That theory makes no testable predictions so it isn't scientific.
Can you think of some observable aspect of reality that would be different depending on whether the universe was designed vs. not designed?
Frankly, the universe looks pretty darn random and chaotic to me. Thrown in some weird stuff like duck billed platypuses and humans all having appendixes for no apparent reason and the most generous conclusion you can come to is that if there was an Intelligent Designer he wasn't very good at it.
It's fine to malloc/free in 16M, just keep a grasp on the situation.
* install a malloc debug library on all test boxes to find all unfreed, double freed, or overwritten chunks. don't ship until all mem is accounted for.
* don't allow, or at least discourage, "little" mallocs like strdup() or allocation of singleton structs.
* do something sensible when malloc starts returning NULL. don't just seg fault or abort.
* destruction test: malloc 15.5MB (and incr until failure) of space at the start just to establish the limit, then back it down 50K and see how long your embedded app can live in that space and watch how it dies. as you fix things, expect it to die in different places each time.
* nm (or elfdump these days) your c lib and look for library functions that link to malloc or calloc. if your embedded OS ships with source, you can just grep for it.
* if your app has a log file, put a printf(sbrk(0)) in every few minutes (or less as appropriate) so you can watch for unexpected growth over time and spikes related to usage.
I think I was being super-simple minded and super-complex minded at the same time.
You can make a really cheap low power DC->DC regulator out of a zener and an emitter follower. I used to do that for home built projects a long time ago. A lot of stuff isn't picky and that'll work fine, but some stuff (CPUs, memory) *is* picky and needs to be regulated anyway and other stuff (disks, graphics cards) draw so much power you need smart xor bulky components to handle it.
But what I am really thinking of is power that can be delivered as is, then regulated on-board, on-package, or on-chip with some cheap (but smart) silicon and one tiny xformer and one tiny cap. So long as the power distrubition network fans out really fine-grained, the individual regulators can be small, on-chip things that just handle 100mA or less, often far less.
Obviously there are some hurdles to this but you'd save one conversion/regulation step, or at least put into silicon where it's super cheap to manufacture yet can still be very sophisticated.
The small matter of design and implementation is the only problem:-)
Thought you had to be up in the MHz to get much skin effect. At say 30KHz (low end of switching PS frequencies) I'd think it would march right up your arm just like DC.
I suspect there is a gruesome study out there somewhere and I really don't want to see it.
I could still see 30Khz/24V-48V as useful within the data center, even within a server itself. It would simplify wiring and PS count greatly if power could be shipped right down to each board or chip.
At the grid level there is no contest, khz would be impossible plus there is so much installed equipment that it could take $100B to switchover--on the other hand we may be wasting that much $ worth of electricity on line losses and waste heat. I think something like well under half the electricity produced at a typical power plant ends up actually performing useful work on the first order, and several more percent are spent on a/c to evacuate the waste heat from homes and businesses.
You almost seem to be arguing for power distribution to be higher frequency AC, so that the DC conversion componenents can be small, cheap and localized to the point of use. If the HF AC is well chosen for voltage and frequency then some of the intermediate conversion steps can be skipped as it will be "ready" for step down and rectification.
50/60Hz originally was chosen because the low frequencies were easier to generate with the mechanical equipment available 100 years ago but also to avoid a lot of inductive loss into every bit of ferrous metal along the right of way; of course ridiculously high frequencies would actually radiate power away.
I think this could actually work if the power cables were balanced and jacketed at all well, and the voltage was high enough (24v) to carry w/o huge cables. The frequency should probably be whatever switching PS's use these days, ~30Khz?
Also I wonder if there is an increased electrocution risk from HF vs 60Hz.
The unmentioned downside is that ping times via geosynhcronous orbiting sats are awful.
* Gaming is impossible * web surfing is click...wait...wait...wait...display. and wait twice as long if javascript tries to load more images. * FTP and email work fine, as do most file sharing networks. * VOIP has intolerable delays, plus the delays are so long some echo cancellation algorithms give up
I would think that a parabolic and a corner cube with the same cross-sections would have about the same gain and frequency response on-axis. off axis the cc wins by so much that even some on-axis loss would
1Kz ~= 1 foot wavelength so 2ft diameter reflectors should do a nice job of any plain old acoustic countermeasures. however if the beam width is 50 ft at your position you're only going to send back a fraction of a percent of the power anyway, parabolic or not, so unless the entire side of your ship is lined with cc's it's pretty much a waste of time.
focused ultrasonics are a bit more of a problem. the interference frequency will just kind of appear on any surface nearby so you won't be able to reflect that, plus the ultrasonics are likely coming from offset positions so the parabola will not be able to reflect both at the same time, and even if the cc does reflect both it won't result in anything destructive reflecting back since it will be out of focus.
Legal documents filed with a court of law are supposed to be carefully reviewed.
The fact that this document was not reviewed by anyone competent is a sign that the plaintiffs are not taking this case seriously. That's not funny, that's just frustrating to all involved.
Yes, I got one of those (actually, I got the claim form, not the voucher).
