AF_UNIX sockets are JUST LIKE AF_INET sockets!!
on
DRI Comes to DirectFB
·
· Score: 1
AF_UNIX sockets work just like AF_INET sockets once you bind(2) that file handle! That's not "nothing to do"! That's PLENTY to do with AF_INET! What I'm talking about is a facility meaning it makes something easier. It's not just a bonus, it's the whole point! If you do dumb things like program AF_UNIX sockets without thinking about network transparency, some day you will be cursed by the guy who has to hack in all the missing network-byte-order stuff, etc..in order to get it to work over future-speed networks that are faster than the memory bus on your computer.
One thing you have to tackle when you get into the Unix world is the concept that users and programmers are discrete sets. In Unix, the greater the union of those two sets, the better! If you're a programmer, you should "think user." If you're a user, you should "think programmer."
Oh, and BTW your spam (in the original Usenet sense of repetition)
X WINDOWS DOES NOT USE NETWORKING FOR THE LOCAL SERVER
does not contribute to the reasoning of your argument.
This is not optional. The majors just do not get me off. All they do is sledgehammer the latest fad "sound" that they glean from indie artists that do something new and catch peoples' attentions trhough creativity.
What I really want to see is the return of the "single." I'm tired of buying 2-3 blood-and-sweat songs on a full priced album full of sparse, sloppy half-done tracks. I like the indie artists that are working for the songs, not just doing product development on the albums.Sometimes sloppy and half-done is fun, but it HAS to be fun. I'd like to avoid paying for the "filler" material.I want to use my money to twist the arms of the recording artists.
WORK, DAMNIT! (Ka-ching...) At this point I'm unimpressed with iTunes music store. Maybe later when I'm in the mood to flesh out my collection with stuff I never would buy on CD... It would take some thought.
This isn't an electronics problem. This is a physics problem. We don't even have enough information to tell if the scope of the problem is sane enough to justify considering and discriminating between consumer devices and avionics hardware.
Avionics hardware has to pass rigorous tests for tolerance of harsh operating conditions and even human error in order to recieve FAA clearance for operation on an aircraft. Since the EMI environment of an airplane in flight is unknown, these tolerances must assume a certain amount of EMI.
Consumer electronics produce EMI at known energies at known frequencies. The only unknown is amplification from multipath aggregation characteristics of an airplane.
Assume every passenger had and operated their own consumer electronic device, and this device jams everything except other devices like it across its entire operating bandwidth in the ISM and Cellular bands. Assume these devices transmit at twice the power of any known consumer devices.
Calculate the total energy of all of these devices. Assuming the signals in a worst case will amplify each other, add that to the maximum tolerable level of external EMI that the aircraft encounters in operation. You now have the maximum possible EMI noise energy that any device in the aircraft could possibly encounter under unrealistic and totally idealized conditions.
Can the existing avionics accept this idealized worst-case EMI? Is this even a reasonable question? Could avionics be redesigned to accept this level of EMI?
Your implication that this issue is beyond science because of the three-fold unknown:
a)you have no idea what's going to come onto the plane. There are hundreds of thousands of different electronic devices. b)you have no idea what avionics systems are in the plane c)you have no idea how the device will get used(and RF emissions from a laptop alone can vary on processor/ram activity, screen brightness, peripheral activity...) d)nobody has done even basic studies to see what general kinds of equipment cause interference.
Steve Ballmer is not a technology authority. He doesn't know the family tree. He read the Cliff's Notes on Unix terminology right before he gave this inerview, and you can expect him to screw up a few details.
What you need to zero in on here is "the way things are structured today", meaning the status-quo, as if things aren't going to break that structure, as provided by Microsoft monopoly.
We need to vote for a President who will prosecute antitrust cases. The "structure" is GW's leash on the DoJ.
No, no no no... When Ballmer says "innovation" it means what we think of as "taking credit for other people's pioneering work." So, I think what he's taking about is a stripped-down FreeBSD box running a DOTNET interpereter for non-GUI DOTNET stuff so they can sell grid-node servers and embedded servers because WINCE has no potential for small embedded devices, and Sun's J2EE will embarass them for it.
Except that the instruction decode unit is a modular part of AMD's chip offerings... As in, the AMD chips execute their own native sub-instruction internal code, and they have a programmable front-end to translate the x86 stuff. If you wanted a PPC chip from AMD, Apple would only have to give them the specs, and EXISTING AMD chip cores could be quickly (software) adapted to execute PPC machine code.
All that's left is the pinouts and power/heat-dissipation requirements to sort out. The AltiVec is another side issue, to which I'm sure AMD would gladly offer a next-gen solution...
Nobody in this thread had a single argument to dispute or refute the "market forces" alluded to by the root of this thread.
I read that post to say [irrespective of ethnicity or origin] you get what [qualities] you pay for. I work with some talented Russian and American and Indian developers. The nature of the educational system each person came from *IS* apparent in their different talents and approaches. The ones with really sharp wits rise to the top, getting seniority titles, responsibilities, and pay, and the other ones need more hand-holding.
Some people say that most Indian developers in their experience are weak problem solvers. Unless you have first hand knowledge of these peoples' experiences, you have to accept their statements of fact for arguments' sake. You can only criticise them for not being clear about potential biases that weaken the correlation between the stated experience and everyone else's experience in general.
