That implies a strange type of liberalism, I'm guessing you're American.
Guilty as charged. You probably wouldn't think so from this post
A good round-up of Liberal* can be found here and the American flavour here.
Interesting too. Not a political scientist; wasn't aware of that sort of nuance in the definition of liberalism.
It's always struck me as ironic that calling something or someone a liberal is viewed as a bad thing in the "Land of the Free". I assume it's just a FUD campaign by the ultra-wrongwingers to make people think "Liberal"==="Evil Satanist".
To further compound the irony, I consider myself more liberal in political orientation. While I consider most American conservatives (that would vote for GWB) as blustering, whiney, cheerleading morons, I do think that American liberals will need to rehabilitate their image by precisely defining the label in more pragmatic terms and letting go of their policy failures. Gov't can't end poverty by wealth redistribution. LBJ's Great Society was a failure as a "liberal" program. Stop depending on traditional gov't structures to correct societal problems (except health care).
Example: if you're going to push for things like school lunch subsidy programs, make sure you point out how relatively inexpensive it is, and its results. Then do not go and say if you have feeding programs for anyone poor, that you'll get the same results and it will be worthwhile. Any "liberal" that does so needs to be verbally abused, badly. Or, carefully point out that the "progressive" aspects of tax policy is a good, "liberal" thing, and conservatives are thieves that don't want to pay any taxes, just stick the bill on the non-rich. (Assuming of course, one doesn't "want" a flat tax policy.) Start attacking gov't subsidy of corporations, and attach that thievery as "conservative". (Of course it won't happen, because the system is rigged.)
If you want more power, just bundle another couple on.
Nope. The rocket is only engineered for what is attached. There is a base in place to perhaps attach a few more rocket motors. But more components means a greater percentage of failure. This route has been taken before with the Soviets.
The sad thing is that it will still be cheaper just to pay the Russians to throw something up into space. Perhaps the more realistic and effective use for NASA's money would be to invest and manage (by customer requirements) the operations end of Russia's launch facilities. In other words, outsource.
You still need to put money into a D-9 program for domestic considerations (the military). Also, with an active rocket program, you have an infrastructure to propel development of rockets to take care of unique missions which the Soviets would not be interested in developing.
The worst thing is that the average American voter is a crippled mind. Not just is it substandard in science knowlege, but now Americans lack imagination and vision. The answer to a future space launch platform is not rockets. Its the space elevator. You're not going to colonize planets or develop an extended presence in space with rockets.
Its as doable now as thermonuclear weapons were back in the '40's. NASA should outsource launch facilities, maintain its current scientific missions, and put the bulk of its money into engineering a space elevator. If its too expensive, get the rest of the world to kick in for its development. Get those cheap Indians and Russians crunching out the numbers, and let the Americans specialize in the design and engineering.
Even if you keep the project in-house, think of the boon it would be for American infrastructure. Hey MIT, Caltech, Los Alamos, etc., here's a billion dollars, go engineer a space elevator. Then put it out to bid for Lockheed, Boeing, etc. to actually build it. Nope, Americans are too stupid to see the economic utility and importance of scientific investment. No vision.
For one thing, the Space Shuttle is the only American man-rated launch system in service (well, nearly in service) today. The last one has not been used since the Apollo-Soyuz joint mission, and there is no tooling or production facilities to build an Apollo-style capsule or launch vehicle to carry it aloft.
Blah, blah, blah. And the Empire State Building is now the tallest skyscraper in Manhattan. Meaningless factoid. Guess what, eventually there will be another building which will be taller. The answer is to go build it. 23 ton PAYLOAD. You merely design a new capsule to put on top of the rocket. Hell, probably could kludge a Soyuz to fit on top of it.
Secondly, there are still missions that require both heavy lifting and human beings. For example, if NASA were to choose to repair the HST using a non-robotic mission, it would be the Shuttle that carried the repairmen aloft.
Yes, but you don't need the shuttle to do that. Manned capsule on rocket can do the same damn thing. Why are you so in love with something that crippled the US space agenda? You're like a liberal who thinks the gov't can end poverty by wealth redistribution.
You're talking out of your ass. There hasn't ever been a shuttle mission which required taking a satellite out of orbit and landing it on earth.
There isn't any utility in doing so either. Its cheaper to send up a new satellite.
Shuttle was an engineering marvel, but a white elephant failure. Disposable rockets are cheaper, and it sucked out all that money that could have been used for a manned Mars mission, or a "useful" space station.
As soon as someone can how me ONE study showing ANY danger from irradiated food, and we can start comparing it against the well know risks of all the other preservation methods.
Apparently, you've never seen cautionary tales as "Attack of the Killer Tomatoes". Of course, after seeing "The Toxic Avenger" I'd take irradiation over chemical preservatives any day...)
Two of the main problems of the existing system are the fact it is very difficult to get rid of bad teachers (tenure anyone?) and that there is a increasing loss of local control.
You have to be able to get rid of bad teachers or be able to move them around. for things to ever improve. The problem with making them fully civil servants
Tenure is a big problem, and its ridiculous practice to apply to primary school educators. The whole point of tenure was to allow researchers with a proven record of academic competence to present unpopular fruits of research without getting fired for presenting the new idea. Primary school educators do not present new research to high school children, and technically, they're not supposed to be teaching anything which contradicts the prerogative of the local school board. Unfortunately, its been a contractual practice to incorporate tenure at that instruction level because of convention, so removing it becomes a sticking point in contract negotiations. Thus, we get stuck with underperformers that we cannot remove.
Don't forget our legal system for contributing to this decline. A lot of destructive policy gets put into place to placate the leeches in suits. And finally, yes, the parents are probably the biggest problem. But there's no means testing in handing out marriage licenses or pregnancy. Its also ridiculous not be able to expel students with sufficent cause. On the other hand, you have idiot administrators expelling kids for possession of aspirin or a t-shirt.
The real problem is we live in a country who's people are stupid enough to vote for Bush.
This is quaint nostalgia from limited imaginations. What you want is The Matrix.
With holograms, you can see it. But you can't smell, taste, or touch it.
Yeah, watching Angelina Jolie strip in 3 dimensions is sweet, but with the Matrix, you can be doing AJ.
Take it from Cypher:
You know, I know this steak doesn't exist. I know that when I put it in my mouth, the Matrix is telling my brain that it is juicy and delicious. After nine years, you know what I realize?
[Takes a bite of steak]
Ignorance is bliss.
You make some great points, but we seem to be arguing about two different issues.
We certainly are arguing about two different issues, but I can't seem to nail down what your issues are either.:o( "Once more into the breach..."
the possibility the financial industry may switch to Linux is not PROOF (in itself) that Solaris "opened up" the kernel as a reaction to Linux.
