Is US becoming the first nation in the world to be dominated by options wielding technocrats? Or are the elite merely waiting for their life-extending biotech rah-rah while - 50% of the world are still impverished - a large majority don't even have a phone - computers are useless because they can't read - baby boomers waiting for the massive transfer of wealth and anticipate living off the tax-sweat of the next generation of the young - can't afford health care much less the exotic drugs the pharmaceuticals charge to recoup R&D (plus hefty margin) costs (nothing like a captive market) - still waiting for the US to pay off its $6 trillion dollar debt while addicting third world nations and various corrupt governments to a consumer lifestyle they can't afford
Yes, it's nice getting 6 figure salaries designing the next smart sweatshop sneaker and worrying about biobabble. I'd like to point out a little newspaper article that caught my eye when a reader ask her son (serviing in the East Timor Peace Keeping Forces) what the people over there would appreciate as a present from her and he replied that for the Timorese, Xmas is a sacred period for celebrating with kin and giving thanks for their delivery, not the credit-draining consumerism exercise it is here in the US. For what merit is technology without the moral sense to apply it wisely? Too often we see the glitter of a holy grail without realising the price. DDT, nuclear research, exploration spread exotic weeds, monocultures, derivative based capital flows, all had consequences beyond those intended by their creators. I just hope the Internet makes the human race as a whole a lot more prepared for the next technology wave than historical economic evidence. In particular read up on the past just to see how economic development has been derived from the struggles of various self-interests (of which OpenSource is yet another saga). Read up on books like Carl Sagan's "The Demand Haunted World" just to realise how much the average citizen is fascinated by superstition (its scary when more people believe in a live Elvis than a dead God). Will biotech be any different? Hopefully we will have developed a better sense of moral ethics by then to guide our decision making.
I certainly hope Red Hat have done their market research. Despite the rather gung-ho image projected by the State Department, China is still predominantly an agricultural based economy (average GDP/capita $3.6K, CIA factbook). Hence if they buy computers, it is likely to be cheap knock-down clones from the Taiwanese or white boxes assembled from components sourced inland. Getting the average person to spend a third of their (average) yearly income on a luxury good (to them) seems rather extreme. Microsoft have decided to go for an internet appliance and I beleive the home red team is also flogging their own TV hookup box. While the coastal regions like HK and Shanghai may be good prospects for sales, given that you can get pirated CDs/VideoDiscs/DVDs for near cost of media (if you even look foreign, they'd come up to you and openly offer to bargin titles) I'd wonder about the chances of any foreign firm trying to sell full-priced equipment. It is a really tough environment to do business, just read AsiaMoney archives on the pitfalls of western capitalism meeting eastern greed (not a pretty sight). I wound rank Japan, Singapore, Taiwan, India, Thailand, Korea higher up in the priority list as those countries have got good educational systems and would benefit more from computer automation and there are likely to be more multinationals there who could benefit from Linux support. It will be interesting to see what the WTO agreement will do to the ISP industry in that region though.
In this world, you get paid for a) time b) task c) talent
Time spent is aasy to measure, task-based performance requires some management competence (hah) in defining specifications and quality, and creative talent is more a wild-card in that you can't always predict the outcome.
Now in a risk adverse commercial environment (despite what capital vultures... errr... venturers say, their job is to load up with bear on the best bets), guess which renumeration scheme is favored? Human nature is perverse in that if you make a task look easy (think of an experienced TV repairman coming in, taking one look and replacing the exact part that failed), you tend to underappreciate it. You can see the "effortless" basket scoring or the clever hack but how many people realise the years or decades of training that led to it?
Looks like the gaming engine world is starting on the slippery slope to the degenerate stage of nit-picking over the merits of each gaming engine. However, I would like to point out that it would be a worthwhile endeavour for the gaming companies to define a compatible API/script language/whatever so that other people can develop smarter AIs/bots/whatever that are portable across multiple gaming generations. One particular advantage I can think of is that it allows more persistant player development in that you can build up a character (let's face it, people can become attached to a persona) and allow them to retain them across the different releases. Plus I can see possibilities opening up in terms of new wrinkles.
Speaking of new wrinkles, after observing the recent Quake client look-ahead crack, I've always wondered what tactics would work on a precog (effectively what a client with perfect future knowledge is).
Zico wrote It's not like SGIs stopped being good computers. If you already own the SGIs, it sounds like the more expensive option is buying computers to replace them.