For example: We might all be correct, as in without disagreement, if the vast majority of weak problem solver Indian developers were hired from the applicants who would accept the lowest salaries. H1B Visas are a diversion. Please help tie this all together, or the "faulty logic" thread will need to continue..
The real problem is how the G.W. Administration and the Military defines Terrorist States. It's a dangerous us-and-them attitude that is completely out of date and distorts our relative weaknesses/advantages dealing with "them". If a whole nation-state and most of its people support terrorists and other questionable policy, like nuclear proliferation, then we should take a defensive posture against them. This is no different than the defensive posture against any other potentially threatening nation, like China.
I'm not even going to take the moral high-road here. In pure Machiavellian politics' terms, the best way to neutralize a threat is to become a means to the enemy's ultimate ends. The G.W. hardliners reject this because they refuse to see the humanity of these people. Poor arab states are populated with people who have not learned concepts like "rule-of-law" and "sanctity of human life" and "people have inalienable rights". The poor, uneducated arab hoi-polloi doesn't recognise those particular carrots dangling in front of them. G.W. and his people think that makes it impossible to deal with them. No, that makes it difficult.
What we need is an enlightened policy towards terrorist states that includes the entire spectrum from friendly, stable, peaceful, democratic nations all the way to unpredictable, dangerous, junta-of-the-day, dictatorships. Things like promoting human rights works to pacify entire nations. Promoting the idea of "limits on powers of government" doesn't sit well with dictatorships' ideas on sovreignty, but in a purely Machiavellian tone: it is wisdom.
You hit on a very important concept, but you fail to underline the dangerous truth behind it. The reason things are so tense between the USA and all these poor countries that harbor terrorists is that our societies are so different. We are advancing and delivering a richer curriculum to more and more of our own people, while "they" are getting less and less. The difference, the gap, is widening, and it has as much to do with our advances as it does with their stagnation. The gap is why G.W. and his people feel so frustrated trying to find foreign policy that fits.
G.W.'s foreign policy is a step away from Bill Clinton's "engagement" policy wherein the USA compromised little by little to achieve social and economic footholds in societies with which we have weak diplomatic security. To my knowledge, we contiune that with China. They are an advanced society with much more in common with the USA than others. The tough ones take a couple of generations to make that kind of progress.
Conceded: the older people in the power seats of those nations are just too slow to chnage their attitudes. However, their children, especially their teenagers, are psychologically and developmentally receptive to American doctrine of rights and limited government, etc.. The real crime is that THOSE ARE THE PEOPLE WE SHOOT! We shoot our own opportunities. We shoot ourselves in the foot. If they begin to percieve us as the means to a better life, the support for terrorism will evaporate (over time). Can we wait to deal with immediate threats? Maybe not. Should we undermine our own long-term agenda because we won't be in office to take credit for the success? NO.
The bottom line is that YOU must get OFF YOUR LUMP and vote G.W. out of office. Aside: we must concede the right to keep and bear arms to the rural America that supported G.W. so staunchly.
Law Enforcement people can't mentally handle the shades of gray, so their struggle is to paint everything in black and white, goodguy vs. badguy colors.
If a criminal were a criminal and that was that we wouldn't need juries and judges or even trials. Ask yourself what kind of world they're really trying to sell you when they say "a criminal is a criminal." What is happening to all the criminals at Enron, or the other dot-bomb scams, or any other company bilking the ma-and-pa investors with their snobbier-than-thug-life white collar crime?
If you need a simpler life, just sit there. The more you try to do, the more complicated it gets. If you're a Cop, you need special appreciation for moral quandaries. If you don't have it SIT DOWN AND SHUT UP!
Ultrasparc places in the low, mid, and high-end server ranges on TPC-H, but IBM and HP will sell you a PC to compete with Sun on the low end, and IBM doesn't even place in the 3TB high-end systems. HP will sell you a PA-RISC system to compete with SunFire 15K. With IBM and HP, you have to rewrite software from the ground up to scale from low to high end of their Sun-competitive server offerings. It's not the same. Apples and oranges.
All that fault isolation crap costs 1.5 - 2 times what you would pay for two non-fault-isolating internally-redundant servers. Very few actually need contunuous-operation guarantees. If it was all that great, we'd all be running HPUX on Stratus PA-RISC systems. SGI is in a tiny-tiny niche occupied by their Cray legacy clients, and also in a niche for render-farms and technical compute farms that are highly (but not as you put it "embarassingly") paralell for software designed from the ground up for N-way concurrency.
The rest of the world wants a fat database server that will be alternately Disk-IO/Network-IO/Memory bound. BTW, Sun has the best, most mature, most stable 64bit addressing architecture to help the transition from a 32bit address space. You dont gain or lose your lead in something like that over one or two generations of hardware. It takes 5-10 years to gain and assert that kind of dominance. Will Sun be so laggard for 5-10 years that it justifies switching now? The world will wait and see, thus Sun holds the brass ring. 64bit needs haven't even been widely expressed, but as storage, and bandwidth gets bigger and faster, UltraSparc will continue to be the standard by which others are judged.
I'm ranting. I have to stop now. Please try to pick the good stuff out for criticism and ignore the ranting crap as much as possible.
The only thing that differentiates Sun hardware is that Solaris runs on it, and thus the multitude of Solaris hosted applications runs on it. Take away Solaris and Sun doesn't lead in anything.