No, not proof. Evidence, but not proof.
Its not even evidence. Keep in mind that the industry can switch to linux without it being about open-source; it can be for other advantages. If it switches, it will be because one platform offers more "bang" for the buck than the other one. It could be because everone else is doing it, and there is a negative effect in going it alone. But is still can be because of those reasons and not have anything to do with open source.
Open source is an industry concept. It exists to allow a certain type of transaction to occur which would not be possible in the legal environment of business today. No business entity wants to "horde" source code IF it doesn't provide them a business advantage. But because of basic capitalist theory and the legal profession making the practice of sharing difficult, there needed to be a new legal doctrine/theory to be able to delineate and secure rights to the vendor and the buyer they felt to be important. Open source came about to be a new legal/business paradigm to supplant legal/business practice of the day.
GPL, is not "open source". GPL secures certain rights to the owner (such as modifications to owners property become property of the owner, user is compelled to make all modifications public if they distribute the owners property, etc.) which open source does not. So when we talk about linux (GPL), we are not necessarily talking about open source philosophy as both of us understand it to be; they can be subtly different.
A couple statements you make seem a bit contradictory to me, like these two:
But I believe Sun would *not* have "open-sourced" Solaris if linux *never* existed.
and... Sun did not "open-source" its code AS A RESPONSE to linux threatening its market...
It seems strange to me that you would find the open sourcing of Solaris to be a reaction to Linux, yet not see it as a defensive move -- in light of your comment that Linux does not (really, CAN NOT) threaten Solaris at the high end, it sort of makes a certain kind of sense.
They are not. My position is that Solaris did not open source as a reaction to linux, they open sourced for different reasons. linux did not CAUSE Solaris to go open source. Sun did it for other reasons.
Different point, if linux did not exist, Solaris could not go open source, because linux helped pioneer the legal concepts of open source. Without linux, there would be no "open source license" for Sun to adopt for its Solaris kernel.
Think of Linux as "The Romans" that "invented" the bridge building. Think of "open source" as the bridge. Think of Sun as "the Egyptians". Sun decides to make a bridge to span a river so they can invade their neighbor, the Carthaginians. Sun could NOT have made that bridge (used opensource) unless Linux existed (to invent it). More important, Sun is building the bridge (using opensource) to attack Carthaginians, not because they feel the need to compete with the Romans. Get it now?
As for the rest of your positions, they attribute a sense of importance to open source, which does not exist. There are no improvements to a piece of software without programmers. If the programmers aren't willing to make improvements that a business desires FOR FREE, then the business extracts no benefit from open sourcing their software. The "network effect" you talk about is more like an environment where software is a form of grease, and opensource is a way to create more grease and share it, to get from point a to point b. But grease in its
You guys are totally missing the point. Whedon didn't incorporate cowboy stories because they sell or the Western is a beloved theme. Hell, no TV Western series could get pilot money from a broadcast network nowadays.
Whedon's goal was to avoid formulaic, crap science fiction perpetuated by the Star Trek factory. The future is not aliens humanoid enough for Terrans to portray them, in a galactic alliance, run by the military with a culture patterned after 19th century navy ships, looking to be do-gooders. (That is not what the Federation taxpayers blew quintillions of credits for.) As much as I loved ST:TOS, it has probably done more to destroy quality sci-fi than any other show. Star Wars is the same pattern, BattleStar Galactica is the same pattern, Space: Above & Beyond(?), same damn military shit. And the one sci-fi show that dared to be different (Babylon 5), got cloned by ST:DS9.
Inbuing a Western theme was a way to differentiate it from the tired Star Trek formula. And it has a lot to be argued for. Space colonization will not be done by rich people looking to spread Western Civilization values across the galaxy. It will be done to make money, by common people looks to strike it rich, seizing an opportunity they can't get in the developed world, or escape the soul sucking crush of "the Man". No saving the galaxy shit, no lets start armed conflict on secular humanist principles, no shagging alien chicks that are biologically closer to an armadillo than an ape. Yeah, sci-fi as a Western is an old theme, but it really wasn't about copying something people don't remember.
Its more like $50K generating 1 million. They show more concern about the net profit than you purport. But you're still better off making 950K with less investment, than 1M with greater risk. Oddly enough, even TV execs are realizing the limits of "reality show" programming. Unfortunately, they are replacing those slots with evening soap operas that are supposed to pattern "Dawson's Creek" season 1. Money is so important, it subverts art.
Eh, I watch too much TV anyway, and advertisers do not care about my demographic. I do more hacking and TV programmers plug-in more profitable programs for pod people. Win win. *sigh*
Your loss. I liked the 2nd season. Joshua was not Jar Jar. My two favorite episodes in the series was in the 2nd season. The only way you can justify dropping DA after the fourth episode is if you dropped Voyager and Enterprise in the same amount of time. Or perhaps your thought patterns closely resemble Fox executives.
I don't even understand why people are in such "mourning" over Firefly. Yes it was good and I liked it, but hey, I preferred ogling Jessica Alba and finding paranoid parallels to our current reality. Firefly seemed to be awfully partial to the losers of the Civil War, and after this year's general election (and the Red Sox beating the Yankees), I really don't see any endearing themes to take from the show. I'd like to think more hetero chicks are on slashdot generating this Firefly > DA vibe, but I'm thinking its probably something else...
Dude, get a grip. "Vastly superior"??? What weed are you smoking? You can get on your soapbox and rattle off details on how firefox is more W3C compliant, etc. etc., but you think the layperson cares??? All they know is the world runs on Microsoft, IE works on web pages that firefox doesn't, and it renders fast. That's all anyone gives a crap about. Firefox's big advantages, popup blocking and I forget the other one, got put into IE via SP2. There's only one aspect that Firefox is measurably better than IE, and that's security. Why good is that if no one is buying? No lame-ass user gives a rat's ass about security. That's why they hit porn sites and install the spyware themselves. Worst of all, if the firefox bug recapped on Slashdot is true, I'll be ripping out Firefox for Mozilla. Twice my machine has mysteriously locked up. It did not even occur to me to associate the problem with Firefox until reading the topic.
Microsoft cares about money, they do not give a sh*t about quality of their code. They have yet to be penalized for that attitude. It would cost them millions to refactor IE, and no one has presented a killer option for web browsing that would make that refactor a selling point for IE.
Improving IE would not make other people think Firefox was inferior anyway. They'd sooner bury Firefox in patent lawsuits; that would give them effective marketing against Firefox. Thankfully, this is all moot. I don't believe M$ cares if Firefox ate up a larger share of the browser market.