I'd second that thought are suggest that people start looking at the total costs. IMHO hardware only tends to be 10-15% of the final IT bill. I wish there were some formal studies to quantify the issue (instead of the usual benchcraft or benchmarketing) but my rule of thumb is that for every $10 you spend - $1 goes to the base hardware - $2 for storage/connectivity/peripherals - $3 for software/operationas/maintenance - $4 for user training and ongoing help
Now, you may think that your high up-front capital costs can be reduced but then have you really looked at what it encapsulates? From working with the Octane, I know it comes bundled with a mass of multimedia software that you'd need to support separately. I confess a bias in that I really dig their modular physical layout. Anyone who've disassembled an O2 will know what I'm talking about. And a lot of work and R&D has gone into their software to make the I/O one of the best (admittedly you have to have knowledgeable people in tuning stuff). I would say that for certain types of jobs, SGI can produce some really balanced and well-engineered architectures.
Now it may well be that thinking about your setup would gain better total costs. If you have a high turn-over of people then you'd be pouring money down a training black-hole unless you retain some institutional memory (or outsource). Also as the number of machine footprints expand, operational scalability issues come into play. Heck a decent sysadmin/programmer would cost $100K/year compared with the $300K/Octane/user which can be amortised over 5-6 years. An alternative may be getting a network of O2s for each user and an Origin 2100 (cheaper than their Origin 2000s) and using their new OpenGL Visualiser to overlay video across the network but concentrate processing within the SMP. Is the work you're doing continuous or bursty. Is time more critical than cost? Who is in charge of data integrity and associated risks? How do you respond to outrages and disruptions to routine? Are then plans for responding to threats and disasters? These should be the policy questions you should be looking at rather than whether the mouse is Microsoft or not.
Anyway, I'm not the best person to ask about configuration as every site and their requirements is different. You guys have got various labs (LLNL, LBL, etc) who do nothing except benchmark machines and some of the smartest code cutters in the world. Spend some money on a consultancy exercise instead of buy the popular hardware-du-jour. Nothing can replace some critical thinking about what setup mix is best for your needs. I'd point out that substituting souped up tricyles is no replacement for industrial strength trucks. Make sure you understand what is needed where as a diversified transport system is designed to provide a high throughput system without bottlenecks. Above all, don't skimp on the training otherwise all you'd end up doing is buying overpriced junk which will be obsolete in 1 year's time.
It's an unfortunate habit that people only notice the flash floods and not the rising tides that affect everyone, particularly those born of many dedicated and nameless people. It is too bad that we can't pay tribute to those guiding the social transformations of this century
war against widespread diseases such as TB and polio that crippled large segments of the population
creation of orphanages and social support systems during the Industrial revolution
creation of non-political supra-national organisations such as Red Cross
recognition of individual freedoms and energy as a means out of poverty such as Deng Xioaping in giving the peasants of China a path out of the disasterous Great Leap Forward. If you think changing a global mindset is difficult, try convincing Wall Street that extreme capitalisation with a flexible labor force (and associated breakup of extended family bonds) is socially destructive.
the pill and associated feminist movement (e.g. vote after WW1) which gave 50% of the population the freedom and responsibility to plan their lives at their convenience
the recognition of wider classes of humanity (us vs them), breaking down artificial mental barriers such as caste (Ghandi), race (slavery, South Africa), and more challenged (mental, physical, etc)
acceptance of universal education, especially higher tertiary levels which prior to WW2 was only reserved for the elite and well off (ivy league aside)
the concept of punishment, restitution and rehabilitation to society (e.g. community service). Prior to decent computer record keeping it would be impossible. Note that this is still an ongoing process.
the consistent applciation of legal principles (e.g. Nullas Terra of Australia, Waitangi Treaty of NZ) that force governments to acknowledge past injustices even though it creates financial pain in the immediate term. The rule of law is probably one of the most fundamental advances this century, giving a flexible framework (though rancuous and noisy at times) for society to define persistent norms and non-acceptable behaviour. Now if we can only get politicians and special interests off the economic levers....
Technology comes and goes but social advances are forever.
Profit may seem to be incongruous with free (speech) software but in fact it is intrinsic to the system when you consider it as a proxy measure for the efficiency of delivering a final good/service. For example instead of hiring 100 water carriers to hustle bottled water, I borrow a billion dollars and build a pipe network maintained by 5 engineers and charge people a lower cost for bulk water, then the profit is a measure of freeing up 95 people to do more useful tasks. Thus profit is a complex formulae combining risk (entrepreneurship) + talent (skill at creating a more efficient business system addressing a market need) + management (motivating everyone and keeping out competitors). The point is not that you can cook a better hamburger than McDonalds (not very difficult) but that they can provide fast food at an affordable cost at a consistent level of standard (no comments about quality please) and franchise it globally.
