i just ditched nwlink at the end of janurary. i had no idea this was coming, but its nice that it didn't affect me.
i was sick of my dsl line randomly getting throttled back to like 44kbps or so until i powere cycled it. i didnt like the lame metered bandwidth crap. i didnt like how they could never keep me from getting other people emails.
any company that charges me bandwidth for a home line these days is ridiculous. is it fair to charge me for the traffic of shitholes portscanning my box ?
Write your own AGI script that takes into account things like time of day, ringing caller id, etc etc.
If every handset in your house is an asterisk extension, and asterisk is the only telco-facing system, you will have 100% control of when your phones ring (and how, as asterisk will send any ring-pattern you want)
For legal reasons, you may want to have one other traditional phone plugged into the telco-facing network, with its ringer permanantly off. That way in the event of a PBX failure, you can still dial 911, but the phone will never ring.
I planned on doing this so my grandmother would _always_ go straight to voicemail, with no internal extensions ringing, whenever she tried to call me before noon.
The only link that matters in any discussion of linux telephony is
www.asteriskpbx.org
I am on the mailing list. It is an _awesome_ project. I'm just waiting for enough disposable income to buy the 24port FXO devkit.
If you've ever wanted to build your own customizable, modular PBX system, asterisk is the thing to have. That it happens to be in production _today_ at multiple homes and businesses as an H.323 to POTS gateway is what makes it related to this story.
If you want to use gnomephone for _free_, use asterisk, _today_.
Someone could setup an IAX box in each of the NXX/NPA's in the usa and build thier own FREE ld network _today_ if they wanted to.
Nobody that is experienced enough to make an objective computer security threat analysis works in the insurance industry.
Insurance underwriters aren't fantastically bright, and the actuaries that keep insurance companies running don't have computer security expertise.
Just running FreeBSD vs NT4.0 SP1 isn't enough to make any kind of a policy decision.
If all aspects of network, administration, training, etc were identical, then, perhaps, a premium for a policy might be different based on OS choice.
But, if you know anything at all, the choice of technology has a lot less to do with computer security than the choice of humans and procedures.
Furthermore, the number of variables involved in something like writing a anti-hacker policy is so large that many of them will be ignored, as the models that the actuaries use to do their risk analysis work on terms of masses of people, and all of the variables in those models are not exposed to the underwriters.
Here's an example -- for an auto insurance policy, they ask you what kind of car it is, how often you get a trafic violation, how old you are, and how far you drive in an average day.
Wouldn't you say there are more variables that are statistically significant than that ? The underwriters don't care that you drink, own a cell phone, etc etc (although they may start on the cell phone matter). Yet those are things that may be modelled by the actuaries, on broad terms to set the safety margins in the premiums.
Similarly, there are all kinds of factors that relate to insuring some kind of computer operation. First off, if you're talking about a computer network, then the OS on a specific host doesn't seem relevant (or were you envisioning that each host would be a multi-valued line item on the policy, just like each car is a line item on an insurance policy ?)
Re:But how would they cover the debt....
on
[H|Cr]acker Insurance
·
· Score: 2, Insightful
Good question.
This is what actuaries do. They determine how to make money off of policies, they determine risk exposure and how to mitigate that risk, etc etc.
To have an actuary that could successfully do a plausible job at this, you'd need one that was a computer security and loss expert.
My father was the youngest person to become an FSA (Fellow of the Society of Actuaries) and last year was the Computer Science chairperson for the SOA (Society of Actuaries).
As both an accomplished actuary (to say the least) and an accomplished computerphile (are you fluent in 360 assembler ?) i feel like he's pretty well versed to speak on this matter.
I can tell you quite confidently that the cross section of actuaries, and people who are computer security experts in the united states is roughly:
0 persons.
When the "hacker insurance costs more for IIS" article came out a year ago i talked it over with my dad. He said it was, "bullshit", and went into a small rant about how ridiculous and sensationalist it was.
the word file format has not changed in the last three versions.
People were running X11 on windows long before OSX was still nextstep.
"Best Page Layout" software is a matter of opinion. What software do you refer to, that doesn't run on windows ?
If you want to talk about problems, lets talk about hooking non-apple hardware to older macs (say, a toshiba cdrom). Lets talk about hooking your APPLE BRANDED serial printer up to your machine running OSX.
Please name some things macOS will do that windows wont. Please be specific.
what sucks about it ? I've used SFU 2.2, 3.0 beta, and 3.0 release. I love it. For putting a useful commandline environment on a w2k/xp box, nothing else comes close.
i agree with everything you've said here, except about nintendo being #2...of which im not necessarily convinced. consider:
in the USA xbox has sold more units that Gamecube. Additionally, Splinter Cell has sold more copies than Metroid Prime. (metroid was the game people were _waiting_ for on the gc and splinter cell comes out of no where and crushes it (crushing meaning "beating it":)
in the USA, of "current gen" consoles, it goes like this:
PS2 XBOX Gamecube
imnot sure where PS1 ranks, but in japan ps1 still outsells xbox afaik.
so as far as global rankings, i think GC has better market share in japan and europe, but not USA. So i dont know what the "world wide" rank is.
I think it's important to consider the value-add of a console maker.
Traditionally, it was a way to publish first party titles and if you published enough of them, you got all of the software revenue.
