Why can't someone honeypot a bot, move the system time forward and intercept NTP queries, and watch the traffic to see what DNS queries it generates? [Sorry for the bad grammar, grammar nazis need not reply]
In Britain, waiting in line is called being in queue. What's a non-verbal queue, then? A line where nobody speaks? Sounds good! When I butt in line, nobody will say anything:)
Oh wait... they'd just punch me. Never mind, then.
Hasn't MIT been working on projects like this for over a decade? I heard that they had a robot that was very convincing of its emotions to others.
Unless the terms are unconscionable (extremely unreasonable, very overly broad, effectively forced upon you, conflicting with irrevocable rights, etc). But it takes a lot to convince a judge that terms are unconscionable unless it is immediately obvious, and companies almost always have more money to spend on good lawyers than you do. Of course, a good judge might see the relationship between imbalance of bargaining power when the contract was negotiated and lack of imbalance of legal resources in the courtroom, and throw out requests from a company's high-paid lawyers that would give them a ridiculously unfair advantage, or find subtle, appropriate, and legal ways to advise you as to options you could pursue that the other lawyers are bound to know about and that you and your attorney might not know about.
First, I was wondering if fish ought to file a class-action lawsuit if the inventor tries to patent it. After seeing this, though, it might be more effective to bring a lawsuit by resurrecting Davinci as a zombie.
Of course, I'm wondering if they can import the contents of the Mizar electronic library for tracking the contents of the Journal of Formalized Mathematics (http://www.cs.ualberta.ca/~piotr/Mizar/mirror/http/library/), which would be pretty cool. The big questions are where the wiki and journal both stand on the Axiom of Choice. Some of the corollaries to the axiom of choice that show up in model theory (which has some interesting things to say about the dimension of the real line versus aleph numbers, beta numbers, and omega numbers) have some very odd consequences and can cause problems when trying to create a formal system that relies on the axiom. Reduced forms of the axiom have been proposed that are less prone to paradoxes, but there's a huge body of formal proof work based on the axiom that would have to be re-worked to use a different set of axioms.
Oh, I know what you mean. I fabricated this neat thing that was like a MOSFET array on a Type I superconductor, using some ideas from core memory to handle the reading/writing of the B-fields. I couldn't afford enough liquid He to run it for very long...:(
But then I got this great idea of poking a little in one of the sector 34 pipelines at the LHC and draining off some of theirs into a thin-film-silvered (for >98% FIR reflectivity) vacuum flask. I don't think anyone noticed, though.
*grin*
I got some hundreds-of-gigabit throughput on a single (though thousands of bits wide) bus that would make Sun Enterprise and SGI systems wet their pants and even make Crays and IBM z/90's break out into a cold sweat. Now if only I could find a RAID system with dozens of 10Gb-ethernet (overclocked, if I can figure out a way to use triaxial shielding techniques to make cables with a twisted-pair CAT6e electrical spec) jacks that could write to a vast array of 10K Ultra-640 SCSI drives (differential bus overclocked, of course). I don't think I'm going to be able to get 32 Hyper8-transports in a cross-bar interconnect, nor find fast enough PCIExpress controllers for the computer's network connection. Oh, crap, wait, I can just fabricate my own controllers on a Type II superconductor substrate and cool them with Liquid N2! DUH!
Hey, does anyone know if ZFS can keep up with media-partition write speed to XFS (I think I would hit some limits on a 64-bit fs pretty quick)? Also, what routers can manage the amount of fiber I'd need to keep the system busy? There's a few thousand uncompressed DVD's on BitTorrent I'd like to download, as well as the complete record of all full-resolution artificial satellite and astronomical telescope and probe data that is publicly available, and the complete edit history of Wikipedia. (Bonus points for anyone who wants to convert that to Libraries-of-Congress!)
I'm slightly impressed by the automatically-scaled logarithmic plot. Any and every dumb chart-producing app can auto-scale a linear scale correctly, but, check it yourself, int(e(n*(l(9000)/10))) [n=0..10] is correct (at least at first glance to someone who's more interested in his tomato soup at the moment).
And, as people above have stated, it's not the bandwidth to the 'net that I'd want at an *{info,secur,...}*con, it's the inter-attendee (and attendee-presenter) bandwidth.
Having a phone is nearly a necessity for the modern world (how do you call out sick when you can't place calls?), and often a landline just does not cut it (need to be locatable 24/7?), unless you want to have both a landline and a pager (plus, what good is a pager when fortress phones are getting harder and harder to find?).
I think there should be some legal restrictions to discourage the excessive use of "fine print." (For example: All wording on a contract must be legible to someone with 20/20 vision at a distance of two feet, or the contract is not enforceable.) A silly law? Well, yes, but when corporations choose to behave in a silly way, the only way to try to maintain some semblance of fairness is usually through silly laws.