Microsoft developed IE as a direct response to Netscape (and the Internet). Netscape's CEO (Barksdale? or Andreessen?) was warbling about making Ihe internet the personal computer, with Netscape becoming the gui interface. Given the big booming unknown that was the Internet, Gates decided that he had to dominate the Internet and cut the legs out from under Netscape. The lynchpin of that strategy was putting out a competing web browser, and distributing it for free. Obviously, in order to take market share, M$ had to keep pumping out improvements into IE.
Fast forward to 2004. Netscape is long dead. The Internet is not the desktop computer OS of the general public. The M$ OS still dominate the desktop market. The dire necessity to keep improving IE is long gone, as is its ability to leverage advantage on the Internet battlefield is minimal at this point. Firefox succeeding, having a variety of webbrowsers to utilize may be of significant impact for users, but I don't see why Microsoft would care enough to plow more money into the product.
It is ironic when taken from the mythical perspective of "the Establishment"; who adjudicated a legal precident to spy on the unwashed masses with impunity, only to later see the decision used by a member of the unwashed masses to spy on "it" with impunity.
(And I only belabor the obvious because its apparent so many people need to buy a clue.)
NO ONE in the financial industry is EVER going to consider switching from Solaris to Linux, and therefore the opening of Solaris cannot POSSIBLY be seen as a reaction to Linux?
Before I start, let me point out that the possibility the financial industry may switch to Linux is not PROOF (in itself) that Solaris "opened up" the kernel as a reaction to Linux.
Let me further qualify my position by saying I don't think its *impossible* for linux to supplant Solaris in the "big iron" environment Sun currently caters to. What I will say is that Linux (and its kernel) will have to make changes to support application architectures popular to those industries. Because I do not see the kind of changes being planned *now* with current kernel development, I do not believe certain goals will be realized within two years from now. Therefore I don't believe Linux will be replacing Solaris in the "big iron" arena anytime soon.
First, to belabor the obvious, financial institutions have different computational requirements than the linux hobbyist. They HAVE to have a high degree of reliability, they need to operate "efficiently", and they need to do it as "cost-effective" a manner as possible. Cheapest hardware, employee, or less of each is not the most cost effective approach.
Its an industry given, there is going to be a certain level of computational infrastructure needed to be able to "reliably" process the X millions of transactions and information for every customer. This means they're forced to hire a certain number of meatbags (that's mostly us) to maintain this infrastructure. Meatbags are currently very expensive compared to hardware. So anything that allows the least amount of meatbags and still allow them to achieve their processing requirements is their desired computing platform. Sometimes they will look at purchasing a system that will allow them to pay for cheaper meatbags (but Microsoft was not sucessful in the niche I'm talking about). Sometimes they look towards buying software/hardware that can take over some the jobs, thus the percentage of time saved means the less meatbags are needed to meet the requirements.
The idea of "grid" linux replacing Solaris is not convincing to me. Granted, you don't need to find that "small", esoteric pool of Solaris meatbags. And granted Sun hardware & support contracts suck a lot more money than PCs. But it just may be that expensive hardware, with expensive support, and a FEW expensive meatbags result in a cheaper, more reliable system than a "grid" linux infrastructure. "Grid" linux *will* replace Solaris, but *only* when its cheaper than Solaris with the same or better performance goals.
Now look at the problems of "grid" linux. When component X fails, it has to be detected and corrected in a punctual manner. In order to have a computing/reliability advantage over current Sun architectures, you need a lot of machines. That means statistically, you are guaranteed X number of failures, and thus X numbers of full-time salaried meatbags will be needed to correct the problem. Recall the vaccuum tube mainframes of the 40's & 50's, and you may start to get an idea of the scope of the problem. Then also consider, 1000 PCs consume a hell of a lot more power, and still require A/C for the server room, so that's an infrastructural cost to be considered.
No, the market opening will be to "consolidate" as many computing platforms as possible, to hire as few meatbags as possible. Also realize that 100 cpus on a miniframe like Sun's products gives you the same kind of distributed processing configuration as 100 PC boxes. Solaris software
is designed to "make pretend" its 100 little boxes, to monitor those little boxes, and to even distribute those computing tasks over *all* those CPUs. So even when you have 1000 different tasks on 500 boxes, all the Sun CPUs are working constantly. It then has added architectural advantages, like bus communication (much faster than netw
The USA has used the Echelon global surveillance system for the purposes of industrial espionage, to give its failing corporations an unfair advantage over more-competitive foreign operations.
Oh, cut the crap you French/Israeli hypocrite! Its documented how your nations use its national intelligence collection agencies to benefit national industries.
At least produce some proof echelon is being used to extract economic intelligence for non-gov't corporations.
That is based on the presumption that if you're richer, you can pay more tax and not experience as negative results as a poor person ("progressive" tax policy).
But we're not talking about taxes, we're talking about denial of technology that generates CO2 in order to reduce aggregate CO2 production by a country. That is pegged by physical laws, not subjective perception of financial wellbeing.
Its not fair if the US would have to sacrifice all its cars in order to meet the same CO2 production as France. (Granted, requirements are per capita related, so its not quite as bad, but you get the picture (if you have a brain).)
What makes Kyoto UTTERLY pointless is that China is not being held to any MEANINGFUL standard to control its CO2 production. ("I'm a poor country, I'm entitled to pollute more...") Anyone with an ounce of common sense knows that as China grows to be in the top 5 production economies, it must produce a CO2 level commesurate to what a top 5 economy will produce. Do I think China must be compelled to match CO2 generation standards AS IT GROWS? No. But at a certain point, China would have to be compelled to meet standards commesurate to an industrialized nation. China refuses to agree to anything would result in "capping" its economic growth (until its as economically productive as the US).
Look, you semi-educated, granola-munching, earthy hippies have to realize that half measures won't result in jack squat. Eventual 50% reduction in CO2 generation does not mean the problem presented by global warming will go away or be ameliorated. At very least, scientists need to show PROOF that will be the case. Logical presumptions are meaningless in a physical world; the world runs on physical laws. (Sometimes relationships are logarithmic in nature, or a hard set number.) What is certain is that following a law like Kyoto will have definite short-term economically negative results. Why should Europe and the US face an economic depression THAT WILL NOT FIX ANYTHING BECAUSE the 3rd world will take up our CO2 generation?
I'm not an opponent of Kyoto on principle; only as its currently formulated. I'm not devastated that Bush won't sign Kyoto. I am displeased he won't continue to engage in negotiation to design an agreement that could be meaningful and "reasonable". But this is a zit level irritation compared to the destructive policies he champions. I don't set expectations for a chimp to do calculus. Its a waste of time to expect the chimp in office to be concerned about the environment.
Not if a majority of the WTO members vote to enforce the tariff.