Nintendo must be in a position where they think the additional share of revenue they get from not paying sony or MS to develop for PS2 or xbox outweights the cost to design/build/support a hardware platform. There's no compelling reason to reason to target only Gamecube apart from exclusivity deals with nintendo. If you had to pick just one console, the cube loses on market share and features (ps2 wins on share, xbox wins on features)
its good that sony broke the strangle hold on the console market. its bad that sony gained it from nintendo. it remains to be seen if xbox will break the stranglehold from sony within "this round"
in any case,. nintendo confirming a console in 2005 is probably one of the worlds biggest non-announcements ever. it's almost as premature, ridiculous, and outlandish as sony spewing off some mumbo jumbo about a 1tflop gaming machine that uses biotechnology and P2P networks.... oh wait!
bosch L-Jetronic is an analog fuel injection computer. As seen on chassis E12 BMWs, and other german cars of the early 80s.
It was replaced on BMWs in the early 80s with Motronic, which actually has a microprocessor and software "maps" instead of analog functions.
Nowadays, all ECU systems are program + data. Some chips modify just some of the data, some chips modify most of the data, and some chips modify some code _and_ data.
They haven't dropped IRIX or MIPS. They aren't repackaging anyone elses stuff - this is still the most advanced hardware platform in existance, and who do you think it was that made Linux non-laughable on > 32 cpus (check for all the kernel work sgi has done)
Finally, SGI _used_ to sell x86 boxes running NT, and has since quit. And even then, they were nonstandard parts.
Basically, everything in your post was wrong, except maybe the part about SGI getting desperate, and how what has happened to SGI is a shame.
first off, with 16" on center stud walls, constructed of 2x4s, and an average studheight of 92 and 5/8ths inches, you can see that the volume inside a stud wall "cell" is pretty piss poor - roughly 5800 cubic inches.
There are a few issues that make this "solution" stupid.
1) the heat doesn't go anywhere. there should be a correspondingly large diameter cut out in the top plate of the wall, so that the air can escape in the attic (where it might do some good, as the attic is cold and properly ventilated, unlike the interior of a wall)
2) there may be cold water supply pipes in wall. do you want to heat your cold water ? especially if they're copper pipes with a very effective heat transfer characteristic
2a) there may be runs of NM-B (romex) electrical cable in that wall cell. The ampacity of electrical wire is a function of its rated capacity, and while most ampacity ratings are given up to 70C, if this thing were _seriously_ efficient at cooling a computer, then it would perhaps begin to cause problems with in-wall structures
3) how does the national fire code feel about stuffing heat into closed interior walls (made of flame-retardant drywall, typically)
4) if the excess heat it dispells isn't enough to cause any code violations, then it clearly isn't sucking enough heat to be worth installing
5) this does little to eliminate the overall heat+noise of _systems_
My idea for this was to find an abandoned refrigerator, or better yet, freezer, and just putting whole systems inside there, and then running flue-spec double-walled exhaust vent pipe elsewhere. Having all the PCs stuck inside a fridge/freezer (shut off, of course) that was properly vented should make things cool _AND_ quiet. Don't beleive me ? Try putting your battery powered alarm clock in your freezer, and see if you can still hear it once the door shuts. You want whole-system noise cancellation ? Then you need real insulation. Want to keep your office cool? then you'll need to do a lot more than putting a turbluent undersized vacuum hose on the back of your PC....if i ever find a fridge and hook this up, i'll be sure to post pictures:)
one optimization - there are numerous projects that are calculating all known prime numbers. It should be a matter of a few moments to see if any of these projects has reached the 1024bit mark, in other words, has any project found all known primes which can be represented in 1024 bits or less.
i see that http://www.utm.edu/research/primes/ftp/all.txt
lists primes in the form of 2^13466917-1
which is clearly larger than a 2048 bit number... so if "all the gaps" have been filled in than the entire set of numbers for p and q are known...
also, tehcnically the prime could be as large as 2047 bits. i realize thatn to check for primality, you only need to check factores up to the square root of the number, but the "other factor" (and we want both of them) is guaranteed to be larger than 1024 bits (unless the key is a perfect square of primes).
so anyway, if theres a project that lists all primes up to 1024bits or 2048bits in length, that is the "input" to the problem - the products of all such pairs of primes is the brute force space, not all possible 2048 bit numbers.
so actually, if you take the "forward" approach, you can choose p and q from the prime database, see if they multiply to N, and if so, you've got d. Distributing "the list" into pq product pairs for calculation is massively distributable, and it seems like knowing some things about N gives you some additional advantages in terms of products you dont need to calculate.
that said, i think the resultant number of actual p's and qs there are is an important number to actually spell out.
the prime number theorem says that this should be roughly x / log(x-1)
turns out that this grows pretty fast. for 2^128, the numberspace is about 3.4E38, while the number of primes is 8E36. Also, the ratio of integerspace:primespace grows roughly linearly with 2^n at about.301 from n=1..n=128.
when you move up to a 512 bit integerspace, the ratio of integers to primes has increased to about 154:1.
excel gives up with 1024bit numbers, but 1023 is showing about a 308:1 ratio. so assume that for a 2048 bit space, there will be a 1:600 ratio of primes to non primes.
not that dividing such a huge space by 600 is significant in terms of space size, but it does make it a bit nicer when approaching it from the brute force aspect. you either finish 600 times faster or with 600 times less machines:)
incidentally, sqrt(2^1024) is about 153 digits long:)
i dont need a MIPS history lesson. I didn't "forget" any of those CPUs. The R8000 was almost non-existant across SGI's product line. While it was the first implementation of MIPS4, and it was an FP monster, and had a huge TLB for the time, it really wasn't so hot as a general purpose CPU. A far as "true 64 bit" in the R4000, which version of IRIX ran on R4k with 64 bit pointers ? 6.2 and 6.5 certainly don't on my IP22.