Having never actually bought a microsoft product, I was unable to take advantage of the offer. As I recall it maxed out around $300 and to get that you had to have bought odd combinations of things, like MS Office and also Excel separately (which is nearly impossible for average joe users) and a retail copy of windows (pre-installed didn't count).
>> TVs had a warm, satisfying orange glow eminating from >> the rear of the set, and a peculiar smell as well.
And a sound! Older sets had notoriously cheap flyback transformers and if your ears were any good you could hear the 15KHz horizontal sync very clearly. Newer CRT's seal the transformer in epoxy or similar plastics but back then they were often made with paper.
I've always want to try a winter vs. summer solution.
In the winter, just vent the heat towards the room you spend the most time in.
In the summer, take in air in near ground level (if not prone to floods!) and venting it up and out by convection or just out the nearest wall if fan forced. A steady flow of air, even if it's humid and 90F will keep your servers pretty happy so long as the amount of air is sufficient. A muffin fan won't be enough, but a 6'' duct fan should do the trick.
A more "green" solution would be to switch to fewer, smaller, lower powered servers so that you generate less heat in the first place. A modest laptop running linux/bsd/solaris can actually be quite a powerful server and can handle several services at once.
For those few that don't get it, the satellites aren't geostationary, so a failed satellite (once the minimum # is reached) will probably create a dead spot that moves around the planet, in which GPS service is degraded.
I was thinking along the lines of some raised bumps along the painted lines. I have mixed feelings though. If people become to aware of the issue they might get a feeling that they are entitled to the area in front of them. This conflicts with the general presumption that the person in front has the right of way.
>> (why are people so much in a hurry to get to a red light?).
My observation of local drivers is that they do it to prevent other drivers from changing into their lane at the light. There is no law (or no enforced law at least) to keep people from jockeying for position at the red light. The only way to maintain your spot in line is to get to the light first. Once you do that, you're now in a position to consider changing lanes yourself if that puts you closer to the beginning of the queue and you become part of the problem.
My preference would be that drivers should not be allowed to change lanes when approaching a red light unless they're going to turn at the intersection.
Slashdot Mirror
That theory makes no testable predictions so it isn't scientific.
Can you think of some observable aspect of reality that would be different depending on whether the universe was designed vs. not designed?
Frankly, the universe looks pretty darn random and chaotic to me. Thrown in some weird stuff like duck billed platypuses and humans all having appendixes for no apparent reason and the most generous conclusion you can come to is that if there was an Intelligent Designer he wasn't very good at it.
Don't forget that RIM sued several companies over the years, enough to get the nickname "Lawsuits in Motion" on theregister.co.uk.
Live by the sword, die by the sword.
It's fine to malloc/free in 16M, just keep a grasp on the situation.
* install a malloc debug library on all test boxes to find all unfreed, double freed, or overwritten chunks. don't ship until all mem is accounted for.
* don't allow, or at least discourage, "little" mallocs like strdup() or allocation of singleton structs.
* do something sensible when malloc starts returning NULL. don't just seg fault or abort.
* destruction test: malloc 15.5MB (and incr until failure) of space at the start just to establish the limit, then back it down 50K and see how long your embedded app can live in that space and watch how it dies. as you fix things, expect it to die in different places each time.
* nm (or elfdump these days) your c lib and look for library functions that link to malloc or calloc. if your embedded OS ships with source, you can just grep for it.
* if your app has a log file, put a printf(sbrk(0)) in every few minutes (or less as appropriate) so you can watch for unexpected growth over time and spikes related to usage.
just some ideas.
I think I was being super-simple minded and super-complex minded at the same time.
:-)
You can make a really cheap low power DC->DC regulator out of a zener and an emitter follower. I used to do that for home built projects a long time ago. A lot of stuff isn't picky and that'll work fine, but some stuff (CPUs, memory) *is* picky and needs to be regulated anyway and other stuff (disks, graphics cards) draw so much power you need smart xor bulky components to handle it.
But what I am really thinking of is power that can be delivered as is, then regulated on-board, on-package, or on-chip with some cheap (but smart) silicon and one tiny xformer and one tiny cap. So long as the power distrubition network fans out really fine-grained, the individual regulators can be small, on-chip things that just handle 100mA or less, often far less.
Obviously there are some hurdles to this but you'd save one conversion/regulation step, or at least put into silicon where it's super cheap to manufacture yet can still be very sophisticated.
The small matter of design and implementation is the only problem
Thought you had to be up in the MHz to get much skin effect. At say 30KHz (low end of switching PS frequencies) I'd think it would march right up your arm just like DC.
I suspect there is a gruesome study out there somewhere and I really don't want to see it.
I could still see 30Khz/24V-48V as useful within the data center, even within a server itself. It would simplify wiring and PS count greatly if power could be shipped right down to each board or chip.
At the grid level there is no contest, khz would be impossible plus there is so much installed equipment that it could take $100B to switchover--on the other hand we may be wasting that much $ worth of electricity on line losses and waste heat. I think something like well under half the electricity produced at a typical power plant ends up actually performing useful work on the first order, and several more percent are spent on a/c to evacuate the waste heat from homes and businesses.