Re:Still can't see how Sun will survive
on
Sun-isms Debunked
·
· Score: 1
Then its 5 boxes here, 3 boxes there, etc. etc. that match the same requirements. Instead of scaling with 1U boxes, scale with adding CPUs, memory, and drive space. You're mindlocked in 2004. Its not going to stay 2004, nor will the technology, or methods.
Re:You do not understand server farms at all
on
Sun-isms Debunked
·
· Score: 1
Disks fail. Networks get flooded with crap packets and DOS attacks. NICs get flooded with outbound traffic.
And you're telling me lots of PCs handle these situations better than a server with as many CPUs, that can hot swap failing hardware, and the advantage of using the bus for communications, rather than individual network interfaces that can only communicate at a maximal GB/sec? What the heck do you think big iron vendors are doing with grid computing and virtual domains? MOST important, is it more COST effective to have more personnel herding around those PCs, expending power to maintain those machines, etc. etc.
There are REASONS people use massively redundant arrays of rack PCs. Do you think we have these setups just because the first PC is cheap?
No, I think those setups came about because they evolved to address operation requirements and costs. And no, Sun/IBM servers for such an enterprise were not cost effective, or capable for dealing with all requirements THEN (and perhaps now).
But now you have warehouses full of running PCs sucking up lots of power, and requiring a certain level of staffing to maintain. Why purchase 30 boxes and requisite networking hardware when you can purchase one server with 30 CPUs, RAID, and enough multiple network interfaces to provide the same bandwidth? There were a ton of individual deficiencies in Sun servers that prevented this from happening circa 2000.
But Sun can only sell big hardware in the nearish future. They have to invest in improving scalability anyway. All they need to do is produce a midsized server that can scale like a rack of PCs and make it cost effective. If its technologically *IMPOSSIBLE* for this to come about, then it won't happen. What I'm seeing is that server farms are evolving towards cost efficiency, and consolidation is a way to achieve this. Sun can either see an opportunity and move to exploit it, or sit dead in the water.
Re:Still can't see how Sun will survive
on
Sun-isms Debunked
·
· Score: 1
Actually IBM manages direction by providing the developers to provide the new kernel code. Linus is an adept project manager, but he is not going to unwilling incorporate changes into *his* kernel because the guys with money want to shove the changes down his throat. From my point of view, its not a matter whether Linux will incorporate changes, its whether it will evolve fast enough to suit commercial interest's long term plans. I don't think there will be significant scalability or more efficient threading without a redesign of the kernel. I see progress, I see building a skyscraper on a shaky foundation that will need to be rebuilt, but I don't see 2.6 development or 2.7 development changing the way it needs to be changed to start competing with Solaris.
Re:From the memory hole...
on
Sun-isms Debunked
·
· Score: 3, Insightful
If this is true, and also, as you posted earlier, that "big financial institutions (or software houses that cater to them) have access to the kernel source so they can add value to the kernel", then wasn't their ability to do that with Linux, and NOT Solaris before now, one of the possible reasons for them to switch to Linux?
Because Linux is not Solaris. It does not support hardware clustering or scale beyond 4 CPUs, it is not as efficient with threaded applications, it is not as reliable as Solaris for transactional processing, etc. etc. The reality is that Linux was hobby kernel, designed ad-hoc, and does not match the quality of a sucessful commercial one (Sun,IBM, etc.). Furthermore, it will not be able to do those cool things that Solaris does without a total redesign. Torvalds, if you been following Linux's evolution, tends to set conservative goals with each kernel change, partial towards monolithic kernel design, and is not predisposed towards favoring commercial vendor's goals. You get a pretty good kernel for a standalone PC, but its unlikely that it will go beyond that until someone big (IBM) forks the kernel towards goals favoring enterprise hardware.
Financial institutions do not want to sink money into something that will have to be redesigned to support big iron features, and sit indefinitely hoping Torvalds will accept their kernel changes. Nor will they want to support an effort which would have them "fork" the kernel and then have to hand over any changes to their competitors (GPL).
You seem to be confusing us saying "compete with Linux in one of Sun's key markets", which Solaris IS NOW doing, with "become exactly like Linux for EVERY market", which Solaris is oviously NEVER going to do.
I can't discern what you claim I am confusing because your statement makes no sense. Provide a context, regurgitate relevant statements.
Riiight. So, this is different from Linux how, exactly? Linux code is open, therefore large firms can add value to it. Now the same can be said about Solaris. By leveling this playing field, you're claiming this isn't an attempt to compete how exactly?
Sun isn't attempting to compete with Linux by opensourcing its kernel. Opensourcing Sun's kernel will not drive down its development/support costs or harvest linux's GPL advantages. It will keep them more closely tied to their big-ticket customers. It enables a startup to use its expertise to develop a product that will support Sun's product line. Its not a response specifically targetted to address linux.
SCO's legal expense capping is merely a memorandum of understanding between SCO and its bank rollers. (The real news was that they're going to use stock to help compensate it's legal firms. That extends their ability to extend their legal offensive.) The alternative for Sun not paying off SCO was to halt its strategy open sourcing its kernel. It wasn't the make or break financing that allowed SCO to proceed with its lawsuits. Sun didn't create an alibi so they could attack linux using SCO. And even if it was more than a million, its still chickenfeed compared to what IBM could pony up in its legal defense. The amusing thing is that I believe IBM is playing with SCO with its legal defense. IBM *wants* a trial so they can set legal precident to solidify linux's legal issues.
Re:Shoot your marketing department.
on
Sun-isms Debunked
·
· Score: 1
Word. Can't say your opinion converges mine 100%, but its the best response I've seen here today.
Re:Still can't see how Sun will survive
on
Sun-isms Debunked
·
· Score: 4, Insightful
Why not?
- for every useful feature that Sun adds in, someone in Linux-land will eventually see that feature as a good thing and work will be done to port that feature to Linux. The porting to Linux of an existing Sun feature can be done faster than Sun can think up and build new features, and as Linux pushes more and more into the enterprise, the focus will become more and more on replicating
Sun's advantages in Linux.
That is presuming that realizing that feature will not require redesigning the linux kernel to implement it. (No chance of that happening quickly until IBM can sucessfully fork linux.)
- to a very large extent, you can achieve uptime by scaling "wide" i.e. throwing more boxes at the problem. It's absolutely not a panacea to all uptime issues, but it's an approach that fits particularly well with Linux/Intel due to the low incremental cost of the hardware. Whatever "uptime smarts" Sun can add to their OS, I and many others can achieve the same results (in pure uptime terms) by bolting a bunch of new Intel boxes into a rack
The key to that strategy is that everything Sun can do with its upscalable platforms can be matched by linux running on another box. That is just not the case. You're enhancing reliablity by adding another point of failure? It may be possible to add redundancy to improve uptime, but that doesn't come without a physical cost. And how are those boxes going to consume less power than an integrated server?