What Sun sells is a scalable platform. It is a *hardware* target for developers to *write software* for. It is not *primarily* an Integrated OS/Hardware stack as you suggest. It is trivial to port human-readable Linux code to Solaris, and thus the success of Linux as a development platform does not preclude anyone's ability to implement that software on Solaris. If you can write good, MT code for Linux, then you can test it on a small sub-$1000 UltraSparc system (Running Linux, FreeBSD, OpenBSD, NetBSD... or even Solaris), or maybe a *little* bigger system if you're paranoid about testing CPU concurrency. Once compiled, that software's performance can be made to scale with the harware it runs on in a near linear way. Code for 2 CPUs and 1GB RAM, but the unaltered software will support 32x the load on a 64CPU 32GB RAM box, without re-engineering anything. Instead of worrying about the MPP (Beowulf) style clustering architecture and optimizing the software for that type of system (if it can even be done), Sun provides a stable ABI across all tiers of servers.
"Take away Solaris and Sun doesn't lead in anything." Well, that's true as far as you know anyways. It's a hardware performance niche that Sun occupies.
Sparc performance has always been and still is a laggard compared to the rest of the industry. Pricing at the low-end is wiped by PCs, and features at the high-end, where the profit margins are still fat, are wiped by IBM and HP in the commercial sector and SGI in the technical markets.
This is all very much subject to debate. First of all, you seem to have very narrow definitions of "performance" (which seems like you mean price/performance in comparison to PCs), and "features". IBM can't provide a platform with the scalability potential of UltraSparc. HP doesn't even have a hardware platform any more! Compaq killed Alpha, and HP killed PA-RISC, and even since buying Compaq, HP can't guarantee a niche for Itanium as they are *more* vulnerable to AMD Opteron competition (high-end to low-end) than Sun is to low-end-PCs in their workstation market. This situation is made MORE significant as Linux gains credibility as an OS. Sun is sitting pretty in the niche they currently occupy.
This really isn't a funny joke. It is a serious legal implication that should be presented to every judge that hears a Microsoft copyright infringement complaint.
Microsoft is making a veiled threat. Also, M&A could raise their stock price with a good purchase.
If I were Google, I would offer Microsoft a support contract and license to use the Google engine and name to promote MSN without selling them any Google IP. Google keeps the patents. I would charge Microsoft a flat royalty off the top of all new MSN revenue, including revenue from bundled products like MSWindows (TM) and MSOffice(TM). Then I would write in a clause to allow rate-review semiannually. If things went sour, another 6 months would see the sunset on the contract.
If MSN wants to compete with AOL, all they need is to buy Larry Flynt. Do they think Google will give them a Pr0n edge? The big-boards (Compuserve, Prodigy, GeNie, AOL) were aways basically supported by pr0n paid for by people who needed extra handholding that only a slick GUI that comes preinstalled on your PC. Like a new club or bar, they should cater to the women who populate the chat rooms. Free webcams, and free service if you get enough "popularity" or something...
Is Google a #1 pr0n portal? I don't understand how they can be competitors...
If ICANN tried to do this (they are a committee of random people who do not run all the root servers), it would require the cooperation of all the root namserver operators. A fork in the DNS authority means some root namserver operators will gladly serivce those requests.
The dissenters would update their root zone to delete NS records pointing to noncooperative root DNS servers, and the root dns servers who got removed from that zone would prolly do the same. The rest of the internet would be split based on who they got their root domain update from.
People would come out of the woodwork to offer hosting for a re-unified DNS because of the confusion that would ensue from forked DNS authority. People would quickly put up DNS servers with "glue" records to choose which root server fork has the right NS records for which domain. It would be a little hacking on the ISC code, but people would have this in distribution within days if not hours.
In laymans terms, DNS would mostly work like it did, after a short hiccup. Some people would get the impression that domains changed hands back-and-forth during the hiccup, but only a few domains would be affected. The vast majority of the people would not understand the controversy. It would look like battling between the parties which swapped apparent control of the affected domains. SOMEBODY would get labelled "hacker" or "rougue".
Lets say you are in a group of people that regularly gets asked obvious questions. Think of it as something like the RGB.txt file in Xwindows. All day long people ask you how much blue is in "LemonChiffon" named color, and you tell them someone else to ask. You refer those trivial queries to other people all day long. The ICANN doesn't pay your salary, but they try (or want) to tell you who and how you can answer those queries. You have your list of people, and you share that list amongst your peers with a handshake agreement: we will all use the same list of who gets referrals for which class of colors. You are completely free to do whatever you want with your list. You have no control over your peers. ICANN has no control over you. Oh, but they whine about it a lot.. "How are we going to make a difference if those guys won't do what we say.. WAAAAAAAA...".
The design of the DNS system makes ICANN unnecessary. The whole idea of ICANN was founded by people who did not understand how this was so for the purpose of establishing privilege ( as in from Latin privilegium, a law affecting one person : privus, single, alone + lex, leg-, law ) for certain minorities to exert control over the DNS namespace.
Large corporations and cadres of lawyers are just as happy as the rest of us about domain squatting. They are even less happy about the whole somethingSUCKS.com court decisions which (by interpereting the US Constitution 1st Amendment) allow people to set up very spiffy parody sites to lampoon their hard-fought corporate images. How are they going to get control of this nasty thorn in their side?