(While we're on the subject, how about requiring contracts to be at a 12th-grade reading level or below to be enforceable? "Use a dictionary," "ask your English-major friend for clarification," you say? I say, "Oh, miss Cingular rep handing me something to sign, go fetch a dictionary for me, pronto. And, while you're away from your desk, I'm going to borrow your phone to call an English major I know." On a computer, you might have a dictionary at your disposal 24/7---in real life, you should not be expected to carry one for day-to-day affairs if you have a high-school level of education!)
If the legal term for signing a contract is going to continue to be "contract negotiation," rather than "bending over and letting a corporation stick it to you without any lube," I would like more emphasis on the *negotiation*.
"Okay, miss Cingular rep. Now, I'll sign your paper if you'll sign mine saying that, before you charge me more than 50% above my minimum monthly pay ment, someone from Cingular is expected to receive confirmation from me *in writing* that I acknowledge and am fully aware of the additional charges. See, this is called NEGOTIATION. You have your terms, I have mine."
Oh, attn large corporations: you would not have to re-invent yourselves as "the new ____" (ex. the new AT&T) if people felt they could trust "the old ____." Let your PR branch drool over that for a while, eh?
[Disclaimer: I am very careful about signing contracts, and have been known to (inadvertently) annoy whoever is asking me to sign something for hours and hours on end until they clarify every detail of the fine print. I have not yet reached the point of asking if I can tape record their explanations of the fine print, but that's just because I have not yet been forced to that point! *g*]
"Most of research is simply discovering new problems for others to solve." -- A very important point. In fact, research that uncovers a problem but not a solution is an exciting opportunity for all related researchers in the field, because there is now one more problem to study.
"Nobody told me I had to find a solution as well." Hmm. If you look at research into pure math, it would be an unfortunate situation indeed if you could not publish until you had worked out a complete solution. Consider the age-old problem of the distribution of the primes...
unrelated notes: "Parents came to visit and brought a full bottle of single-malt whiskey" -- umm, so, when are your parents going to visit my apartment? give me an e-mail address and I'll send a list of what I like;) "Classical gas simplified for beginners [youtube.com]" -- What kind of 'classical gas' are we talking about? You know, even 'ideal gas' has pitfalls as an ambiguous phrase...
Ha! I bet you don't get that six times on the:x8's!
Of course, for people who haven't seen TheWeatherChannel, you have to wonder what the remainder of the time is spent broadcasting (and, no, it's not a bunch of people in a bar painted like cold and hot fronts colliding, as some commercials might lead you to believe).
My dad calls it 'weather porn', and I think that's a rather accurate description. If your inner geek has its mouth watering looking at pictures of Flourinert-immersed Crays, your inner weathergeek will be glued to the sofa for the broadcasts of houses and businesses getting ripped apart in Hurricane Alley. Yes, my friend, if you thought goatse leaves a lasting impression, if you just haven't erased 2girls1cup from your mind, if you still have nightmares of tubgirl... call your local cable provider and find out what channel is TWC!
-- oh, obligatory ontopic comment: someone posted about IBM not being able to deliver in-time and within-budget. Well, duh, but the sultans of the mainframe were still making progress, no? Hire a few of the ex-Multics Stratus VOS guys if you've already burned the bridges with IBM---they know reliability pretty well, too. (Until recently, I would have advocated more for IBM due to their strong endorsement of open-source, but, now that Bull-France has given us a license to Multics source, I'm happy to say nice things about Stratus, because I know some of the ex-Multicians who work there were the ones pushing Bull to make sure that everything was followed through to get Multics freed (although it did take 5 or 6 more years than we [alt.os.multics folk] were all hoping for, but, when the weight of {making absolutely sure that no other potential Multics IP owners would turn around and sue Bull} fell on just one lone employee who had a few hours of spare time every now and then to work on the Multics-going-open-source project, I'd say the time it took was actually quite good! Hey, who enjoyed my run-on sentence? *g*)
Seriously. If you want to please a woman, (1) try ribbed condoms, (2) work on your stroke, (3) work on overall intercourse techniques, (4) experiment more with foreplay and sex play, (5) remember that the brain is the strongest sex organ -- what can you say to enhance the experience?, and (6) remember that the sensation of touch is what excites most women the most (whereas most men are excited the most by visual stimulation).
Man, I should develop on these ideas and sell them in a book. I bet I'd have more happy customers than Enzyte and great word-of-mouth advertising (cheaper and often more effective than commercials).
ISO Wakes from 350-Day Drinking Binge, Wonders What the Hell It Just Approved (from the oh-man-what-a-hangover dept.)
ISO and IEC have just woken up from a 350-day drinking binge during which Microsoft footed the bill. ``It's great to be drinking with pals like ISO and IEC. Our bartenders told about 6,500 pages worth of stories, and they lapped up everything,'' stated a Microsoft spokesman. ISO and IEC simplied replied with, ``Wow, I've never seen such a liquor cabinet! Wait, did we approve something last night? Microsoft kept asking if we wanted another round of shots, and then if we wanted another round of word-wrapping styles. I can't remember what we were saying yes to... owww, my head...''