Don't be shocked if five years from now, PC's aren't used at server farms. Why have thousands of PCs running linux, consuming all that electricity in computing and air conditioning, and physical space? Instead, have 5 "Sun Server Bazillion"s. You need more computing power, slap in a hotpluggable CPU, rather than another PC machine. No need to implement a networking grid for all those PCs. The only networking needed is the server to the outside world router. Have two-four overpaid sysadmins or a battery of employee salaries to maintain a battery of PCs
In piecemeal ways, webserver companies are already moving this way with low powered CPUs and fiddling with "blade" machines. A smart marketing team with a smart engineering team could easily bring Sun back into the server market. Not the mom & pop ISPs, but the AOLs and Verisigns of the world. Their problem is that their hardware is not quite designed to hotswap CPUs and memory like hard drives, they haven't configured a software product to realize this vision, their OS is still relatively esoteric, and they margin themselves out of profitability. But none of those things are impossible to correct.
Slashdot Mirror
Guilty as charged. You probably wouldn't think so from this post
Interesting too. Not a political scientist; wasn't aware of that sort of nuance in the definition of liberalism.
To further compound the irony, I consider myself more liberal in political orientation. While I consider most American conservatives (that would vote for GWB) as blustering, whiney, cheerleading morons, I do think that American liberals will need to rehabilitate their image by precisely defining the label in more pragmatic terms and letting go of their policy failures. Gov't can't end poverty by wealth redistribution. LBJ's Great Society was a failure as a "liberal" program. Stop depending on traditional gov't structures to correct societal problems (except health care).
Example: if you're going to push for things like school lunch subsidy programs, make sure you point out how relatively inexpensive it is, and its results. Then do not go and say if you have feeding programs for anyone poor, that you'll get the same results and it will be worthwhile. Any "liberal" that does so needs to be verbally abused, badly. Or, carefully point out that the "progressive" aspects of tax policy is a good, "liberal" thing, and conservatives are thieves that don't want to pay any taxes, just stick the bill on the non-rich. (Assuming of course, one doesn't "want" a flat tax policy.) Start attacking gov't subsidy of corporations, and attach that thievery as "conservative". (Of course it won't happen, because the system is rigged.)
Nope. The rocket is only engineered for what is attached. There is a base in place to perhaps attach a few more rocket motors. But more components means a greater percentage of failure. This route has been taken before with the Soviets.
The sad thing is that it will still be cheaper just to pay the Russians to throw something up into space. Perhaps the more realistic and effective use for NASA's money would be to invest and manage (by customer requirements) the operations end of Russia's launch facilities. In other words, outsource.
You still need to put money into a D-9 program for domestic considerations (the military). Also, with an active rocket program, you have an infrastructure to propel development of rockets to take care of unique missions which the Soviets would not be interested in developing.
The worst thing is that the average American voter is a crippled mind. Not just is it substandard in science knowlege, but now Americans lack imagination and vision. The answer to a future space launch platform is not rockets. Its the space elevator. You're not going to colonize planets or develop an extended presence in space with rockets.
Its as doable now as thermonuclear weapons were back in the '40's. NASA should outsource launch facilities, maintain its current scientific missions, and put the bulk of its money into engineering a space elevator. If its too expensive, get the rest of the world to kick in for its development. Get those cheap Indians and Russians crunching out the numbers, and let the Americans specialize in the design and engineering.
Even if you keep the project in-house, think of the boon it would be for American infrastructure. Hey MIT, Caltech, Los Alamos, etc., here's a billion dollars, go engineer a space elevator. Then put it out to bid for Lockheed, Boeing, etc. to actually build it. Nope, Americans are too stupid to see the economic utility and importance of scientific investment. No vision.
Blah, blah, blah. And the Empire State Building is now the tallest skyscraper in Manhattan. Meaningless factoid. Guess what, eventually there will be another building which will be taller. The answer is to go build it. 23 ton PAYLOAD. You merely design a new capsule to put on top of the rocket. Hell, probably could kludge a Soyuz to fit on top of it.
Yes, but you don't need the shuttle to do that. Manned capsule on rocket can do the same damn thing. Why are you so in love with something that crippled the US space agenda? You're like a liberal who thinks the gov't can end poverty by wealth redistribution.
You're talking out of your ass. There hasn't ever been a shuttle mission which required taking a satellite out of orbit and landing it on earth.
There isn't any utility in doing so either. Its cheaper to send up a new satellite.
Shuttle was an engineering marvel, but a white elephant failure. Disposable rockets are cheaper, and it sucked out all that money that could have been used for a manned Mars mission, or a "useful" space station.
Apparently, you've never seen cautionary tales as "Attack of the Killer Tomatoes". Of course, after seeing "The Toxic Avenger" I'd take irradiation over chemical preservatives any day...)
Tenure is a big problem, and its ridiculous practice to apply to primary school educators. The whole point of tenure was to allow researchers with a proven record of academic competence to present unpopular fruits of research without getting fired for presenting the new idea. Primary school educators do not present new research to high school children, and technically, they're not supposed to be teaching anything which contradicts the prerogative of the local school board. Unfortunately, its been a contractual practice to incorporate tenure at that instruction level because of convention, so removing it becomes a sticking point in contract negotiations. Thus, we get stuck with underperformers that we cannot remove.
Don't forget our legal system for contributing to this decline. A lot of destructive policy gets put into place to placate the leeches in suits. And finally, yes, the parents are probably the biggest problem. But there's no means testing in handing out marriage licenses or pregnancy. Its also ridiculous not be able to expel students with sufficent cause. On the other hand, you have idiot administrators expelling kids for possession of aspirin or a t-shirt.
The real problem is we live in a country who's people are stupid enough to vote for Bush.
This is quaint nostalgia from limited imaginations. What you want is The Matrix.
With holograms, you can see it. But you can't smell, taste, or touch it.
Yeah, watching Angelina Jolie strip in 3 dimensions is sweet, but with the Matrix, you can be doing AJ.
Take it from Cypher:
You make some great points, but we seem to be arguing about two different issues.
We certainly are arguing about two different issues, but I can't seem to nail down what your issues are either. :o( "Once more into the breach..."
the possibility the financial industry may switch to Linux is not PROOF (in itself) that Solaris "opened up" the kernel as a reaction to Linux.
No, not proof. Evidence, but not proof.