When the R3k came out it was the first real example of commercially FAST and successful RISC design. It was used in multiple machines from multiple companies. SGI didn't "really" up the ante again until R10k, which was their first offering that was superpipelined and superscalar.
Finally, regarding SGI and clustering:
SGI is not price-competitive with shared-nothing clusters of PCs or Alphas. Nor is it trying to be. You probably know what the O2k/O3k systems are good at and how they differ from any other system being sold today, othewise you wouldn't have responded to me. I think my statement is valid --- the SGI big iron solves problems that shared nothing clusters CANT. Furthermore, they're so much more expensive than shared nothings that if you need shared nothing and buy origin, you're silly.
SGI will continue to make investments in IRIX and MIPS until it makes sense to move all of their products and customers to Linux on IA64, and that may not happen until theres something better than Linux+IA64 out:) So who knows when MIPS IRIX will officially go away.
Linux isn't there yet for the bread and butter SGI customers. Neither is IA64.
processor performance has never been SGI's strong point, except for breifly after the R3000 and R10000 were introduced.
SGI's workstation line is largely unimpressive, especially for the 99% case of computer users, hell, even engineers.
The problem is, for a small set of jobs, for a small set of people, nothing else is suficient - at any price. You're either using an SGI, or the work isn't taking place.
That market is continuing to erode, but i dont think it will ever dissolve completely. I think eventually SGi will effectively become a US govt subsidized entity. SGI continues to build the systems that only governments need and only government agencies can afford.
Clustering has nothing to do with the markets SGI sells in. Please don't mention it, it makes me think you don't know what you're talking about.
Do you ?
Re:The last of the V8 Interceptors
on
New Mad Max Film
·
· Score: 3, Informative
the problem with NOS is that its not movie authentic.
You may remember that humongous's turbine powered truck seemed to have a NOS bottle on it that gave it the power to overtake mad max and then the quasi-gay biker used a vertical truck exhaust pipe to bash in the windshield on the interceptor, forcing max off the road and ultimately causing the car to be destroyed and his dog to be shot.
putting NOS on a car would be like sleeping with the enemy. I don't care if it would go faster, max's car didn't have it:)
im not sure if this should get +1 Funny (why do you know that shit, loser) or -1 Get A Life (why do you know that shit, loser)
The last of the V8 Interceptors
on
New Mad Max Film
·
· Score: 5, Interesting
The Road Warrior was my favorite movie from the first time i saw it (at the tender age of 5) up until "The Devil's Advocate" came out.
Nobody in kindergarten knew what the hell i was talking about if i asked them if they needed a guy to haul this rig.
To this day, i still want the car he had in mad max. That supercharger (albeit fake) was the coolest thing i've ever seen. And i want a gear lever with a red button on it that makes the most glorious sound i've ever heard. I had a whole section on my website about the mad max car and some guy in.au emailed me about it with lots of details and info while i was in college. There are clone mad max cars up for auction from time to time. Hopefully, when im old and loaded, i'll be able to pick up a perfect replica mad-max car, with a _working_ super charger that somehow makes that incredible whine when i engage the supercharger.
I've actually asked a couple of tuners about that functionality, apparently it was pretty suspect. To run a boosted motor you need to run lower compression pistons to avoid predetonation, which means that when the SC was disengaged you'd be making shit for power , (although i guess technically you'd be using less gas, but the engine would be way less than optimally efficient). that makes it basically a tradeoff, theres probably some crossover point where you're actually getting better "bang for your gas" with the SC engaged than with it off.
Also, whowever invented the wrist-gauntlet mounted mini-crossbow is a diety.
Not only is proprietary softare inherently insecure, it's inherently more expensive, inherently doesn't work as well
Get over yourself. security, cost, and "work as well" are not inherent characteristics of any method of managing who gets to look at source code.
Open Source does NOT equate better security. Never think so, never say so, never tell anyone this again. It's pure bullshit. Security is a disciplined mindset. And the opensource mindset is "someone will look at this and email me a patch if i fucked up". Lovely.
Free software is free if you don't value your time. Of _course_ contracting agencies are going to be in favor of open source systems that cost them nothing to procure or redistribute and have less functionality out of the box so that they get more billable hours re-inventing the wheel (again).
How's that open source sound coming along in linux these days ? Working much better than say, sound in MacOS ? Windows ? Irix ? Again, this is pure bullshit.
Why were new versions of HP-UX required ? What did it offer for systems that were already in-place ? What was so compelling about a new version of HP-UX that justified the considerable man hours of rolling it out (note that the human cost here usually dwarfs the cost of acquiring new HPUX hardware)
The concept of absolute morality is the first unresolved question of first semester political science.