ISO and IEC lamented that there were ``these four dudes trying to crash the party'' who kept insisting that they stop drinking, sober up, and pay attention to that to which they were agreeing.
An anonymous Microsoft programmer wrote in: ``People have always just thought it was code and feature bloat, but the higher-ups have been strategizing about this for years. Nobody else can compete with a programming team of this size. Management always says more code is better... right?''
What are you saying, that when I eat a dessert I give off methane gas for a few hours afterward, much to the dismay of the roommates, who do their best to discourage me from eating so many dairy confections when they're around?
No stinky overnight garbage bags? Oh, the solution to that is to take the half-full big black bag out at the end of the day and leave it on the lawn of one of the responsible lawmakers. Then, in the morning, you can bring it back in and use it, unless of course the lawmaker's lawn now has a swarm of flies, or ripped up plastic and scattered garbage from a hungry animal visitor in the night---in which case, it definitely wasn't a waste of a bag, it was well worth it;)
Oops! I misread the spec! I thought when you called to check for features, you passed in '0' in %eax and checked the return string for the letter 'n' in character offsets 2 and 6. D'oh! - anon FutureMark developer
Pfff. There's nothing that a large capacitor bank or stored-inertia motor-generator pair can't do. Who needs some tiny little solid-state UPS? If you run over it with a truck, it stops working! And, forget this DIMM memory. A large cabinet of core memory is not only economical and practical (keeps your server room warm, also functions as an EMF detector and cosmic ray detector---four uses in one!) but is inherently non-volatile and stylish to boot. Lost power so long that your stored energy systems got exhausted? No need to fear, thanks to magnetic hysterisis technology, core memory deteriorates quite slowly when kept in good field isolation!
No, you mean slackware. My 386/33MHz still has just 4MiB of RAM and a Hercules card for video, not VGA, you insensitive clod! I hope I get the 387 chip for Christmas this year, though.
Unfortunately, Congress has the power to 'regulate' Interstate commerce in just about any way they see fit. Of course, any matter of national policy/might/ have an effect on Interstate commerce, so they mostly get to do what they like.
A much better stipulation would be, "As this Bill pertains unequivocally and indubitably to national interests *strictly necessary* to Interstate commerce and is furthermore restricted in action to minimal involvement necessary unless regularly reviewed and granted further extension of power and influence by a [large] margin of the States,... (insert the rest of the text of the Bill here)"
It's more difficult to build something spherical than something cubic with nice right angles. Hence, the Borg ship is a cube, but, they still get the structural efficiency/stability of a sphere by circumscribing the ship in an imaginary sphere, then assimilating all the unfilled space in the sphere.
Seriously, though: since the Universe is only 6,000 years old, we know that any Earth "debris" on the Moon was just put there as a test of faith.;)
DC power transmission over short distances is feasible. Over long distances, it isn't. Look up the Current Wars and AC power distribution. For DC, P=I^2*R=V^2/R. For AC, Prms=Irms^2*Z=Vrms^2/Z. Ignoring phase shift and comparing RMS AC quantities with DC quantities, the equations look the same. The longer the lines, the larger R, so the larger your power lost to heat. DC-DC power conversion is a modern solid-state technology (using charge pumps?) and still tends to be expensive, intolerant of transients (without proper filtering), and limited to small voltages and/or currents. AC-AC power conversion is simple, cheap, and can handle huge currents and voltages. Also, it's much harder to go from DC to AC than the other way around. Until modern solid state, in fact, there was no reliable, efficient way to convert DC to AC in any significant quantity.
If you had a shed and a house on two sides of a large property, and you wanted to put solar panels on one and bring some power to the other (perhaps it is in the shade), inverter + step-up transformer + step-down transformer + AC-to-DC is going to have a noticeable improvement in efficiency over trying to carry DC long distances. If you're generating any serious amount of power, you're also going to need some thick, thick cables to carry lo-volt hi-amp DC around in order to safely dissipate the heat, especially for wires running indoors. Even if you don't care about losses, converting to/from AC is much cheaper than replacing everything in a burned-down house.