You almost seem to be arguing for power distribution to be higher frequency AC, so that the DC conversion componenents can be small, cheap and localized to the point of use. If the HF AC is well chosen for voltage and frequency then some of the intermediate conversion steps can be skipped as it will be "ready" for step down and rectification.
50/60Hz originally was chosen because the low frequencies were easier to generate with the mechanical equipment available 100 years ago but also to avoid a lot of inductive loss into every bit of ferrous metal along the right of way; of course ridiculously high frequencies would actually radiate power away.
I think this could actually work if the power cables were balanced and jacketed at all well, and the voltage was high enough (24v) to carry w/o huge cables. The frequency should probably be whatever switching PS's use these days, ~30Khz?
Also I wonder if there is an increased electrocution risk from HF vs 60Hz.
globalstar and iridium sat phones use(d) LEO sats and had much lower delays.
old-school sat phones (with the briefcase sized antennas) have the delay, but when you're "embedded" in afghanistan you put up with it.
The unmentioned downside is that ping times via geosynhcronous orbiting sats are awful.
* Gaming is impossible
* web surfing is click...wait...wait...wait...display. and wait twice as long if javascript tries to load more images.
* FTP and email work fine, as do most file sharing networks.
* VOIP has intolerable delays, plus the delays are so long some echo cancellation algorithms give up
I would think that a parabolic and a corner cube with the same cross-sections would have about the same gain and frequency response on-axis. off axis the cc wins by so much that even some on-axis loss would
1Kz ~= 1 foot wavelength so 2ft diameter reflectors should do a nice job of any plain old acoustic countermeasures. however if the beam width is 50 ft at your position you're only going to send back a fraction of a percent of the power anyway, parabolic or not, so unless the entire side of your ship is lined with cc's it's pretty much a waste of time.
focused ultrasonics are a bit more of a problem. the interference frequency will just kind of appear on any surface nearby so you won't be able to reflect that, plus the ultrasonics are likely coming from offset positions so the parabola will not be able to reflect both at the same time, and even if the cc does reflect both it won't result in anything destructive reflecting back since it will be out of focus.
it would make a great precedent except that it happened afterward.
parabolic would have to be perfectly aimed in order to work.
corner cube reflector would be a better choice.
Legal documents filed with a court of law are supposed to be carefully reviewed.
The fact that this document was not reviewed by anyone competent is a sign that the plaintiffs are not taking this case seriously. That's not funny, that's just frustrating to all involved.
Why did they not make a resolution-independent standard for HDTV?
Yes, I got one of those (actually, I got the claim form, not the voucher).
Having never actually bought a microsoft product, I was unable to take advantage of the offer. As I recall it maxed out around $300 and to get that you had to have bought odd combinations of things, like MS Office and also Excel separately (which is nearly impossible for average joe users) and a retail copy of windows (pre-installed didn't count).
It will probably get slashdotted immediately, but there is a video here, in QT format (12MB):
http://www.robotcombat.com/video_oldglory_hi.html
>> TVs had a warm, satisfying orange glow eminating from
>> the rear of the set, and a peculiar smell as well.
And a sound! Older sets had notoriously cheap flyback transformers and if your ears were any good you could hear the 15KHz horizontal sync very clearly. Newer CRT's seal the transformer in epoxy or similar plastics but back then they were often made with paper.
>> But laptops suck for 10-drive RAID arrays and such...
Heh, seriously though if he's serving up terabytes maybe it's time to colo anyway.
I've always want to try a winter vs. summer solution.
In the winter, just vent the heat towards the room you spend the most time in.
In the summer, take in air in near ground level (if not prone to floods!) and venting it up and out by convection or just out the nearest wall if fan forced. A steady flow of air, even if it's humid and 90F will keep your servers pretty happy so long as the amount of air is sufficient. A muffin fan won't be enough, but a 6'' duct fan should do the trick.
A more "green" solution would be to switch to fewer, smaller, lower powered servers so that you generate less heat in the first place. A modest laptop running linux/bsd/solaris can actually be quite a powerful server and can handle several services at once.
How to uninstall macromedia flash:
n dex.cfm?id=tn_14157
http://www.macromedia.com/cfusion/knowledgebase/i
In other words, people break into your box at their own RISC?
Like we haven't hear that joke 80386 times before...
You're joking right?
For those few that don't get it, the satellites aren't geostationary, so a failed satellite (once the minimum # is reached) will probably create a dead spot that moves around the planet, in which GPS service is degraded.
Just the fact they they have a CEO tells me everything I need to know about their software.
I was thinking along the lines of some raised bumps along the painted lines. I have mixed feelings though. If people become to aware of the issue they might get a feeling that they are entitled to the area in front of them. This conflicts with the general presumption that the person in front has the right of way.
>> (why are people so much in a hurry to get to a red light?).
My observation of local drivers is that they do it to prevent other drivers from changing into their lane at the light. There is no law (or no enforced law at least) to keep people from jockeying for position at the red light. The only way to maintain your spot in line is to get to the light first. Once you do that, you're now in a position to consider changing lanes yourself if that puts you closer to the beginning of the queue and you become part of the problem.
My preference would be that drivers should not be allowed to change lanes when approaching a red light unless they're going to turn at the intersection.