Its not even evidence. Keep in mind that the industry can switch to linux without it being about open-source; it can be for other advantages. If it switches, it will be because one platform offers more "bang" for the buck than the other one. It could be because everone else is doing it, and there is a negative effect in going it alone. But is still can be because of those reasons and not have anything to do with open source.
Open source is an industry concept. It exists to allow a certain type of transaction to occur which would not be possible in the legal environment of business today. No business entity wants to "horde" source code IF it doesn't provide them a business advantage. But because of basic capitalist theory and the legal profession making the practice of sharing difficult, there needed to be a new legal doctrine/theory to be able to delineate and secure rights to the vendor and the buyer they felt to be important. Open source came about to be a new legal/business paradigm to supplant legal/business practice of the day.
GPL, is not "open source". GPL secures certain rights to the owner (such as modifications to owners property become property of the owner, user is compelled to make all modifications public if they distribute the owners property, etc.) which open source does not. So when we talk about linux (GPL), we are not necessarily talking about open source philosophy as both of us understand it to be; they can be subtly different.
A couple statements you make seem a bit contradictory to me, like these two: But I believe Sun would *not* have "open-sourced" Solaris if linux *never* existed. and ... Sun did not "open-source" its code AS A RESPONSE to linux threatening its market...
It seems strange to me that you would find the open sourcing of Solaris to be a reaction to Linux, yet not see it as a defensive move -- in light of your comment that Linux does not (really, CAN NOT) threaten Solaris at the high end, it sort of makes a certain kind of sense.
They are not. My position is that Solaris did not open source as a reaction to linux, they open sourced for different reasons. linux did not CAUSE Solaris to go open source. Sun did it for other reasons.
Different point, if linux did not exist, Solaris could not go open source, because linux helped pioneer the legal concepts of open source. Without linux, there would be no "open source license" for Sun to adopt for its Solaris kernel.
Think of Linux as "The Romans" that "invented" the bridge building. Think of "open source" as the bridge. Think of Sun as "the Egyptians". Sun decides to make a bridge to span a river so they can invade their neighbor, the Carthaginians. Sun could NOT have made that bridge (used opensource) unless Linux existed (to invent it). More important, Sun is building the bridge (using opensource) to attack Carthaginians, not because they feel the need to compete with the Romans. Get it now?
As for the rest of your positions, they attribute a sense of importance to open source, which does not exist. There are no improvements to a piece of software without programmers. If the programmers aren't willing to make improvements that a business desires FOR FREE, then the business extracts no benefit from open sourcing their software. The "network effect" you talk about is more like an environment where software is a form of grease, and opensource is a way to create more grease and share it, to get from point a to point b. But grease in its
You guys are totally missing the point. Whedon didn't incorporate cowboy stories because they sell or the Western is a beloved theme. Hell, no TV Western series could get pilot money from a broadcast network nowadays.
Whedon's goal was to avoid formulaic, crap science fiction perpetuated by the Star Trek factory. The future is not aliens humanoid enough for Terrans to portray them, in a galactic alliance, run by the military with a culture patterned after 19th century navy ships, looking to be do-gooders. (That is not what the Federation taxpayers blew quintillions of credits for.) As much as I loved ST:TOS, it has probably done more to destroy quality sci-fi than any other show. Star Wars is the same pattern, BattleStar Galactica is the same pattern, Space: Above & Beyond(?), same damn military shit. And the one sci-fi show that dared to be different (Babylon 5), got cloned by ST:DS9.
Inbuing a Western theme was a way to differentiate it from the tired Star Trek formula. And it has a lot to be argued for. Space colonization will not be done by rich people looking to spread Western Civilization values across the galaxy. It will be done to make money, by common people looks to strike it rich, seizing an opportunity they can't get in the developed world, or escape the soul sucking crush of "the Man". No saving the galaxy shit, no lets start armed conflict on secular humanist principles, no shagging alien chicks that are biologically closer to an armadillo than an ape. Yeah, sci-fi as a Western is an old theme, but it really wasn't about copying something people don't remember.
Its more like $50K generating 1 million. They show more concern about the net profit than you purport. But you're still better off making 950K with less investment, than 1M with greater risk.
Oddly enough, even TV execs are realizing the limits of "reality show" programming. Unfortunately, they are replacing those slots with evening soap operas that are supposed to pattern "Dawson's Creek" season 1. Money is so important, it subverts art.
Eh, I watch too much TV anyway, and advertisers do not care about my demographic. I do more hacking and TV programmers plug-in more profitable programs for pod people. Win win. *sigh*
Your loss. I liked the 2nd season. Joshua was not Jar Jar. My two favorite episodes in the series was in the 2nd season. The only way you can justify dropping DA after the fourth episode is if you dropped Voyager and Enterprise in the same amount of time. Or perhaps your thought patterns closely resemble Fox executives.
I don't even understand why people are in such "mourning" over Firefly. Yes it was good and I liked it, but hey, I preferred ogling Jessica Alba and finding paranoid parallels to our current reality. Firefly seemed to be awfully partial to the losers of the Civil War, and after this year's general election (and the Red Sox beating the Yankees), I really don't see any endearing themes to take from the show. I'd like to think more hetero chicks are on slashdot generating this Firefly > DA vibe, but I'm thinking its probably something else...
Dude, get a grip. "Vastly superior"??? What weed are you smoking? You can get on your soapbox and rattle off details on how firefox is more W3C compliant, etc. etc., but you think the layperson cares??? All they know is the world runs on Microsoft, IE works on web pages that firefox doesn't, and it renders fast. That's all anyone gives a crap about. Firefox's big advantages, popup blocking and I forget the other one, got put into IE via SP2. There's only one aspect that Firefox is measurably better than IE, and that's security. Why good is that if no one is buying? No lame-ass user gives a rat's ass about security. That's why they hit porn sites and install the spyware themselves. Worst of all, if the firefox bug recapped on Slashdot is true, I'll be ripping out Firefox for Mozilla. Twice my machine has mysteriously locked up. It did not even occur to me to associate the problem with Firefox until reading the topic.
Microsoft cares about money, they do not give a sh*t about quality of their code. They have yet to be penalized for that attitude. It would cost them millions to refactor IE, and no one has presented a killer option for web browsing that would make that refactor a selling point for IE.
Improving IE would not make other people think Firefox was inferior anyway. They'd sooner bury Firefox in patent lawsuits; that would give them effective marketing against Firefox. Thankfully, this is all moot. I don't believe M$ cares if Firefox ate up a larger share of the browser market.