Depending on the size of your organization, and how "corporate" the network is (are workstations in software lock-down?), you may have to spend a lot of time designing software that can ensure some level of code authenticity for any deployed work. Otherwise, don't expect to get approval to run this on "most every" random workstation. I had a project where machines would automatically download code and data sets to run as requested, synchronized by a central server. Three things stood in my way: (a) limitations in the development environment made debugging difficult (MFC is such a buggy piece of crap!), (b) eventually the machines that originally took hours to run the simulations were upgraded to bleeding-edge systems that ran them in the duration of a coffee break, meaning I would have had to do some serious research into tight, fast synchronization and networking code in order to gain any improvements by using a cluster due to the nature of the data blocks (they were very small, and before a new set could be deployed, all results of the last set had to be tallied); another approach could have been to redesign the algorithm to let part of the network move on to a new data set while everything was tallied (but depending on the tally, some results would have to be thrown out), and (c) since the workstations were starting to go to software lock-down during the latest refresh, a lot of additional time would have to be spent developing some very high guarantees of security (my code did md5summing of all code modules, including the client itself, but someone could have potentially written a hacked client that returned "the right" md5sums when asked while running unsafe code, so a different system would have had to be designed).
Sorry for the run-ons, I'm a little busy/distracted recovering data on a server with failed disks:( I was *just about* to upgrade it to mirroring RAID, too (at least I did regular backups over the network, though).
I am not a physicist, but it seems to me that this experiment could raise some interesting possibilities, multiple ones of which could be simultaneously true. (1) Quantization is much more complex than thought, occurs on multiple levels, relies on several decoherence phenomena, and Schroedinger's equation is a special case of something else. (2) Special relativity is quantized and has its own decoherence, and is also a special case of some larger system. (3) Quantum statistics do rely on hidden variables, and these hidden variables are so numerous as to be immeasurable and have their own special behavior involving a decoherence phenomenon that is different from previously observed forms of decoherence. (4) The notion of information, as we understand it, is a special case of something much more complicated. (5) Quantized spin is a special case of something else not before seen in experiments. (6+) Other things.
No matter what, it should yield some new ideas and possible ways to investigate ToEs, GUTs, and high-energy physics orders of magnitude larger than anything we have any idea about to date.
What's interesting to me is that, the whole time, the prosecution was playing these dramatic court-room antics (maybe that's standard in capital murder, I don't know) about how obviously guilty the defendant was. Now that guilt is presumed by verdict, they're bartering with the defendant for information, which is a dead giveaway that the prosecution wasn't really on sure footing. They apparently *need* a body (and all the details about how the murder was executed therein entailed) to make things tidy from an investigative standpoint, or why bother cutting the defendant a deal, especially if the criminal act is as egregious as the prosecution always made it out to be? Emotional closure for the children would be a nice added benefit, but I doubt that would be the primary motivation in doing this kind of thing. Of course, all of this is going on the premise that the prosecution really is solidly interested in finding a body, and that the defendant isn't bluffing or playing mind games. Side-note: it doesn't really matter whether this came about from the defendant offering a body in exchange for lessened charges or the prosecution offering lessened charges in hopes of getting more information; the key point is, either way, the prosecution is heavily interested in the body, or they wouldn't bother, since it otherwise wouldn't add much to the record if it were to be accidentally exhumed (for example, in land development).
On a less serious note: always keeping an ace up the sleeve with a missing body... maybe that's the reason why reiserfs doesn't have a/lost+found?
I'm guessing that you're Louis Slavain. You say that the computing industry needs to listen to you, but you're hard on the ears. That's not going to warrant you much success. Academia will be much more likely to discuss your proposals if you patiently listen to criticism and defend against it without engaging in personal attacks. Don't respond to the laughter or harsh words of others with harsh words of your own; professionals will not respect you unless you can maintain professionalism even when your opponents don't. You can support your ideas intensely without being uncomfortably intense on your audience. If you channel your personal drive into productivity rather than ranting, you will find followers much more quickly.
You do have some interesting ideas. I think that many problems in computing can be answered by some of the reactive models that you are developing for fine-grained parallelism, but the days of algorithms and functions are not numbered. You might also want to investigate large-scale asynchronous systems whose individual modules are only locally synchronous; to me, it seems like some of your ideas can be compared to the way that nature works on the scale of chemical reactions and particle interactions, which might be synchronized locally on very small scales (e.g. "Planck time") and exhibit some large-scale emergent synchronization but don't indicate any general dependency on nor tendency toward large-scale synchronization.
Yes, time "travel" seems an unlikely possibility. I would suggest you study the philosophical writings of Peter van Inwagen; he has written a number of responses to the ideas presented by David Lewis about the nature of time and relativity.
I would also encourage you to implement more examples of COSA systems that achieve useful goals, then start writing introductions to COSA for people coming from the algorithmic and functional points of view. In actuality, COSA may have more in common with functions (although not necessarily Functional Programming as it is generally taught) than you think. Have you ever conducted an extensive survey of the difference in approach in Scheme versus Lisp? Lambda functions and tail-recursion can often be easily optimized to collapse complex, nested Scheme code into a straightforward iterative sequence; in a similar way, similar transformations may show some classes of functions to be readily changeable to COSA systems. Mature Lisp systems often exploit modularity to automatically implement some coarse-grained parallelism, which is another interesting transformation worth examining.
Slashdot Mirror
Why can't someone honeypot a bot, move the system time forward and intercept NTP queries, and watch the traffic to see what DNS queries it generates?