The correct way for those people to solve their problem is to "fork" the DNS root and create their own set of root servers supervised by their lawyers. They could then begin boycotting the original root servers' registrars, and require end users to use DNS servers that submit to their authority. The first problem with that is how so many corporations will fail to agree on enough details to let that happen. The second problem is that anyone could selectively forward queries to their servers for some lookups, and forward queries to other peoples' servers for other lookups. Each DNS server decides who to delegate what authority to. Each end-user could theoretically run their own DNS server without ever needing to query a root server.
The bottom line is that DNS is anarchy, but there is a de facto consensus to trust several root server operators to be cool. The first step to accomplishing what the IANA wants to do is to convince people to revoke trust in the existing root servers. Instead, they keep trying to bully the root server operators, who roll their eyes and sigh..
The real risk that the IANA faces is that the DNS root server authority gets institutionalized in a widely publicised and debated way. If they can't weasel their way into control quietly, they risk the door being be slammed in their face by a new consensus formed out of "informed consent". It's like the UN where everybody has a veto, and it is terribly uncertain how the vote will go.
The real reason the ICANN is such a joke is that the tootpaste is out of the tube. People are widely aware of the attempted power grab, and the important people know how futile that is once it is widely known. ICANN would only be allowed to operate if it behaved identically to the current system, which begs the question: why are we fixing it if it isn't broke?
Pay attention to Verisgn.com (who bought NSI). They will attempt to leverage DNS authority with their x509 business. Look at how BIND9 signed-zones are supposed to work. It isn't just ICANN we should be worried about.
Learn PGP keyring management. It is complicated. It is very worthwhile though. The PGP trust management system is our defense. We should seek to protect the right to that system in the Supreme Court of the US under the Bill of Rights.
Lasser has a pessimistic view of programmers. He assumes they care about security, and he also assumes they are unable to write secure (abuse-resistant) application code.
Lasser is optimistic of high-level languages. He assumes that the buffer-overflow and input validation problems can be solved by giving programmers complicated components wrapped in high-level language constructs, and forcing programmers to use them, as in making the low-level features unavailable.
I agree with Lasser that sloppy use of low-level language constructs is an invitation to disaster. I disagree that abstraction is the solution. I disagree that application programmers sufficiently *care* about security. I disagree that application programmers are unable to code applications securely in low-level languages.
I suggest that application programmers do not care sufficiently about security. How often do you see "test harness" code written by a programmer to expose possible problems with their own libraries? How often does software get refactored to facilitate finer-grained testing? The problem with the status-quo is that the process of finding these bugs too-often escapes the grasp of the application programmers. In the worst-case the code-review is effected by a widespread Internet worm. It should be effected by the programer(s).
Code is law. This is a problem of readability. With insecure programs, it is not obvious that the security vulnerabilities exist. Bugs == undocumented features. We must not avoid the questions: What are my expectations of this code, and does this code conform to my expectations? OpenBSD is written in C. It uses low-level language constructs, but its record is significantly better than Lasser's examples, and improving.
Using high-level languages *can* be more secure, and they represent a certain kind of discipline, which is helpful. However they are not panacea for buffer-overflows, and they represent lack-of-awareness, which is dangerous. Moreover, most low-level languages have features which allow programmers to create their own abstract constructs. Refactoring applications and creating such abstractions as a part of application development allows programmers to take the advantages of both high and low level languages, and with discipline they can avoid the drawbacks of either.
Programming is intellectual work. It is never going to be easy to do it right. Even a simple programming exercise (that in the end requires no changes or improvements) should recieve the same level of scrutiny as the level of its users' dependence. Coding without that scrutiny--code review--is not programming. It is bare coding. Programming is distinguished by problem solving. Coding is just writing anything that can be validated by a compiler. Without that Philosophy (as in love-of-wisdom), you are dealing with fools' code.
(In sid EE us) Adj. 1. Something which persists or propagates through obfuscated and/or malignant means. "This virus was particularly insidious considering how far it spread before we discovered how to control it." 2. Something which is used to entrap. "The sting is an insidious but effective law-enforcement technique." 3. Bad, but superficially good or attractive. "The insidious nature of heroin and other opoid drugs is often used to characterize and generalize all recreational drugs." 4. A computer program that produces output which is shared between people, but is useless without compensating the authors of the computer program. "Microsoft Office is insidious in the way it takes a person's work and leverages it to sell more copies of Microsoft Office to others who cannot otherwise use the work."
Proprietary file formats are a thing of the past. They used to be a necessary evil, but now they're just evil. If people send you a document you can't read it is THEIR FAULT for not knowing what formats you require. It's a kind of bigotry, and we all have to show a little backbone when it counts.
DISCLAIMER:
This somewhat editorialized post is meant to provide some possible answers for you to ponder. I don't know if any of this is true. I do wonder though... and YOU SHOULD TOO!
Security firms have nothing better to do most of the time except sit around and think of a way to scare up another emergency. If they think it up (and make the claim publicly), they get headlines. If we don't rush to pay them for their services, then they will release the exploit to the kiddies (aka start breakin' kneecaps). It's totally conflict-of-interest and its totally protection-racket.
That's why they give us dilligent sysadmins a break. They release a warning that's just good enough for the 5-10% of the market who would never pay them for their dubius services willingly. The rest of the people get "You should be more like them. Pay us to show you how."... or something BAD might happen...
That's what they call "white-hat" cracking. You sit around with computers full of the most popular software/versions and hammer the protocols with garbage until you get some kind of scream-in-agony from the victim servers' debugger/log windows. Then you get a cup of coffee. Then you go back to your source-code review, and look for ways to tweak your brute-force buffer-overflow hammer perl script. You drink a lot of coffee in that job.