Microsoft developed IE as a direct response to Netscape (and the Internet). Netscape's CEO (Barksdale? or Andreessen?) was warbling about making Ihe internet the personal computer, with Netscape becoming the gui interface. Given the big booming unknown that was the Internet, Gates decided that he had to dominate the Internet and cut the legs out from under Netscape. The lynchpin of that strategy was putting out a competing web browser, and distributing it for free. Obviously, in order to take market share, M$ had to keep pumping out improvements into IE. Fast forward to 2004. Netscape is long dead. The Internet is not the desktop computer OS of the general public. The M$ OS still dominate the desktop market. The dire necessity to keep improving IE is long gone, as is its ability to leverage advantage on the Internet battlefield is minimal at this point. Firefox succeeding, having a variety of webbrowsers to utilize may be of significant impact for users, but I don't see why Microsoft would care enough to plow more money into the product.
(And I only belabor the obvious because its apparent so many people need to buy a clue.)
Before I start, let me point out that the possibility the financial industry may switch to Linux is not PROOF (in itself) that Solaris "opened up" the kernel as a reaction to Linux.
Let me further qualify my position by saying I don't think its *impossible* for linux to supplant Solaris in the "big iron" environment Sun currently caters to. What I will say is that Linux (and its kernel) will have to make changes to support application architectures popular to those industries. Because I do not see the kind of changes being planned *now* with current kernel development, I do not believe certain goals will be realized within two years from now. Therefore I don't believe Linux will be replacing Solaris in the "big iron" arena anytime soon.
First, to belabor the obvious, financial institutions have different computational requirements than the linux hobbyist. They HAVE to have a high degree of reliability, they need to operate "efficiently", and they need to do it as "cost-effective" a manner as possible. Cheapest hardware, employee, or less of each is not the most cost effective approach.
Its an industry given, there is going to be a certain level of computational infrastructure needed to be able to "reliably" process the X millions of transactions and information for every customer. This means they're forced to hire a certain number of meatbags (that's mostly us) to maintain this infrastructure. Meatbags are currently very expensive compared to hardware. So anything that allows the least amount of meatbags and still allow them to achieve their processing requirements is their desired computing platform. Sometimes they will look at purchasing a system that will allow them to pay for cheaper meatbags (but Microsoft was not sucessful in the niche I'm talking about). Sometimes they look towards buying software/hardware that can take over some the jobs, thus the percentage of time saved means the less meatbags are needed to meet the requirements.
The idea of "grid" linux replacing Solaris is not convincing to me. Granted, you don't need to find that "small", esoteric pool of Solaris meatbags. And granted Sun hardware & support contracts suck a lot more money than PCs. But it just may be that expensive hardware, with expensive support, and a FEW expensive meatbags result in a cheaper, more reliable system than a "grid" linux infrastructure. "Grid" linux *will* replace Solaris, but *only* when its cheaper than Solaris with the same or better performance goals.
Now look at the problems of "grid" linux. When component X fails, it has to be detected and corrected in a punctual manner. In order to have a computing/reliability advantage over current Sun architectures, you need a lot of machines. That means statistically, you are guaranteed X number of failures, and thus X numbers of full-time salaried meatbags will be needed to correct the problem. Recall the vaccuum tube mainframes of the 40's & 50's, and you may start to get an idea of the scope of the problem. Then also consider, 1000 PCs consume a hell of a lot more power, and still require A/C for the server room, so that's an infrastructural cost to be considered.
No, the market opening will be to "consolidate" as many computing platforms as possible, to hire as few meatbags as possible. Also realize that 100 cpus on a miniframe like Sun's products gives you the same kind of distributed processing configuration as 100 PC boxes. Solaris software is designed to "make pretend" its 100 little boxes, to monitor those little boxes, and to even distribute those computing tasks over *all* those CPUs. So even when you have 1000 different tasks on 500 boxes, all the Sun CPUs are working constantly. It then has added architectural advantages, like bus communication (much faster than netw
The USA has used the Echelon global surveillance system for the purposes of industrial espionage, to give its failing corporations an unfair advantage over more-competitive foreign operations.
Oh, cut the crap you French/Israeli hypocrite! Its documented how your nations use its national intelligence collection agencies to benefit national industries.
At least produce some proof echelon is being used to extract economic intelligence for non-gov't corporations.
(No argument with the other statements...)
That is based on the presumption that if you're richer, you can pay more tax and not experience as negative results as a poor person ("progressive" tax policy).
But we're not talking about taxes, we're talking about denial of technology that generates CO2 in order to reduce aggregate CO2 production by a country. That is pegged by physical laws, not subjective perception of financial wellbeing.
Its not fair if the US would have to sacrifice all its cars in order to meet the same CO2 production as France. (Granted, requirements are per capita related, so its not quite as bad, but you get the picture (if you have a brain).)
What makes Kyoto UTTERLY pointless is that China is not being held to any MEANINGFUL standard to control its CO2 production. ("I'm a poor country, I'm entitled to pollute more...") Anyone with an ounce of common sense knows that as China grows to be in the top 5 production economies, it must produce a CO2 level commesurate to what a top 5 economy will produce. Do I think China must be compelled to match CO2 generation standards AS IT GROWS? No. But at a certain point, China would have to be compelled to meet standards commesurate to an industrialized nation. China refuses to agree to anything would result in "capping" its economic growth (until its as economically productive as the US).
Look, you semi-educated, granola-munching, earthy hippies have to realize that half measures won't result in jack squat. Eventual 50% reduction in CO2 generation does not mean the problem presented by global warming will go away or be ameliorated. At very least, scientists need to show PROOF that will be the case. Logical presumptions are meaningless in a physical world; the world runs on physical laws. (Sometimes relationships are logarithmic in nature, or a hard set number.) What is certain is that following a law like Kyoto will have definite short-term economically negative results. Why should Europe and the US face an economic depression THAT WILL NOT FIX ANYTHING BECAUSE the 3rd world will take up our CO2 generation?
I'm not an opponent of Kyoto on principle; only as its currently formulated. I'm not devastated that Bush won't sign Kyoto. I am displeased he won't continue to engage in negotiation to design an agreement that could be meaningful and "reasonable". But this is a zit level irritation compared to the destructive policies he champions. I don't set expectations for a chimp to do calculus. Its a waste of time to expect the chimp in office to be concerned about the environment.
Not if a majority of the WTO members vote to enforce the tariff.
Then its 5 boxes here, 3 boxes there, etc. etc. that match the same requirements. Instead of scaling with 1U boxes, scale with adding CPUs, memory, and drive space. You're mindlocked in 2004. Its not going to stay 2004, nor will the technology, or methods.
Disks fail. Networks get flooded with crap packets and DOS attacks. NICs get flooded with outbound traffic.
And you're telling me lots of PCs handle these situations better than a server with as many CPUs, that can hot swap failing hardware, and the advantage of using the bus for communications, rather than individual network interfaces that can only communicate at a maximal GB/sec? What the heck do you think big iron vendors are doing with grid computing and virtual domains? MOST important, is it more COST effective to have more personnel herding around those PCs, expending power to maintain those machines, etc. etc.