You are free to beleive what you like (that there is an absolute right and an absolute wrong). That you seem to think it is unfathomable to differ from your opinions is unfortuneate comedy.
The only right that is guaranteed is the right to suffer. Every other right has a price. Today, like every other day, I thank my lucky stars that my rights as an American have been bought and paid for by the blood of heroes.
If Joe China wants the right to speak his mind politically, he knows what the costs associated with getting to the US are, and he knows what the costs of trying to change his home regime are. He'll likely choose, as many others, to get to the USA any way he can (its cheaper)
The notion that what a specific person thinks (usually the "speaker") is absolutely correct and should be swiftly mandated into law that affects the masses is a sure sign of non-fitness of said speaker for ANY position in public policy.
At work, i have 2 XP boxes and a W2k Advanced server machine all powered up in my office. There's a linux box and an openbsd box i used for a few investigative projects, but they're powered down right now.
I've got a "General machine" running XP that i do my web surfing, mail deleting, and the majority of my work on (it usually has 2 or 3 isqlw windows open at all times, not to mention all those instant messenger windows)
My other xp machine is a much faster dedicated machine for work in vs.net, as well as working with 3 separate source trees for forks of a testing harness that i manage. I have outlook completely unconfigured on this machine, as outlook only makes things slow:) I dont run any IM software on the machine, infact, the only office component i use on that box is excel.
At home, my main machine is a dual proc sparcstation 10. (my SGI indigo^2 ate its disk and i have no install media). I have an openBSD IDE fileserver, and another openBSD dedicated firewall (p166mmx, no less). I've also got a 486-133 distributed across the floor. Finally, there's a p233mmx running win2k server. My wife has a powerbook g3. I have an IBM Z50 CE laptop.
If i read slashdot from work, its IE6. From home, its either IE6 (via TS connection to the windows box from my wireless Z50) or links on the sparc 10. no version of any netscape product is fast enough on any machine i own.
at home my main apps are mutt, irc, and rdesktop, sprinkled with links, and gnuplot and xfig occasionally.
The biggest thing the windows machine gets used for is for our photoprinter, and running the BMW ETK software (You can buy the mobile traditions CDROM which is a snapshot of the BMW parts database for older cars. includes technical drawings/diagrams.. helps me keep my 1988 M5 running)
Basically, i dont care what os I use - whatever fits the needs the easiest. Fuck trying to get a photoprinter working on anything besides windows.. i mean.. why bother ? And i know that some people can get netbsd running on the ibm Z50, but again, why bother ? CE has some annoying quirks (like no soft reset, and forgetting what my Aironet 352 card looks like about every 5th power-wake).
Basically, any OS i've mentioned here - OS X, W2k, XP, OpenBSD, and solaris -- they're all passable and don't get in my way enough to cause frustration for the tasks they're doing. On solaris i use CDE because fuck compiling gnome or something else, CDE works great for me and doesn't slow me down.
So at work, its 99% windows, at home, its 70/30 unix/windows. I've got too much of an investment in unix hardware, unix knowledge, and unix comfort, to just throw it away.
One thing that makes Microsoft products a good choice at work is um, that i've been a microsoft employee for about two and a half years. They're sort of the incumbant environment at work. Although there are people here that use exchange connector and Evolution against the corporate exchange servers (mostly to try it out, assess the "competition", etc). And one of the VB developers writes his code in raw emacs for NT, without c-mode.el (he wrote his own c-mode a long time ago and prefers it).
So, i dont really care what platform i use. What im doing is more important than the toolchain, because most toolchains and platforms these days are passable. I missed all MS operating systems after win 3.1 and went straight to w2k, but our lab team has to support everything 95osr2 and above, so i get to see how shitty W9x is from time to time:)
realistically, W2k and XP are good enough that i dont miss anything about unix, except the unix-specific things. For that, i install MS Services for UNIX on my machines. the big things i use are "tail -f" and one-off sed/awk/grep/head/tail/cut constructs that are just stupidly frustrating to do in Cygwin or with the dos ports of the GNU tools. SFU does all that stuff perfectly, and gives me a real single-rooted fs and a real tsch hosted on my box. I can paste data from sed into excel or notepad. It's really the best of both worlds.
the spending on xbox is an investment. if it pays off, microsoft will have a few things
1) a huge source of licensing revenue 2) a huge source of recurring service revenue (xbox live) 3) a presence in the living room (this is _huge_) 4) another platform to provide best-of-breed tools, support, etc for
Microsoft's business is not "PC software". It's software. And its migrating from software pure-play to software and software services. XBox gives them a platform to sell first party software, and to collect platform licensing fees for 3rd party titles. It also allows them to adapt software tools to an additional platform and charge for _those_ as well.
If MS jacked up the prices of windows and office, everyone would yell bloody murder (just like they do anyway). cranking up prices on products and calling them "done" is how you lose. investing in R&D is how you win. xbox is a big investment and if it pays off it will do so handsomely. if it doesn't, well, there have been bad investments before. as long as more investments pay off than tank, things will continue to be just peachy in redmond.
COMPAQ's Adapter Teaming drivers for compaq badged server NICs.
See, it takes two physical drivers, inserts itself above them in the stack by consuming their NDIS interface (afaik) and re-exposes a single NDIS interface. That way you can bind tcp/ip to this virtual adaptor, and it does adaptor load balancing, by dispatching work to each of the physical adaptors beneath it in the stack.