When you find something, you ponder heavily whether to brag on IRC #l33t or start a boring vulnerability report. Maybe you can get a payroll wannabe to do your work while you brag on #l33t.
Go read back-issues of PHRACK or PHRACK or cDc for some introduction to the boring-ass-world of trying to boost your self-esteem at the expense of dumber folks.
I moved sixteen Gigs and what do I get?
Another day older and deeper in debt!
Honey don't you call me, 'cause I can't come:
I owe my soul to the company store.
Slashdot Mirror
AF_UNIX sockets work just like AF_INET sockets once you bind(2) that file handle! That's not "nothing to do"! That's PLENTY to do with AF_INET! What I'm talking about is a facility meaning it makes something easier. It's not just a bonus, it's the whole point! If you do dumb things like program AF_UNIX sockets without thinking about network transparency, some day you will be cursed by the guy who has to hack in all the missing network-byte-order stuff, etc..in order to get it to work over future-speed networks that are faster than the memory bus on your computer.
One thing you have to tackle when you get into the Unix world is the concept that users and programmers are discrete sets. In Unix, the greater the union of those two sets, the better! If you're a programmer, you should "think user." If you're a user, you should "think programmer."
Oh, and BTW your spam (in the original Usenet sense of repetition)
does not contribute to the reasoning of your argument.This is not optional. The majors just do not get me off. All they do is sledgehammer the latest fad "sound" that they glean from indie artists that do something new and catch peoples' attentions trhough creativity.
What I really want to see is the return of the "single." I'm tired of buying 2-3 blood-and-sweat songs on a full priced album full of sparse, sloppy half-done tracks. I like the indie artists that are working for the songs, not just doing product development on the albums.Sometimes sloppy and half-done is fun, but it HAS to be fun. I'd like to avoid paying for the "filler" material.I want to use my money to twist the arms of the recording artists.
WORK, DAMNIT! (Ka-ching...) At this point I'm unimpressed with iTunes music store. Maybe later when I'm in the mood to flesh out my collection with stuff I never would buy on CD... It would take some thought.
This isn't an electronics problem. This is a physics problem. We don't even have enough information to tell if the scope of the problem is sane enough to justify considering and discriminating between consumer devices and avionics hardware.
Avionics hardware has to pass rigorous tests for tolerance of harsh operating conditions and even human error in order to recieve FAA clearance for operation on an aircraft. Since the EMI environment of an airplane in flight is unknown, these tolerances must assume a certain amount of EMI.
Consumer electronics produce EMI at known energies at known frequencies. The only unknown is amplification from multipath aggregation characteristics of an airplane.
Assume every passenger had and operated their own consumer electronic device, and this device jams everything except other devices like it across its entire operating bandwidth in the ISM and Cellular bands. Assume these devices transmit at twice the power of any known consumer devices. Calculate the total energy of all of these devices. Assuming the signals in a worst case will amplify each other, add that to the maximum tolerable level of external EMI that the aircraft encounters in operation. You now have the maximum possible EMI noise energy that any device in the aircraft could possibly encounter under unrealistic and totally idealized conditions.
Can the existing avionics accept this idealized worst-case EMI? Is this even a reasonable question? Could avionics be redesigned to accept this level of EMI?
Your implication that this issue is beyond science because of the three-fold unknown:
WRONG.Steve Ballmer is not a technology authority. He doesn't know the family tree. He read the Cliff's Notes on Unix terminology right before he gave this inerview, and you can expect him to screw up a few details.
What you need to zero in on here is "the way things are structured today", meaning the status-quo, as if things aren't going to break that structure, as provided by Microsoft monopoly.
We need to vote for a President who will prosecute antitrust cases. The "structure" is GW's leash on the DoJ.
No, no no no... When Ballmer says "innovation" it means what we think of as "taking credit for other people's pioneering work." So, I think what he's taking about is a stripped-down FreeBSD box running a DOTNET interpereter for non-GUI DOTNET stuff so they can sell grid-node servers and embedded servers because WINCE has no potential for small embedded devices, and Sun's J2EE will embarass them for it.
Except that the instruction decode unit is a modular part of AMD's chip offerings... As in, the AMD chips execute their own native sub-instruction internal code, and they have a programmable front-end to translate the x86 stuff. If you wanted a PPC chip from AMD, Apple would only have to give them the specs, and EXISTING AMD chip cores could be quickly (software) adapted to execute PPC machine code.
All that's left is the pinouts and power/heat-dissipation requirements to sort out. The AltiVec is another side issue, to which I'm sure AMD would gladly offer a next-gen solution...
Nobody in this thread had a single argument to dispute or refute the "market forces" alluded to by the root of this thread.
I read that post to say [irrespective of ethnicity or origin] you get what [qualities] you pay for. I work with some talented Russian and American and Indian developers. The nature of the educational system each person came from *IS* apparent in their different talents and approaches. The ones with really sharp wits rise to the top, getting seniority titles, responsibilities, and pay, and the other ones need more hand-holding.
Some people say that most Indian developers in their experience are weak problem solvers. Unless you have first hand knowledge of these peoples' experiences, you have to accept their statements of fact for arguments' sake. You can only criticise them for not being clear about potential biases that weaken the correlation between the stated experience and everyone else's experience in general.