There are REASONS people use massively redundant arrays of rack PCs. Do you think we have these setups just because the first PC is cheap?
No, I think those setups came about because they evolved to address operation requirements and costs. And no, Sun/IBM servers for such an enterprise were not cost effective, or capable for dealing with all requirements THEN (and perhaps now).
But now you have warehouses full of running PCs sucking up lots of power, and requiring a certain level of staffing to maintain. Why purchase 30 boxes and requisite networking hardware when you can purchase one server with 30 CPUs, RAID, and enough multiple network interfaces to provide the same bandwidth? There were a ton of individual deficiencies in Sun servers that prevented this from happening circa 2000.
But Sun can only sell big hardware in the nearish future. They have to invest in improving scalability anyway. All they need to do is produce a midsized server that can scale like a rack of PCs and make it cost effective. If its technologically *IMPOSSIBLE* for this to come about, then it won't happen. What I'm seeing is that server farms are evolving towards cost efficiency, and consolidation is a way to achieve this. Sun can either see an opportunity and move to exploit it, or sit dead in the water.
Actually IBM manages direction by providing the developers to provide the new kernel code. Linus is an adept project manager, but he is not going to unwilling incorporate changes into *his* kernel because the guys with money want to shove the changes down his throat. From my point of view, its not a matter whether Linux will incorporate changes, its whether it will evolve fast enough to suit commercial interest's long term plans. I don't think there will be significant scalability or more efficient threading without a redesign of the kernel. I see progress, I see building a skyscraper on a shaky foundation that will need to be rebuilt, but I don't see 2.6 development or 2.7 development changing the way it needs to be changed to start competing with Solaris.
If this is true, and also, as you posted earlier, that "big financial institutions (or software houses that cater to them) have access to the kernel source so they can add value to the kernel", then wasn't their ability to do that with Linux, and NOT Solaris before now, one of the possible reasons for them to switch to Linux?
Because Linux is not Solaris. It does not support hardware clustering or scale beyond 4 CPUs, it is not as efficient with threaded applications, it is not as reliable as Solaris for transactional processing, etc. etc. The reality is that Linux was hobby kernel, designed ad-hoc, and does not match the quality of a sucessful commercial one (Sun,IBM, etc.). Furthermore, it will not be able to do those cool things that Solaris does without a total redesign. Torvalds, if you been following Linux's evolution, tends to set conservative goals with each kernel change, partial towards monolithic kernel design, and is not predisposed towards favoring commercial vendor's goals. You get a pretty good kernel for a standalone PC, but its unlikely that it will go beyond that until someone big (IBM) forks the kernel towards goals favoring enterprise hardware.
Financial institutions do not want to sink money into something that will have to be redesigned to support big iron features, and sit indefinitely hoping Torvalds will accept their kernel changes. Nor will they want to support an effort which would have them "fork" the kernel and then have to hand over any changes to their competitors (GPL).
You seem to be confusing us saying "compete with Linux in one of Sun's key markets", which Solaris IS NOW doing, with "become exactly like Linux for EVERY market", which Solaris is oviously NEVER going to do.
I can't discern what you claim I am confusing because your statement makes no sense. Provide a context, regurgitate relevant statements.
Riiight. So, this is different from Linux how, exactly? Linux code is open, therefore large firms can add value to it. Now the same can be said about Solaris. By leveling this playing field, you're claiming this isn't an attempt to compete how exactly?
Sun isn't attempting to compete with Linux by opensourcing its kernel. Opensourcing Sun's kernel will not drive down its development/support costs or harvest linux's GPL advantages. It will keep them more closely tied to their big-ticket customers. It enables a startup to use its expertise to develop a product that will support Sun's product line. Its not a response specifically targetted to address linux.
SCO's legal expense capping is merely a memorandum of understanding between SCO and its bank rollers. (The real news was that they're going to use stock to help compensate it's legal firms. That extends their ability to extend their legal offensive.) The alternative for Sun not paying off SCO was to halt its strategy open sourcing its kernel. It wasn't the make or break financing that allowed SCO to proceed with its lawsuits. Sun didn't create an alibi so they could attack linux using SCO. And even if it was more than a million, its still chickenfeed compared to what IBM could pony up in its legal defense. The amusing thing is that I believe IBM is playing with SCO with its legal defense. IBM *wants* a trial so they can set legal precident to solidify linux's legal issues.
Word. Can't say your opinion converges mine 100%, but its the best response I've seen here today.
Why not? - for every useful feature that Sun adds in, someone in Linux-land will eventually see that feature as a good thing and work will be done to port that feature to Linux. The porting to Linux of an existing Sun feature can be done faster than Sun can think up and build new features, and as Linux pushes more and more into the enterprise, the focus will become more and more on replicating Sun's advantages in Linux.
That is presuming that realizing that feature will not require redesigning the linux kernel to implement it. (No chance of that happening quickly until IBM can sucessfully fork linux.)
- to a very large extent, you can achieve uptime by scaling "wide" i.e. throwing more boxes at the problem. It's absolutely not a panacea to all uptime issues, but it's an approach that fits particularly well with Linux/Intel due to the low incremental cost of the hardware. Whatever "uptime smarts" Sun can add to their OS, I and many others can achieve the same results (in pure uptime terms) by bolting a bunch of new Intel boxes into a rack
The key to that strategy is that everything Sun can do with its upscalable platforms can be matched by linux running on another box. That is just not the case. You're enhancing reliablity by adding another point of failure? It may be possible to add redundancy to improve uptime, but that doesn't come without a physical cost. And how are those boxes going to consume less power than an integrated server?
Don't be shocked if five years from now, PC's aren't used at server farms. Why have thousands of PCs running linux, consuming all that electricity in computing and air conditioning, and physical space? Instead, have 5 "Sun Server Bazillion"s. You need more computing power, slap in a hotpluggable CPU, rather than another PC machine. No need to implement a networking grid for all those PCs. The only networking needed is the server to the outside world router. Have two-four overpaid sysadmins or a battery of employee salaries to maintain a battery of PCs
In piecemeal ways, webserver companies are already moving this way with low powered CPUs and fiddling with "blade" machines. A smart marketing team with a smart engineering team could easily bring Sun back into the server market. Not the mom & pop ISPs, but the AOLs and Verisigns of the world. Their problem is that their hardware is not quite designed to hotswap CPUs and memory like hard drives, they haven't configured a software product to realize this vision, their OS is still relatively esoteric, and they margin themselves out of profitability. But none of those things are impossible to correct.