise, the diploma isn't going to do you much good.
Were you perhaps thinking of Napster? Gnutella is about as P2P as it gets; there's no central server, and once two nodes have been introduced (e.g., at least one of them has added the other to its host list) they can reconnect even if everyone else is shut down. Granted, it may take awhile if the original network was >> 2 nodes--but it doesn't take a very large fraction of the network to self-connect within a reasonable time.
Let me explain. I was responding to the post I followed up to; one about AI recognition of SPAM. I was specifically not writing about the community spam identification effort with hashes, as my post clearly shows.
There is an abundance of wasted energy ready to tap at many points....
It's not perpetual motion, but an attempt to retain energy that is now simply radiated away.
But heat isn't free energy (free in the physics sense, not in the open source sense). True, you can get energy from a difference in temperature, but only by slowing the flow of heat that would have otherwise taken place (just like damming a river) and thus raising the entropy (in this case, temperature). Now, doing so will make your engine run hotter, and thus less efficient, and you have a net loss.
Suppose you do something to cool the heat sink to make up for this. Then you have two cases to consider: either 1) you are using energy to do this, or 2) you have a passive way to do it. In case 1 you are still at a net loss, but in case 2 you might well be doing better than the original system. But you've then changed the base case--if you used the passive cooling trick (a heat sink or whatever) on the original system, you would have gotten a greater gain, so your gizzmo is still costing you.
This has the same problem as the things that generate electricity from your body heat/motion/whatever. By adding such a device to the system you make the original system harder to cool (because your gizzmo acts as an insulator) or harder to move (because your gizzmo has mass) or whatever (details vary depending on how you're trying to get energy out of the system) and in the end you will reduce the efficiency by an amount that will require you to put more fuel/power/food/whatever into the original system. If your parasitic gizzmo were 100% efficient you still wouldn't gain anything, and in any real case you'll face a net loss.
Example: You put a heat-based gizzmo on your car's exhaust pipe. The temerature (and thus pressure) in the exhaust system goes up, making the engine less efficient and making you use more fuel to go the same distance.
Example: You put one on your CPU. Same deal, except your cooling system now has to work harder to keep it at a reasonable temperature, and thus uses more power.
Example: You wear a swatch. It takes a little bit more energy each time you move your arm. If you want to power a computer the same way, you'll soon be too tired to type.
The key point is in every case you will have to put more energy in than you get back out. That's why perpetual motion machines do not and can not work.
To have any hope of working over the long term, this kind of an approach must include the ability to distribute not just the hashes themselves, but the hash function as well, so that the hash function itself can be adjusted, when needed.
Yes! In fact, why not have many fuzzy hash functions floating around at once? That way, their task would be to come up with something that yielded a different hash against all of the hash functions at once, a much harder problem. If some spammer figures out a way to do it, an anti-spammer can devise a function (looking at lots of copies of the spam, which shouldn't be hard to come by) that would catch it, and now that trick won't work any more.
Distributing the functions with the hash (with a few safe guards, e.g. re: the halting problem) would make this darned near imposible to beat.
Slashdot Mirror
It may seems so, to people coming from the corporate culture that uses language as a bludgeon (e.g. "issue" as a euphemism for "problem")--but having worked in both worlds I'm convinced that "independence" and "freedom" are better words for it than "anarchy" and "chaos."
I think the folks at IBM and Oracle ought to seriously have a LONG talk with Linux Torvalds himself and convince him to create a true clearinghouse where every improvement is approved by a committee. That way, Linux improvements happen in an organized fashion, which makes things way easier for developers and IT managers.
Convince him how? Work him over? Impress him with their org charts? Offer him the chance to work on a bigger project? Imply darkly that it won't look good on his annual review? If a variety of very smat people haven't been able to convince him with their best rational arguments, what exactly are the corporations supposed to do? And more to the point, do we really want to "convince" people of things that way?
And do you honestly believe that committees speed things up, or make them more organized, or easier to deal with? (This isn't a rhetorical question; for all I know you could be in full agreement with me, but writing in heavy sarcasm mode here and I'm just not getting it.)
Having managed a number of large projects I certainly find Linus's arguments more convincing; I'd much rather get on a ship run by one good captain than a ship where a variety of gizmos and rubber stamp committees had been put in charge for the express purpose of making the ship go faster than the captain thought prudent.
-- MarkusQ
Edible concrete has been with us for years:
coarse aggregate - dried fruit
fine aggregate - flour
binding agent - eggs.
It proved so popular, it's got its own name - cake!
Let's face it, Marie Antoinette would've looked a bit of a dork saying "let them eat edible concrete."
I haven't any real mod points, but this AC post deserves notice in the context.
-- MarkusQ
Try SETEDIT; it's pretty darn close.
-- MarkusQ
Furthermore, heating issues can be handled in the way they are handled in any electrical or mechanical system (in this case decoupling the ratchet, using active cooling, or whatever).
And how to you propose to power this "active cooling" system? If it and your ratchet are both 100% efficient you can break even; otherwise, you'll be operating at a net loss.
Before anyone else (the poster to whom I'm responding seems to understand this point) suggests passive cooling, that won't work either; your device is surrounded on all sides by the heat bath, otherwise you wouldn't be seeing the Brownian motion, remember?
-- MarkusQ
I agree.
I suppose you think windmills don't work, either.
They only work if the air has net motion relative to the windmill. You can't run one off of still air just because the air is hot.
-- MarkusQ
Or even in addition to DirectX.
-- MarkusQ
The Brownian Ratchet you describe won't work, because of the second law of thermodynamics. The second law is potent enough that even evoking Feynman's name won't make it go away. Besides, what Feynman described was why this won't work.