For example: We might all be correct, as in without disagreement, if the vast majority of weak problem solver Indian developers were hired from the applicants who would accept the lowest salaries. H1B Visas are a diversion. Please help tie this all together, or the "faulty logic" thread will need to continue..
The Romanians saw Nikolai Ceaucescu (sp?) holding those carrots in Romania.
The real problem is how the G.W. Administration and the Military defines Terrorist States. It's a dangerous us-and-them attitude that is completely out of date and distorts our relative weaknesses/advantages dealing with "them". If a whole nation-state and most of its people support terrorists and other questionable policy, like nuclear proliferation, then we should take a defensive posture against them. This is no different than the defensive posture against any other potentially threatening nation, like China.
I'm not even going to take the moral high-road here. In pure Machiavellian politics' terms, the best way to neutralize a threat is to become a means to the enemy's ultimate ends. The G.W. hardliners reject this because they refuse to see the humanity of these people. Poor arab states are populated with people who have not learned concepts like "rule-of-law" and "sanctity of human life" and "people have inalienable rights". The poor, uneducated arab hoi-polloi doesn't recognise those particular carrots dangling in front of them. G.W. and his people think that makes it impossible to deal with them. No, that makes it difficult.
What we need is an enlightened policy towards terrorist states that includes the entire spectrum from friendly, stable, peaceful, democratic nations all the way to unpredictable, dangerous, junta-of-the-day, dictatorships. Things like promoting human rights works to pacify entire nations. Promoting the idea of "limits on powers of government" doesn't sit well with dictatorships' ideas on sovreignty, but in a purely Machiavellian tone: it is wisdom.
You hit on a very important concept, but you fail to underline the dangerous truth behind it. The reason things are so tense between the USA and all these poor countries that harbor terrorists is that our societies are so different. We are advancing and delivering a richer curriculum to more and more of our own people, while "they" are getting less and less. The difference, the gap, is widening, and it has as much to do with our advances as it does with their stagnation. The gap is why G.W. and his people feel so frustrated trying to find foreign policy that fits.
G.W.'s foreign policy is a step away from Bill Clinton's "engagement" policy wherein the USA compromised little by little to achieve social and economic footholds in societies with which we have weak diplomatic security. To my knowledge, we contiune that with China. They are an advanced society with much more in common with the USA than others. The tough ones take a couple of generations to make that kind of progress.
Conceded: the older people in the power seats of those nations are just too slow to chnage their attitudes. However, their children, especially their teenagers, are psychologically and developmentally receptive to American doctrine of rights and limited government, etc.. The real crime is that THOSE ARE THE PEOPLE WE SHOOT! We shoot our own opportunities. We shoot ourselves in the foot. If they begin to percieve us as the means to a better life, the support for terrorism will evaporate (over time). Can we wait to deal with immediate threats? Maybe not. Should we undermine our own long-term agenda because we won't be in office to take credit for the success? NO.
The bottom line is that YOU must get OFF YOUR LUMP and vote G.W. out of office. Aside: we must concede the right to keep and bear arms to the rural America that supported G.W. so staunchly.
Law Enforcement people can't mentally handle the shades of gray, so their struggle is to paint everything in black and white, goodguy vs. badguy colors.
If a criminal were a criminal and that was that we wouldn't need juries and judges or even trials. Ask yourself what kind of world they're really trying to sell you when they say "a criminal is a criminal." What is happening to all the criminals at Enron, or the other dot-bomb scams, or any other company bilking the ma-and-pa investors with their snobbier-than-thug-life white collar crime?
If you need a simpler life, just sit there. The more you try to do, the more complicated it gets. If you're a Cop, you need special appreciation for moral quandaries. If you don't have it SIT DOWN AND SHUT UP!
Ultrasparc places in the low, mid, and high-end server ranges on TPC-H, but IBM and HP will sell you a PC to compete with Sun on the low end, and IBM doesn't even place in the 3TB high-end systems. HP will sell you a PA-RISC system to compete with SunFire 15K. With IBM and HP, you have to rewrite software from the ground up to scale from low to high end of their Sun-competitive server offerings. It's not the same. Apples and oranges.
Oh, BTW: TPC-H results
All that fault isolation crap costs 1.5 - 2 times what you would pay for two non-fault-isolating internally-redundant servers. Very few actually need contunuous-operation guarantees. If it was all that great, we'd all be running HPUX on Stratus PA-RISC systems. SGI is in a tiny-tiny niche occupied by their Cray legacy clients, and also in a niche for render-farms and technical compute farms that are highly (but not as you put it "embarassingly") paralell for software designed from the ground up for N-way concurrency.
The rest of the world wants a fat database server that will be alternately Disk-IO/Network-IO/Memory bound. BTW, Sun has the best, most mature, most stable 64bit addressing architecture to help the transition from a 32bit address space. You dont gain or lose your lead in something like that over one or two generations of hardware. It takes 5-10 years to gain and assert that kind of dominance. Will Sun be so laggard for 5-10 years that it justifies switching now? The world will wait and see, thus Sun holds the brass ring. 64bit needs haven't even been widely expressed, but as storage, and bandwidth gets bigger and faster, UltraSparc will continue to be the standard by which others are judged.
I'm ranting. I have to stop now. Please try to pick the good stuff out for criticism and ignore the ranting crap as much as possible.
I have to disagree.