Did you think i was full of it, or are you playing "quiz bowl" with me ?
Slashdot Mirror
i just ditched nwlink at the end of janurary. i had no idea this was coming, but its nice that it didn't affect me.
i was sick of my dsl line randomly getting throttled back to like 44kbps or so until i powere cycled it. i didnt like the lame metered bandwidth crap. i didnt like how they could never keep me from getting other people emails.
any company that charges me bandwidth for a home line these days is ridiculous. is it fair to charge me for the traffic of shitholes portscanning my box ?
What you want is the asterisk pbx system.
http://www.asteriskpbx.org
Write your own AGI script that takes into account things like time of day, ringing caller id, etc etc.
If every handset in your house is an asterisk extension, and asterisk is the only telco-facing system, you will have 100% control of when your phones ring (and how, as asterisk will send any ring-pattern you want)
For legal reasons, you may want to have one other traditional phone plugged into the telco-facing network, with its ringer permanantly off. That way in the event of a PBX failure, you can still dial 911, but the phone will never ring.
I planned on doing this so my grandmother would _always_ go straight to voicemail, with no internal extensions ringing, whenever she tried to call me before noon.
The only link that matters in any discussion of linux telephony is
www.asteriskpbx.org
I am on the mailing list. It is an _awesome_ project. I'm just waiting for enough disposable income to buy the 24port FXO devkit.
If you've ever wanted to build your own customizable, modular PBX system, asterisk is the thing to have. That it happens to be in production _today_ at multiple homes and businesses as an H.323 to POTS gateway is what makes it related to this story.
If you want to use gnomephone for _free_, use asterisk, _today_.
Someone could setup an IAX box in each of the NXX/NPA's in the usa and build thier own FREE ld network _today_ if they wanted to.
this is ridiculous.
Nobody that is experienced enough to make an objective computer security threat analysis works in the insurance industry.
Insurance underwriters aren't fantastically bright, and the actuaries that keep insurance companies running don't have computer security expertise.
Just running FreeBSD vs NT4.0 SP1 isn't enough to make any kind of a policy decision.
If all aspects of network, administration, training, etc were identical, then, perhaps, a premium for a policy might be different based on OS choice.
But, if you know anything at all, the choice of technology has a lot less to do with computer security than the choice of humans and procedures.
Furthermore, the number of variables involved in something like writing a anti-hacker policy is so large that many of them will be ignored, as the models that the actuaries use to do their risk analysis work on terms of masses of people, and all of the variables in those models are not exposed to the underwriters.
Here's an example -- for an auto insurance policy, they ask you what kind of car it is, how often you get a trafic violation, how old you are, and how far you drive in an average day.
Wouldn't you say there are more variables that are statistically significant than that ? The underwriters don't care that you drink, own a cell phone, etc etc (although they may start on the cell phone matter). Yet those are things that may be modelled by the actuaries, on broad terms to set the safety margins in the premiums.
Similarly, there are all kinds of factors that relate to insuring some kind of computer operation. First off, if you're talking about a computer network, then the OS on a specific host doesn't seem relevant (or were you envisioning that each host would be a multi-valued line item on the policy, just like each car is a line item on an insurance policy ?)
Good question.
This is what actuaries do. They determine how to make money off of policies, they determine risk exposure and how to mitigate that risk, etc etc.
To have an actuary that could successfully do a plausible job at this, you'd need one that was a computer security and loss expert.
My father was the youngest person to become an FSA (Fellow of the Society of Actuaries) and last year was the Computer Science chairperson for the SOA (Society of Actuaries).
As both an accomplished actuary (to say the least) and an accomplished computerphile (are you fluent in 360 assembler ?) i feel like he's pretty well versed to speak on this matter.
I can tell you quite confidently that the cross section of actuaries, and people who are computer security experts in the united states is roughly:
0 persons.
When the "hacker insurance costs more for IIS" article came out a year ago i talked it over with my dad. He said it was, "bullshit", and went into a small rant about how ridiculous and sensationalist it was.
the word file format has not changed in the last three versions.
People were running X11 on windows long before OSX was still nextstep.
"Best Page Layout" software is a matter of opinion. What software do you refer to, that doesn't run on windows ?
If you want to talk about problems, lets talk about hooking non-apple hardware to older macs (say, a toshiba cdrom). Lets talk about hooking your APPLE BRANDED serial printer up to your machine running OSX.
Please name some things macOS will do that windows wont. Please be specific.
yes, i have, and it doesn't come close.
for one, there is no tty handling to speak of, as near as i can tell.
what sucks about it ? I've used SFU 2.2, 3.0 beta, and 3.0 release. I love it. For putting a useful commandline environment on a w2k/xp box, nothing else comes close.
i agree with everything you've said here, except about nintendo being #2...of which im not necessarily convinced. consider:
:)
... oh wait!
in the USA xbox has sold more units that Gamecube. Additionally, Splinter Cell has sold more copies than Metroid Prime. (metroid was the game people were _waiting_ for on the gc and splinter cell comes out of no where and crushes it (crushing meaning "beating it"
in the USA, of "current gen" consoles, it goes like this:
PS2
XBOX
Gamecube
imnot sure where PS1 ranks, but in japan ps1 still outsells xbox afaik.
so as far as global rankings, i think GC has better market share in japan and europe, but not USA. So i dont know what the "world wide" rank is.