[Sorry for the bad grammar, grammar nazis need not reply]
In Britain, waiting in line is called being in queue. What's a non-verbal queue, then? A line where nobody speaks? Sounds good! When I butt in line, nobody will say anything :)
Oh wait... they'd just punch me. Never mind, then.
Hasn't MIT been working on projects like this for over a decade? I heard that they had a robot that was very convincing of its emotions to others.
Unless the terms are unconscionable (extremely unreasonable, very overly broad, effectively forced upon you, conflicting with irrevocable rights, etc). But it takes a lot to convince a judge that terms are unconscionable unless it is immediately obvious, and companies almost always have more money to spend on good lawyers than you do. Of course, a good judge might see the relationship between imbalance of bargaining power when the contract was negotiated and lack of imbalance of legal resources in the courtroom, and throw out requests from a company's high-paid lawyers that would give them a ridiculously unfair advantage, or find subtle, appropriate, and legal ways to advise you as to options you could pursue that the other lawyers are bound to know about and that you and your attorney might not know about.
First, I was wondering if fish ought to file a class-action lawsuit if the inventor tries to patent it. After seeing this, though, it might be more effective to bring a lawsuit by resurrecting Davinci as a zombie.
Of course, I'm wondering if they can import the contents of the Mizar electronic library for tracking the contents of the Journal of Formalized Mathematics (http://www.cs.ualberta.ca/~piotr/Mizar/mirror/http/library/), which would be pretty cool. The big questions are where the wiki and journal both stand on the Axiom of Choice. Some of the corollaries to the axiom of choice that show up in model theory (which has some interesting things to say about the dimension of the real line versus aleph numbers, beta numbers, and omega numbers) have some very odd consequences and can cause problems when trying to create a formal system that relies on the axiom. Reduced forms of the axiom have been proposed that are less prone to paradoxes, but there's a huge body of formal proof work based on the axiom that would have to be re-worked to use a different set of axioms.
Oh, I know what you mean. I fabricated this neat thing that was like a MOSFET array on a Type I superconductor, using some ideas from core memory to handle the reading/writing of the B-fields. I couldn't afford enough liquid He to run it for very long... :(
But then I got this great idea of poking a little in one of the sector 34 pipelines at the LHC and draining off some of theirs into a thin-film-silvered (for >98% FIR reflectivity) vacuum flask. I don't think anyone noticed, though.
*grin*
I got some hundreds-of-gigabit throughput on a single (though thousands of bits wide) bus that would make Sun Enterprise and SGI systems wet their pants and even make Crays and IBM z/90's break out into a cold sweat. Now if only I could find a RAID system with dozens of 10Gb-ethernet (overclocked, if I can figure out a way to use triaxial shielding techniques to make cables with a twisted-pair CAT6e electrical spec) jacks that could write to a vast array of 10K Ultra-640 SCSI drives (differential bus overclocked, of course). I don't think I'm going to be able to get 32 Hyper8-transports in a cross-bar interconnect, nor find fast enough PCIExpress controllers for the computer's network connection. Oh, crap, wait, I can just fabricate my own controllers on a Type II superconductor substrate and cool them with Liquid N2! DUH!
Hey, does anyone know if ZFS can keep up with media-partition write speed to XFS (I think I would hit some limits on a 64-bit fs pretty quick)? Also, what routers can manage the amount of fiber I'd need to keep the system busy? There's a few thousand uncompressed DVD's on BitTorrent I'd like to download, as well as the complete record of all full-resolution artificial satellite and astronomical telescope and probe data that is publicly available, and the complete edit history of Wikipedia. (Bonus points for anyone who wants to convert that to Libraries-of-Congress!)
I'm slightly impressed by the automatically-scaled logarithmic plot. Any and every dumb chart-producing app can auto-scale a linear scale correctly, but, check it yourself, int(e(n*(l(9000)/10))) [n=0..10] is correct (at least at first glance to someone who's more interested in his tomato soup at the moment).
And, as people above have stated, it's not the bandwidth to the 'net that I'd want at an *{info,secur,...}*con, it's the inter-attendee (and attendee-presenter) bandwidth.
Having a phone is nearly a necessity for the modern world (how do you call out sick when you can't place calls?), and often a landline just does not cut it (need to be locatable 24/7?), unless you want to have both a landline and a pager (plus, what good is a pager when fortress phones are getting harder and harder to find?).
I think there should be some legal restrictions to discourage the excessive use of "fine print." (For example: All wording on a contract must be legible to someone with 20/20 vision at a distance of two feet, or the contract is not enforceable.) A silly law? Well, yes, but when corporations choose to behave in a silly way, the only way to try to maintain some semblance of fairness is usually through silly laws.