What Sun sells is a scalable platform. It is a *hardware* target for developers to *write software* for. It is not *primarily* an Integrated OS/Hardware stack as you suggest. It is trivial to port human-readable Linux code to Solaris, and thus the success of Linux as a development platform does not preclude anyone's ability to implement that software on Solaris. If you can write good, MT code for Linux, then you can test it on a small sub-$1000 UltraSparc system (Running Linux, FreeBSD, OpenBSD, NetBSD... or even Solaris), or maybe a *little* bigger system if you're paranoid about testing CPU concurrency. Once compiled, that software's performance can be made to scale with the harware it runs on in a near linear way. Code for 2 CPUs and 1GB RAM, but the unaltered software will support 32x the load on a 64CPU 32GB RAM box, without re-engineering anything. Instead of worrying about the MPP (Beowulf) style clustering architecture and optimizing the software for that type of system (if it can even be done), Sun provides a stable ABI across all tiers of servers."Take away Solaris and Sun doesn't lead in anything." Well, that's true as far as you know anyways. It's a hardware performance niche that Sun occupies.
This is all very much subject to debate. First of all, you seem to have very narrow definitions of "performance" (which seems like you mean price/performance in comparison to PCs), and "features". IBM can't provide a platform with the scalability potential of UltraSparc. HP doesn't even have a hardware platform any more! Compaq killed Alpha, and HP killed PA-RISC, and even since buying Compaq, HP can't guarantee a niche for Itanium as they are *more* vulnerable to AMD Opteron competition (high-end to low-end) than Sun is to low-end-PCs in their workstation market. This situation is made MORE significant as Linux gains credibility as an OS. Sun is sitting pretty in the niche they currently occupy.TROLL
This really isn't a funny joke. It is a serious legal implication that should be presented to every judge that hears a Microsoft copyright infringement complaint.
Everyone would include you, your trollness...
Microsoft is making a veiled threat. Also, M&A could raise their stock price with a good purchase.
If I were Google, I would offer Microsoft a support contract and license to use the Google engine and name to promote MSN without selling them any Google IP. Google keeps the patents. I would charge Microsoft a flat royalty off the top of all new MSN revenue, including revenue from bundled products like MSWindows (TM) and MSOffice(TM). Then I would write in a clause to allow rate-review semiannually. If things went sour, another 6 months would see the sunset on the contract.
If MSN wants to compete with AOL, all they need is to buy Larry Flynt. Do they think Google will give them a Pr0n edge? The big-boards (Compuserve, Prodigy, GeNie, AOL) were aways basically supported by pr0n paid for by people who needed extra handholding that only a slick GUI that comes preinstalled on your PC. Like a new club or bar, they should cater to the women who populate the chat rooms. Free webcams, and free service if you get enough "popularity" or something...
Is Google a #1 pr0n portal? I don't understand how they can be competitors...
If ICANN tried to do this (they are a committee of random people who do not run all the root servers), it would require the cooperation of all the root namserver operators. A fork in the DNS authority means some root namserver operators will gladly serivce those requests.
The dissenters would update their root zone to delete NS records pointing to noncooperative root DNS servers, and the root dns servers who got removed from that zone would prolly do the same. The rest of the internet would be split based on who they got their root domain update from.
People would come out of the woodwork to offer hosting for a re-unified DNS because of the confusion that would ensue from forked DNS authority. People would quickly put up DNS servers with "glue" records to choose which root server fork has the right NS records for which domain. It would be a little hacking on the ISC code, but people would have this in distribution within days if not hours.
In laymans terms, DNS would mostly work like it did, after a short hiccup. Some people would get the impression that domains changed hands back-and-forth during the hiccup, but only a few domains would be affected. The vast majority of the people would not understand the controversy. It would look like battling between the parties which swapped apparent control of the affected domains. SOMEBODY would get labelled "hacker" or "rougue".
Lets say you are in a group of people that regularly gets asked obvious questions. Think of it as something like the RGB.txt file in Xwindows. All day long people ask you how much blue is in "LemonChiffon" named color, and you tell them someone else to ask. You refer those trivial queries to other people all day long. The ICANN doesn't pay your salary, but they try (or want) to tell you who and how you can answer those queries. You have your list of people, and you share that list amongst your peers with a handshake agreement: we will all use the same list of who gets referrals for which class of colors. You are completely free to do whatever you want with your list. You have no control over your peers. ICANN has no control over you. Oh, but they whine about it a lot.. "How are we going to make a difference if those guys won't do what we say.. WAAAAAAAA...".
DNS IS ANARCHY THAT WORKS!
The design of the DNS system makes ICANN unnecessary. The whole idea of ICANN was founded by people who did not understand how this was so for the purpose of establishing privilege ( as in from Latin privilegium, a law affecting one person : privus, single, alone + lex, leg-, law ) for certain minorities to exert control over the DNS namespace.
Large corporations and cadres of lawyers are just as happy as the rest of us about domain squatting. They are even less happy about the whole somethingSUCKS.com court decisions which (by interpereting the US Constitution 1st Amendment) allow people to set up very spiffy parody sites to lampoon their hard-fought corporate images. How are they going to get control of this nasty thorn in their side?
The correct way for those people to solve their problem is to "fork" the DNS root and create their own set of root servers supervised by their lawyers. They could then begin boycotting the original root servers' registrars, and require end users to use DNS servers that submit to their authority. The first problem with that is how so many corporations will fail to agree on enough details to let that happen. The second problem is that anyone could selectively forward queries to their servers for some lookups, and forward queries to other peoples' servers for other lookups. Each DNS server decides who to delegate what authority to. Each end-user could theoretically run their own DNS server without ever needing to query a root server.