I think it's important to consider the value-add of a console maker.
Traditionally, it was a way to publish first party titles and if you published enough of them, you got all of the software revenue.
Nintendo must be in a position where they think the additional share of revenue they get from not paying sony or MS to develop for PS2 or xbox outweights the cost to design/build/support a hardware platform. There's no compelling reason to reason to target only Gamecube apart from exclusivity deals with nintendo. If you had to pick just one console, the cube loses on market share and features (ps2 wins on share, xbox wins on features)
its good that sony broke the strangle hold on the console market. its bad that sony gained it from nintendo. it remains to be seen if xbox will break the stranglehold from sony within "this round"
in any case,. nintendo confirming a console in 2005 is probably one of the worlds biggest non-announcements ever. it's almost as premature, ridiculous, and outlandish as sony spewing off some mumbo jumbo about a 1tflop gaming machine that uses biotechnology and P2P networks.
bosch L-Jetronic is an analog fuel injection computer. As seen on chassis E12 BMWs, and other german cars of the early 80s.
It was replaced on BMWs in the early 80s with Motronic, which actually has a microprocessor and software "maps" instead of analog functions.
Nowadays, all ECU systems are program + data. Some chips modify just some of the data, some chips modify most of the data, and some chips modify some code _and_ data.
Stop trolling, and stop being dumb.
They haven't dropped IRIX or MIPS.
They aren't repackaging anyone elses stuff - this is still the most advanced hardware platform in existance, and who do you think it was that made Linux non-laughable on > 32 cpus (check for all the kernel work sgi has done)
Finally, SGI _used_ to sell x86 boxes running NT, and has since quit. And even then, they were nonstandard parts.
Basically, everything in your post was wrong, except maybe the part about SGI getting desperate, and how what has happened to SGI is a shame.
this is woefully unimpressive, and uninspired.
...if i ever find a fridge and hook this up, i'll be sure to post pictures :)
first off, with 16" on center stud walls, constructed of 2x4s, and an average studheight of 92 and 5/8ths inches, you can see that the volume inside a stud wall "cell" is pretty piss poor - roughly 5800 cubic inches.
There are a few issues that make this "solution" stupid.
1) the heat doesn't go anywhere. there should be a correspondingly large diameter cut out in the top plate of the wall, so that the air can escape in the attic (where it might do some good, as the attic is cold and properly ventilated, unlike the interior of a wall)
2) there may be cold water supply pipes in wall. do you want to heat your cold water ? especially if they're copper pipes with a very effective heat transfer characteristic
2a) there may be runs of NM-B (romex) electrical cable in that wall cell. The ampacity of electrical wire is a function of its rated capacity, and while most ampacity ratings are given up to 70C, if this thing were _seriously_ efficient at cooling a computer, then it would perhaps begin to cause problems with in-wall structures
3) how does the national fire code feel about stuffing heat into closed interior walls (made of flame-retardant drywall, typically)
4) if the excess heat it dispells isn't enough to cause any code violations, then it clearly isn't sucking enough heat to be worth installing
5) this does little to eliminate the overall heat+noise of _systems_
My idea for this was to find an abandoned refrigerator, or better yet, freezer, and just putting whole systems inside there, and then running flue-spec double-walled exhaust vent pipe elsewhere. Having all the PCs stuck inside a fridge/freezer (shut off, of course) that was properly vented should make things cool _AND_ quiet. Don't beleive me ? Try putting your battery powered alarm clock in your freezer, and see if you can still hear it once the door shuts. You want whole-system noise cancellation ? Then you need real insulation. Want to keep your office cool? then you'll need to do a lot more than putting a turbluent undersized vacuum hose on the back of your PC.
great information.
.301 from n=1..n=128.
:)
:)
one optimization - there are numerous projects that are calculating all known prime numbers. It should be a matter of a few moments to see if any of these projects has reached the 1024bit mark, in other words, has any project found all known primes which can be represented in 1024 bits or less.
i see that http://www.utm.edu/research/primes/ftp/all.txt
lists primes in the form of
2^13466917-1
which is clearly larger than a 2048 bit number... so if "all the gaps" have been filled in than the entire set of numbers for p and q are known...
also, tehcnically the prime could be as large as 2047 bits. i realize thatn to check for primality, you only need to check factores up to the square root of the number, but the "other factor" (and we want both of them) is guaranteed to be larger than 1024 bits (unless the key is a perfect square of primes).
so anyway, if theres a project that lists all primes up to 1024bits or 2048bits in length, that is the "input" to the problem - the products of all such pairs of primes is the brute force space, not all possible 2048 bit numbers.
so actually, if you take the "forward" approach, you can choose p and q from the prime database, see if they multiply to N, and if so, you've got d. Distributing "the list" into pq product pairs for calculation is massively distributable, and it seems like knowing some things about N gives you some additional advantages in terms of products you dont need to calculate.
that said, i think the resultant number of actual p's and qs there are is an important number to actually spell out.
the prime number theorem says that this should be roughly x / log(x-1)
turns out that this grows pretty fast. for 2^128, the numberspace is about 3.4E38, while the number of primes is 8E36. Also, the ratio of integerspace:primespace grows roughly linearly with 2^n at about
when you move up to a 512 bit integerspace, the ratio of integers to primes has increased to about 154:1.
excel gives up with 1024bit numbers, but 1023 is showing about a 308:1 ratio. so assume that for a 2048 bit space, there will be a 1:600 ratio of primes to non primes.
not that dividing such a huge space by 600 is significant in terms of space size, but it does make it a bit nicer when approaching it from the brute force aspect. you either finish 600 times faster or with 600 times less machines
incidentally, sqrt(2^1024) is about 153 digits long
i dont need a MIPS history lesson. I didn't "forget" any of those CPUs. The R8000 was almost non-existant across SGI's product line.