(While we're on the subject, how about requiring contracts to be at a 12th-grade reading level or below to be enforceable? "Use a dictionary," "ask your English-major friend for clarification," you say? I say, "Oh, miss Cingular rep handing me something to sign, go fetch a dictionary for me, pronto. And, while you're away from your desk, I'm going to borrow your phone to call an English major I know." On a computer, you might have a dictionary at your disposal 24/7---in real life, you should not be expected to carry one for day-to-day affairs if you have a high-school level of education!)
If the legal term for signing a contract is going to continue to be "contract negotiation," rather than "bending over and letting a corporation stick it to you without any lube," I would like more emphasis on the *negotiation*.
"Okay, miss Cingular rep. Now, I'll sign your paper if you'll sign mine saying that, before you charge me more than 50% above my minimum monthly pay ment, someone from Cingular is expected to receive confirmation from me *in writing* that I acknowledge and am fully aware of the additional charges. See, this is called NEGOTIATION. You have your terms, I have mine."
Oh, attn large corporations: you would not have to re-invent yourselves as "the new ____" (ex. the new AT&T) if people felt they could trust "the old ____." Let your PR branch drool over that for a while, eh?
[Disclaimer: I am very careful about signing contracts, and have been known to (inadvertently) annoy whoever is asking me to sign something for hours and hours on end until they clarify every detail of the fine print. I have not yet reached the point of asking if I can tape record their explanations of the fine print, but that's just because I have not yet been forced to that point! *g*]
"Most of research is simply discovering new problems for others to solve." -- A very important point. In fact, research that uncovers a problem but not a solution is an exciting opportunity for all related researchers in the field, because there is now one more problem to study.
"Nobody told me I had to find a solution as well." Hmm. If you look at research into pure math, it would be an unfortunate situation indeed if you could not publish until you had worked out a complete solution. Consider the age-old problem of the distribution of the primes...
unrelated notes: ;)
"Parents came to visit and brought a full bottle of single-malt whiskey" -- umm, so, when are your parents going to visit my apartment? give me an e-mail address and I'll send a list of what I like
"Classical gas simplified for beginners [youtube.com]" -- What kind of 'classical gas' are we talking about? You know, even 'ideal gas' has pitfalls as an ambiguous phrase...
Ha! I bet you don't get that six times on the :x8's!
Of course, for people who haven't seen TheWeatherChannel, you have to wonder what the remainder of the time is spent broadcasting (and, no, it's not a bunch of people in a bar painted like cold and hot fronts colliding, as some commercials might lead you to believe).
My dad calls it 'weather porn', and I think that's a rather accurate description. If your inner geek has its mouth watering looking at pictures of Flourinert-immersed Crays, your inner weathergeek will be glued to the sofa for the broadcasts of houses and businesses getting ripped apart in Hurricane Alley. Yes, my friend, if you thought goatse leaves a lasting impression, if you just haven't erased 2girls1cup from your mind, if you still have nightmares of tubgirl... call your local cable provider and find out what channel is TWC!
-- oh, obligatory ontopic comment: someone posted about IBM not being able to deliver in-time and within-budget. Well, duh, but the sultans of the mainframe were still making progress, no? Hire a few of the ex-Multics Stratus VOS guys if you've already burned the bridges with IBM---they know reliability pretty well, too. (Until recently, I would have advocated more for IBM due to their strong endorsement of open-source, but, now that Bull-France has given us a license to Multics source, I'm happy to say nice things about Stratus, because I know some of the ex-Multicians who work there were the ones pushing Bull to make sure that everything was followed through to get Multics freed (although it did take 5 or 6 more years than we [alt.os.multics folk] were all hoping for, but, when the weight of {making absolutely sure that no other potential Multics IP owners would turn around and sue Bull} fell on just one lone employee who had a few hours of spare time every now and then to work on the Multics-going-open-source project, I'd say the time it took was actually quite good! Hey, who enjoyed my run-on sentence? *g*)
Seriously. If you want to please a woman, (1) try ribbed condoms, (2) work on your stroke, (3) work on overall intercourse techniques, (4) experiment more with foreplay and sex play, (5) remember that the brain is the strongest sex organ -- what can you say to enhance the experience?, and (6) remember that the sensation of touch is what excites most women the most (whereas most men are excited the most by visual stimulation).
Man, I should develop on these ideas and sell them in a book. I bet I'd have more happy customers than Enzyte and great word-of-mouth advertising (cheaper and often more effective than commercials).
On the issue of who gets to decide what 'spam' to drop: Agreed, totally!!
-Mike, of Viagra Resellers, Niger national division
ISO Wakes from 350-Day Drinking Binge, Wonders What the Hell It Just Approved
(from the oh-man-what-a-hangover dept.)