The bottom line is that DNS is anarchy, but there is a de facto consensus to trust several root server operators to be cool. The first step to accomplishing what the IANA wants to do is to convince people to revoke trust in the existing root servers. Instead, they keep trying to bully the root server operators, who roll their eyes and sigh..
The real risk that the IANA faces is that the DNS root server authority gets institutionalized in a widely publicised and debated way. If they can't weasel their way into control quietly, they risk the door being be slammed in their face by a new consensus formed out of "informed consent". It's like the UN where everybody has a veto, and it is terribly uncertain how the vote will go.
The real reason the ICANN is such a joke is that the tootpaste is out of the tube. People are widely aware of the attempted power grab, and the important people know how futile that is once it is widely known. ICANN would only be allowed to operate if it behaved identically to the current system, which begs the question: why are we fixing it if it isn't broke?
Pay attention to Verisgn.com (who bought NSI). They will attempt to leverage DNS authority with their x509 business. Look at how BIND9 signed-zones are supposed to work. It isn't just ICANN we should be worried about.
Learn PGP keyring management. It is complicated. It is very worthwhile though. The PGP trust management system is our defense. We should seek to protect the right to that system in the Supreme Court of the US under the Bill of Rights.
Excuse me, WTF?
Lasser has a pessimistic view of programmers. He assumes they care about security, and he also assumes they are unable to write secure (abuse-resistant) application code.
Lasser is optimistic of high-level languages. He assumes that the buffer-overflow and input validation problems can be solved by giving programmers complicated components wrapped in high-level language constructs, and forcing programmers to use them, as in making the low-level features unavailable.
I agree with Lasser that sloppy use of low-level language constructs is an invitation to disaster. I disagree that abstraction is the solution. I disagree that application programmers sufficiently *care* about security. I disagree that application programmers are unable to code applications securely in low-level languages.
I suggest that application programmers do not care sufficiently about security. How often do you see "test harness" code written by a programmer to expose possible problems with their own libraries? How often does software get refactored to facilitate finer-grained testing? The problem with the status-quo is that the process of finding these bugs too-often escapes the grasp of the application programmers. In the worst-case the code-review is effected by a widespread Internet worm. It should be effected by the programer(s).
Code is law. This is a problem of readability. With insecure programs, it is not obvious that the security vulnerabilities exist. Bugs == undocumented features. We must not avoid the questions: What are my expectations of this code, and does this code conform to my expectations? OpenBSD is written in C. It uses low-level language constructs, but its record is significantly better than Lasser's examples, and improving.
Using high-level languages *can* be more secure, and they represent a certain kind of discipline, which is helpful. However they are not panacea for buffer-overflows, and they represent lack-of-awareness, which is dangerous. Moreover, most low-level languages have features which allow programmers to create their own abstract constructs. Refactoring applications and creating such abstractions as a part of application development allows programmers to take the advantages of both high and low level languages, and with discipline they can avoid the drawbacks of either.
Programming is intellectual work. It is never going to be easy to do it right. Even a simple programming exercise (that in the end requires no changes or improvements) should recieve the same level of scrutiny as the level of its users' dependence. Coding without that scrutiny--code review--is not programming. It is bare coding. Programming is distinguished by problem solving. Coding is just writing anything that can be validated by a compiler. Without that Philosophy (as in love-of-wisdom), you are dealing with fools' code.
Proprietary file formats are a thing of the past. They used to be a necessary evil, but now they're just evil. If people send you a document you can't read it is THEIR FAULT for not knowing what formats you require. It's a kind of bigotry, and we all have to show a little backbone when it counts.
Parable of the talents.
Parable of the mustard seed.
Only the filthy rich can afford to be so stupid.
Why do we think they should be running the country?
DISCLAIMER:
This somewhat editorialized post is meant to provide some possible answers for you to ponder. I don't know if any of this is true. I do wonder though... and YOU SHOULD TOO!
Security firms have nothing better to do most of the time except sit around and think of a way to scare up another emergency. If they think it up (and make the claim publicly), they get headlines. If we don't rush to pay them for their services, then they will release the exploit to the kiddies (aka start breakin' kneecaps). It's totally conflict-of-interest and its totally protection-racket.
That's why they give us dilligent sysadmins a break. They release a warning that's just good enough for the 5-10% of the market who would never pay them for their dubius services willingly. The rest of the people get "You should be more like them. Pay us to show you how." ... or something BAD might happen...
That's what they call "white-hat" cracking. You sit around with computers full of the most popular software/versions and hammer the protocols with garbage until you get some kind of scream-in-agony from the victim servers' debugger/log windows. Then you get a cup of coffee. Then you go back to your source-code review, and look for ways to tweak your brute-force buffer-overflow hammer perl script. You drink a lot of coffee in that job.
When you find something, you ponder heavily whether to brag on IRC #l33t or start a boring vulnerability report. Maybe you can get a payroll wannabe to do your work while you brag on #l33t.
Go read back-issues of PHRACK or PHRACK or cDc for some introduction to the boring-ass-world of trying to boost your self-esteem at the expense of dumber folks.
I moved sixteen Gigs and what do I get?
Another day older and deeper in debt!
Honey don't you call me, 'cause I can't come:
I owe my soul to the company store.