While it was the first implementation of MIPS4, and it was an FP monster, and had a huge TLB for the time, it really wasn't so hot as a general purpose CPU.
A far as "true 64 bit" in the R4000, which version of IRIX ran on R4k with 64 bit pointers ? 6.2 and 6.5 certainly don't on my IP22.
When the R3k came out it was the first real example of commercially FAST and successful RISC design. It was used in multiple machines from multiple companies. SGI didn't "really" up the ante again until R10k, which was their first offering that was superpipelined and superscalar.
Finally, regarding SGI and clustering:
SGI is not price-competitive with shared-nothing clusters of PCs or Alphas. Nor is it trying to be. You probably know what the O2k/O3k systems are good at and how they differ from any other system being sold today, othewise you wouldn't have responded to me. I think my statement is valid --- the SGI big iron solves problems that shared nothing clusters CANT. Furthermore, they're so much more expensive than shared nothings that if you need shared nothing and buy origin, you're silly.
Nope.
:) So who knows when MIPS IRIX will officially go away.
SGI will continue to make investments in IRIX and MIPS until it makes sense to move all of their products and customers to Linux on IA64, and that may not happen until theres something better than Linux+IA64 out
Linux isn't there yet for the bread and butter SGI customers. Neither is IA64.
processor performance has never been SGI's strong point, except for breifly after the R3000 and R10000 were introduced.
SGI's workstation line is largely unimpressive, especially for the 99% case of computer users, hell, even engineers.
The problem is, for a small set of jobs, for a small set of people, nothing else is suficient - at any price. You're either using an SGI, or the work isn't taking place.
That market is continuing to erode, but i dont think it will ever dissolve completely. I think eventually SGi will effectively become a US govt subsidized entity. SGI continues to build the systems that only governments need and only government agencies can afford.
Clustering has nothing to do with the markets SGI sells in. Please don't mention it, it makes me think you don't know what you're talking about.
Do you ?
the problem with NOS is that its not movie authentic.
:)
You may remember that humongous's turbine powered truck seemed to have a NOS bottle on it that gave it the power to overtake mad max and then the quasi-gay biker used a vertical truck exhaust pipe to bash in the windshield on the interceptor, forcing max off the road and ultimately causing the car to be destroyed and his dog to be shot.
putting NOS on a car would be like sleeping with the enemy. I don't care if it would go faster, max's car didn't have it
im not sure if this should get
+1 Funny (why do you know that shit, loser)
or
-1 Get A Life (why do you know that shit, loser)
The Road Warrior was my favorite movie from the first time i saw it (at the tender age of 5) up until "The Devil's Advocate" came out.
.au emailed me about it with lots of details and info while i was in college. There are clone mad max cars up for auction from time to time. Hopefully, when im old and loaded, i'll be able to pick up a perfect replica mad-max car, with a _working_ super charger that somehow makes that incredible whine when i engage the supercharger.
Nobody in kindergarten knew what the hell i was talking about if i asked them if they needed a guy to haul this rig.
To this day, i still want the car he had in mad max. That supercharger (albeit fake) was the coolest thing i've ever seen. And i want a gear lever with a red button on it that makes the most glorious sound i've ever heard. I had a whole section on my website about the mad max car and some guy in
I've actually asked a couple of tuners about that functionality, apparently it was pretty suspect. To run a boosted motor you need to run lower compression pistons to avoid predetonation, which means that when the SC was disengaged you'd be making shit for power , (although i guess technically you'd be using less gas, but the engine would be way less than optimally efficient). that makes it basically a tradeoff, theres probably some crossover point where you're actually getting better "bang for your gas" with the SC engaged than with it off.
Also, whowever invented the wrist-gauntlet mounted mini-crossbow is a diety.
Get over yourself. security, cost, and "work as well" are not inherent characteristics of any method of managing who gets to look at source code.
Why were new versions of HP-UX required ? What did it offer for systems that were already in-place ? What was so compelling about a new version of HP-UX that justified the considerable man hours of rolling it out (note that the human cost here usually dwarfs the cost of acquiring new HPUX hardware)
The concept of absolute morality is the first unresolved question of first semester political science.
You are free to beleive what you like (that there is an absolute right and an absolute wrong). That you seem to think it is unfathomable to differ from your opinions is unfortuneate comedy.
The only right that is guaranteed is the right to suffer. Every other right has a price. Today, like every other day, I thank my lucky stars that my rights as an American have been bought and paid for by the blood of heroes.