ISO and IEC have just woken up from a 350-day drinking binge during which Microsoft footed the bill. ``It's great to be drinking with pals like ISO and IEC. Our bartenders told about 6,500 pages worth of stories, and they lapped up everything,'' stated a Microsoft spokesman. ISO and IEC simplied replied with, ``Wow, I've never seen such a liquor cabinet! Wait, did we approve something last night? Microsoft kept asking if we wanted another round of shots, and then if we wanted another round of word-wrapping styles. I can't remember what we were saying yes to... owww, my head...''
ISO and IEC lamented that there were ``these four dudes trying to crash the party'' who kept insisting that they stop drinking, sober up, and pay attention to that to which they were agreeing.
An anonymous Microsoft programmer wrote in: ``People have always just thought it was code and feature bloat, but the higher-ups have been strategizing about this for years. Nobody else can compete with a programming team of this size. Management always says more code is better... right?''
What are you saying, that when I eat a dessert I give off methane gas for a few hours afterward, much to the dismay of the roommates, who do their best to discourage me from eating so many dairy confections when they're around?
No stinky overnight garbage bags? Oh, the solution to that is to take the half-full big black bag out at the end of the day and leave it on the lawn of one of the responsible lawmakers. Then, in the morning, you can bring it back in and use it, unless of course the lawmaker's lawn now has a swarm of flies, or ripped up plastic and scattered garbage from a hungry animal visitor in the night---in which case, it definitely wasn't a waste of a bag, it was well worth it ;)
Oops! I misread the spec! I thought when you called to check for features, you passed in '0' in %eax and checked the return string for the letter 'n' in character offsets 2 and 6. D'oh!
- anon FutureMark developer
Pfff. There's nothing that a large capacitor bank or stored-inertia motor-generator pair can't do. Who needs some tiny little solid-state UPS? If you run over it with a truck, it stops working! And, forget this DIMM memory. A large cabinet of core memory is not only economical and practical (keeps your server room warm, also functions as an EMF detector and cosmic ray detector---four uses in one!) but is inherently non-volatile and stylish to boot. Lost power so long that your stored energy systems got exhausted? No need to fear, thanks to magnetic hysterisis technology, core memory deteriorates quite slowly when kept in good field isolation!
No, you mean slackware. My 386/33MHz still has just 4MiB of RAM and a Hercules card for video, not VGA, you insensitive clod! I hope I get the 387 chip for Christmas this year, though.
Unfortunately, Congress has the power to 'regulate' Interstate commerce in just about any way they see fit. Of course, any matter of national policy /might/ have an effect on Interstate commerce, so they mostly get to do what they like.
A much better stipulation would be, "As this Bill pertains unequivocally and indubitably to national interests *strictly necessary* to Interstate commerce and is furthermore restricted in action to minimal involvement necessary unless regularly reviewed and granted further extension of power and influence by a [large] margin of the States, ... (insert the rest of the text of the Bill here)"
It's more difficult to build something spherical than something cubic with nice right angles. Hence, the Borg ship is a cube, but, they still get the structural efficiency/stability of a sphere by circumscribing the ship in an imaginary sphere, then assimilating all the unfilled space in the sphere.
Seriously, though: since the Universe is only 6,000 years old, we know that any Earth "debris" on the Moon was just put there as a test of faith. ;)
DC power transmission over short distances is feasible. Over long distances, it isn't. Look up the Current Wars and AC power distribution. For DC, P=I^2*R=V^2/R. For AC, Prms=Irms^2*Z=Vrms^2/Z. Ignoring phase shift and comparing RMS AC quantities with DC quantities, the equations look the same. The longer the lines, the larger R, so the larger your power lost to heat. DC-DC power conversion is a modern solid-state technology (using charge pumps?) and still tends to be expensive, intolerant of transients (without proper filtering), and limited to small voltages and/or currents. AC-AC power conversion is simple, cheap, and can handle huge currents and voltages. Also, it's much harder to go from DC to AC than the other way around. Until modern solid state, in fact, there was no reliable, efficient way to convert DC to AC in any significant quantity.
If you had a shed and a house on two sides of a large property, and you wanted to put solar panels on one and bring some power to the other (perhaps it is in the shade), inverter + step-up transformer + step-down transformer + AC-to-DC is going to have a noticeable improvement in efficiency over trying to carry DC long distances. If you're generating any serious amount of power, you're also going to need some thick, thick cables to carry lo-volt hi-amp DC around in order to safely dissipate the heat, especially for wires running indoors. Even if you don't care about losses, converting to/from AC is much cheaper than replacing everything in a burned-down house.