See Chapter 46 of the Lectures if you want the details, but in short, it would quickly get hot enough that its own shaking (heat=random motion remember?) would drown out the Brownian motion.
-- MarkusQ
by jgerman on 3:32 09 January 2002 (#2808571)
(User #106518 Info)
Quite true. I'd mod you up but I have no mod points save the ones I make up.
--MarkusQ
Speech recognition for hacking Perl, anyone?
Better that than using it for APL I suppose. But I think what you really need for hands-free Perl is a personal (i.e. trainable) facial expression encoding system.
-- MarkusQ
by Nailer on 3:01 08 January 2002 (#2802924)
(User #69468 Info | http://www.cyber.com.au/)
Anyone else see the parallels?
How is this redundant? No one else has said anything remotely similar. I know--I just read all the points. I'd give you back a point of karma you lost if I had any mod points, but alas I do not.
-- MarkusQ
10 do while inkey$ = ""
20 loop
30 print "You can't hack me, f00l!!!"
40 goto 10
Thus proving the point; I would bet that most actual implementations of this would fall to a "random control character" attack (e.g., ^C, ^Z, etc.).
-- MarkusQ
*laugh* I have no points and I must mod.
-- MarkusQ
P.S. As far as I know there is no truth to the rumour that "MQR points" will be phased out in favour of the euro.
Very interesting idea; it raises some obvious questions (e.g., how to code the "modification tags" in a way that doesn't totally break portability). I suspect it would work best at the upper levels (e.g. if you could write "sort X" or "sort [use quick-sort] X" or "sort [if count X lt 20 use bubble-sort else use quick-sort] X", etc.) But then of course the question would be "can't you already do that?"
-- MarkusQ
Not really. In this case the "up to 1 kilomtre" (of air) that it "works over" is the signal medium. Suppose instead we were going over some type of cable (and fliping the length specification to postfix for clarity):
You clearly wouldn't write:
And by the same logic, I would say the word "over" is needed in the original context.
-- MarkusQ
that can work over [a distance of] up to 1 kilometre
-- MarkusQ
You're mostly right, but water is a solvent, it can't be saturated.
Sure it can.
--MarkusQ
This sounds very much like what I was told in the 70's when I was in a start up trying to sell microcomputer turnkey accounting software to small businesses: no one will ever trust a micro, and you're delusional if you think otherwise.
Our main problem turned out to be keeping up with demand and managing explosive growth.
-- MarkusQ
Not quite true; it is likely that the solution could be expressed as an algorithm that would take up much less space than an exhaustive list of state transitions.
For example, if you had a game where the goal was to take turns naming positive integers until someone was forced to name an integer larger than the one their opponent just named (and thus lost), there are an infinite number of states, but the "solution" can be expressed quite simply:
Say "one."
So, at least in this case, the solution is much smaller than the list of all posible cases. I suspect the solution to chess, while much larger than this, is still smaller that you propose.
-- MarkusQ
but I'm sick of this "C for kernel, bloatlang for everything else" BS. I'm writing a ALife sim of language evolution program (for my thesis) in C and I'm thinking about writing agent ai code in *asm*, because compiler generated code won't be fast enough.
I assume you registered at the university instead of attending as an "Anonymous Coward"--otherwise, the diploma isn't going to do you much good.
-- MarkusQ
It's so XP Office compatible it even requires a subscription (~$50/y IIRC) to use it. Now that's what I'd call full compatibility.
-- MarkusQ
Were you perhaps thinking of Napster? Gnutella is about as P2P as it gets; there's no central server, and once two nodes have been introduced (e.g., at least one of them has added the other to its host list) they can reconnect even if everyone else is shut down. Granted, it may take awhile if the original network was >> 2 nodes--but it doesn't take a very large fraction of the network to self-connect within a reasonable time.
-- MarkusQ
I stand corrected. -- MarkusQ
But heat isn't free energy (free in the physics sense, not in the open source sense). True, you can get energy from a difference in temperature, but only by slowing the flow of heat that would have otherwise taken place (just like damming a river) and thus raising the entropy (in this case, temperature). Now, doing so will make your engine run hotter, and thus less efficient, and you have a net loss.
Suppose you do something to cool the heat sink to make up for this. Then you have two cases to consider: either 1) you are using energy to do this, or 2) you have a passive way to do it. In case 1 you are still at a net loss, but in case 2 you might well be doing better than the original system. But you've then changed the base case--if you used the passive cooling trick (a heat sink or whatever) on the original system, you would have gotten a greater gain, so your gizzmo is still costing you.
-- MarkusQ
Example: You put a heat-based gizzmo on your car's exhaust pipe. The temerature (and thus pressure) in the exhaust system goes up, making the engine less efficient and making you use more fuel to go the same distance.
Example: You put one on your CPU. Same deal, except your cooling system now has to work harder to keep it at a reasonable temperature, and thus uses more power.
Example: You wear a swatch. It takes a little bit more energy each time you move your arm. If you want to power a computer the same way, you'll soon be too tired to type.
The key point is in every case you will have to put more energy in than you get back out. That's why perpetual motion machines do not and can not work.
-- MarkusQ
Yes! In fact, why not have many fuzzy hash functions floating around at once? That way, their task would be to come up with something that yielded a different hash against all of the hash functions at once, a much harder problem. If some spammer figures out a way to do it, an anti-spammer can devise a function (looking at lots of copies of the spam, which shouldn't be hard to come by) that would catch it, and now that trick won't work any more.
Distributing the functions with the hash (with a few safe guards, e.g. re: the halting problem) would make this darned near imposible to beat.
-- MarkusQ