If Joe China wants the right to speak his mind politically, he knows what the costs associated with getting to the US are, and he knows what the costs of trying to change his home regime are. He'll likely choose, as many others, to get to the USA any way he can (its cheaper)
The notion that what a specific person thinks (usually the "speaker") is absolutely correct and should be swiftly mandated into law that affects the masses is a sure sign of non-fitness of said speaker for ANY position in public policy.
http://www.microsoft.com/windows2000/hpc/
Cornell has some windows clusters that they seem to like ok.
http://www.tc.cornell.edu/
Ghost Recon is almost exactly what you described - multiplayer coop battles. it rules.
Mech assault is also a lot of fun, and motoGP with 16 players is a blast as well
At work, i have 2 XP boxes and a W2k Advanced server machine all powered up in my office. There's a linux box and an openbsd box i used for a few investigative projects, but they're powered down right now.
:) I dont run any IM software on the machine, infact, the only office component i use on that box is excel.
:)
I've got a "General machine" running XP that i do my web surfing, mail deleting, and the majority of my work on (it usually has 2 or 3 isqlw windows open at all times, not to mention all those instant messenger windows)
My other xp machine is a much faster dedicated machine for work in vs.net, as well as working with 3 separate source trees for forks of a testing harness that i manage. I have outlook completely unconfigured on this machine, as outlook only makes things slow
At home, my main machine is a dual proc sparcstation 10. (my SGI indigo^2 ate its disk and i have no install media). I have an openBSD IDE fileserver, and another openBSD dedicated firewall (p166mmx, no less). I've also got a 486-133 distributed across the floor. Finally, there's a p233mmx running win2k server. My wife has a powerbook g3. I have an IBM Z50 CE laptop.
If i read slashdot from work, its IE6. From home, its either IE6 (via TS connection to the windows box from my wireless Z50) or links on the sparc 10. no version of any netscape product is fast enough on any machine i own.
at home my main apps are mutt, irc, and rdesktop, sprinkled with links, and gnuplot and xfig occasionally.
The biggest thing the windows machine gets used for is for our photoprinter, and running the BMW ETK software (You can buy the mobile traditions CDROM which is a snapshot of the BMW parts database for older cars. includes technical drawings/diagrams.. helps me keep my 1988 M5 running)
Basically, i dont care what os I use - whatever fits the needs the easiest. Fuck trying to get a photoprinter working on anything besides windows.. i mean.. why bother ? And i know that some people can get netbsd running on the ibm Z50, but again, why bother ? CE has some annoying quirks (like no soft reset, and forgetting what my Aironet 352 card looks like about every 5th power-wake).
Basically, any OS i've mentioned here - OS X, W2k, XP, OpenBSD, and solaris -- they're all passable and don't get in my way enough to cause frustration for the tasks they're doing. On solaris i use CDE because fuck compiling gnome or something else, CDE works great for me and doesn't slow me down.
So at work, its 99% windows, at home, its 70/30 unix/windows. I've got too much of an investment in unix hardware, unix knowledge, and unix comfort, to just throw it away.
One thing that makes Microsoft products a good choice at work is um, that i've been a microsoft employee for about two and a half years. They're sort of the incumbant environment at work. Although there are people here that use exchange connector and Evolution against the corporate exchange servers (mostly to try it out, assess the "competition", etc). And one of the VB developers writes his code in raw emacs for NT, without c-mode.el (he wrote his own c-mode a long time ago and prefers it).
So, i dont really care what platform i use. What im doing is more important than the toolchain, because most toolchains and platforms these days are passable. I missed all MS operating systems after win 3.1 and went straight to w2k, but our lab team has to support everything 95osr2 and above, so i get to see how shitty W9x is from time to time
realistically, W2k and XP are good enough that i dont miss anything about unix, except the unix-specific things. For that, i install MS Services for UNIX on my machines. the big things i use are "tail -f" and one-off sed/awk/grep/head/tail/cut constructs that are just stupidly frustrating to do in Cygwin or with the dos ports of the GNU tools. SFU does all that stuff perfectly, and gives me a real single-rooted fs and a real tsch hosted on my box. I can paste data from sed into excel or notepad. It's really the best of both worlds.
the register is dumb. also, you are dumb.
the spending on xbox is an investment. if it pays off, microsoft will have a few things
1) a huge source of licensing revenue
2) a huge source of recurring service revenue (xbox live)
3) a presence in the living room (this is _huge_)
4) another platform to provide best-of-breed tools, support, etc for
Microsoft's business is not "PC software". It's software. And its migrating from software pure-play to software and software services. XBox gives them a platform to sell first party software, and to collect platform licensing fees for 3rd party titles. It also allows them to adapt software tools to an additional platform and charge for _those_ as well.
If MS jacked up the prices of windows and office, everyone would yell bloody murder (just like they do anyway). cranking up prices on products and calling them "done" is how you lose. investing in R&D is how you win. xbox is a big investment and if it pays off it will do so handsomely. if it doesn't, well, there have been bad investments before. as long as more investments pay off than tank, things will continue to be just peachy in redmond.
COMPAQ's Adapter Teaming drivers for compaq badged server NICs.
See, it takes two physical drivers, inserts itself above them in the stack by consuming their NDIS interface (afaik) and re-exposes a single NDIS interface. That way you can bind tcp/ip to this virtual adaptor, and it does adaptor load balancing, by dispatching work to each of the physical adaptors beneath it in the stack.
Did you think i was full of it, or are you playing "quiz bowl" with me ?