Depending on the size of your organization, and how "corporate" the network is (are workstations in software lock-down?), you may have to spend a lot of time designing software that can ensure some level of code authenticity for any deployed work. Otherwise, don't expect to get approval to run this on "most every" random workstation. I had a project where machines would automatically download code and data sets to run as requested, synchronized by a central server. Three things stood in my way: (a) limitations in the development environment made debugging difficult (MFC is such a buggy piece of crap!), (b) eventually the machines that originally took hours to run the simulations were upgraded to bleeding-edge systems that ran them in the duration of a coffee break, meaning I would have had to do some serious research into tight, fast synchronization and networking code in order to gain any improvements by using a cluster due to the nature of the data blocks (they were very small, and before a new set could be deployed, all results of the last set had to be tallied); another approach could have been to redesign the algorithm to let part of the network move on to a new data set while everything was tallied (but depending on the tally, some results would have to be thrown out), and (c) since the workstations were starting to go to software lock-down during the latest refresh, a lot of additional time would have to be spent developing some very high guarantees of security (my code did md5summing of all code modules, including the client itself, but someone could have potentially written a hacked client that returned "the right" md5sums when asked while running unsafe code, so a different system would have had to be designed).
Sorry for the run-ons, I'm a little busy/distracted recovering data on a server with failed disks :( I was *just about* to upgrade it to mirroring RAID, too (at least I did regular backups over the network, though).
I am not a physicist, but it seems to me that this experiment could raise some interesting possibilities, multiple ones of which could be simultaneously true. (1) Quantization is much more complex than thought, occurs on multiple levels, relies on several decoherence phenomena, and Schroedinger's equation is a special case of something else. (2) Special relativity is quantized and has its own decoherence, and is also a special case of some larger system. (3) Quantum statistics do rely on hidden variables, and these hidden variables are so numerous as to be immeasurable and have their own special behavior involving a decoherence phenomenon that is different from previously observed forms of decoherence. (4) The notion of information, as we understand it, is a special case of something much more complicated. (5) Quantized spin is a special case of something else not before seen in experiments. (6+) Other things.
No matter what, it should yield some new ideas and possible ways to investigate ToEs, GUTs, and high-energy physics orders of magnitude larger than anything we have any idea about to date.
What's interesting to me is that, the whole time, the prosecution was playing these dramatic court-room antics (maybe that's standard in capital murder, I don't know) about how obviously guilty the defendant was. Now that guilt is presumed by verdict, they're bartering with the defendant for information, which is a dead giveaway that the prosecution wasn't really on sure footing. They apparently *need* a body (and all the details about how the murder was executed therein entailed) to make things tidy from an investigative standpoint, or why bother cutting the defendant a deal, especially if the criminal act is as egregious as the prosecution always made it out to be? Emotional closure for the children would be a nice added benefit, but I doubt that would be the primary motivation in doing this kind of thing. Of course, all of this is going on the premise that the prosecution really is solidly interested in finding a body, and that the defendant isn't bluffing or playing mind games. Side-note: it doesn't really matter whether this came about from the defendant offering a body in exchange for lessened charges or the prosecution offering lessened charges in hopes of getting more information; the key point is, either way, the prosecution is heavily interested in the body, or they wouldn't bother, since it otherwise wouldn't add much to the record if it were to be accidentally exhumed (for example, in land development).
/lost+found?
On a less serious note: always keeping an ace up the sleeve with a missing body... maybe that's the reason why reiserfs doesn't have a
I'm guessing that you're Louis Slavain. You say that the computing industry needs to listen to you, but you're hard on the ears. That's not going to warrant you much success. Academia will be much more likely to discuss your proposals if you patiently listen to criticism and defend against it without engaging in personal attacks. Don't respond to the laughter or harsh words of others with harsh words of your own; professionals will not respect you unless you can maintain professionalism even when your opponents don't. You can support your ideas intensely without being uncomfortably intense on your audience. If you channel your personal drive into productivity rather than ranting, you will find followers much more quickly.
You do have some interesting ideas. I think that many problems in computing can be answered by some of the reactive models that you are developing for fine-grained parallelism, but the days of algorithms and functions are not numbered. You might also want to investigate large-scale asynchronous systems whose individual modules are only locally synchronous; to me, it seems like some of your ideas can be compared to the way that nature works on the scale of chemical reactions and particle interactions, which might be synchronized locally on very small scales (e.g. "Planck time") and exhibit some large-scale emergent synchronization but don't indicate any general dependency on nor tendency toward large-scale synchronization.
Yes, time "travel" seems an unlikely possibility. I would suggest you study the philosophical writings of Peter van Inwagen; he has written a number of responses to the ideas presented by David Lewis about the nature of time and relativity.
I would also encourage you to implement more examples of COSA systems that achieve useful goals, then start writing introductions to COSA for people coming from the algorithmic and functional points of view. In actuality, COSA may have more in common with functions (although not necessarily Functional Programming as it is generally taught) than you think. Have you ever conducted an extensive survey of the difference in approach in Scheme versus Lisp? Lambda functions and tail-recursion can often be easily optimized to collapse complex, nested Scheme code into a straightforward iterative sequence; in a similar way, similar transformations may show some classes of functions to be readily changeable to COSA systems. Mature Lisp systems often exploit modularity to automatically implement some coarse-grained parallelism, which is another interesting transformation worth examining.