Why should OpenSource be profitable? Because ESR pointed out very clearly that software is a service, not a manufactured good and the bulk of programming time is spent in the maintenance phase. One big problem with corporatism is they have this habit of externalising negativities, what people commonly call privatising the profits but leting risks get lumped on the public. One obvious example is the tobacco industry which has caused ripple efects on the public health system. Hence the domination of a company in addicting the rest of the world on an endless cycle of unnecessary upgrades has got the rest of the industry feeling the impact of the technology arms race and stressed out and aggravated at the code quality. OpenSource can be seen as a natural reaction by the intermediate customers (who are not the hardware manufacturers) but the developers who want a stable and cheap platform in creating new internet services for their end-customers.
Now most people don't object to profits fairly gained and in fact if they are very successful they eventually dominate their market niche and become a natural monopoly. However, bad business practices can easily erode goodwill such as
bundled packages with forced tying
depressing and deprecating competitor's products to force a buyout at lower than market price
deliberate obfuscation of standards/APIs create barriers to subsitution
consistent history of providing bogus delivery dates for non-finished products
defining and imposing in-house specifications rather than common interoperable standards
Fundamentally a company operates in a network of trust (simply to avoid getting expensive lawyers in to guard every silly thing) and creating a toxic wasteground of relationships as a result of dubious activities, no matter how currently profitable in the short term, is fatal to long-term success. Thus it is not so much the profits that annoy people but the means of achieving those profits and the fact that larger organisations carry their fair share of the social consequences of their actions (e.g. oil companies with their environmental policies, manufacturing with their sweatshop practices, software companies with their founder's personality cults.
If OpenSource is successful as a profitable business model then it must indicate that it satisfies the necessary conditions for a sustainable industry. The point is that with the code being freely available, then the only distinguishing factor is the value of their associated services, reputation for sales support and means for attracting skiled staff with a combination of lifestyle and long-term mission/purpose. One obvious analogy is why do doctors volunteer to work for organisations such as Medici Sans Frontiers? Because they believe in the cause. Profits are an interesting metric of value but they can't measure everything.
Fortran still has its strengths and with the new extensions (OpenMP, OO traits) can still be used. However, for teaching purposes, I would hazard that people would prefer coding in higher level languages such as MatLab. Sure Fortran may be that trusty comfortable hammer but when you've got an automated nailgun handy, it makes sense to shoot yourself in the foot faster:-). CPU cycles are getting cheaper whereas human development time is not. Debugging old style Fortran is still a pain even though it doesn't have the pointer aliasing problems of C. Also people at TriLabs have found that with the right approach, numerical libraries written in C++ can be just as good as Fortran. If you look at areas where some heavy number crunching is used, for example real-world physics in game engines, I suspect that C++ would predominate. Hence Fortran will still cruise on like the mainstay Cobol but the use of other languages (like Java and Python) will start to draw attention, especially to utilise commercial development environments.
The big bugbear is that Matlab compilers still don't optimise themselves very well for modern architectures with cache and/or hierarchical memory. If you're interested in the nitty gritty details of cache-line control and low-level latency then hand-tuned Fortran still has a place. As the Alpha processor has the highest raw floating point performane, it is a favorite of many academic and scientific grunt workstations. And if worst comes to worst, you can use the heat it generates as a coffee warmer:-).
The fundamental problem with patenting information is that you cannot exclude other people from having the same (or similar ideas) independently. Technology comes in waves and each wave of ideas builds on the previous infrastructure/memeware left by the previous. Creating a rigid legislative framework for a fluid medium is cause for trouble and only enriches only lawyers. There are a number of problems with the current system. The holy mantra of network effects (ie Metcalfe's Law that the utility of a network increases as the square of the number of participants) encourages companies to create Ponzi schemes to try and create the maximum customer base for their wares (can we say.con here?) on the hope that they can exclude any competitors (a la AOL instant messaging) and sell their customers to advertisers. Patents are just one weapon in this scheme as the speed of penetration translations to market share and any delaying tactics on your competition hurts their growth rates. There are actually some rather interesting theoretical work on modelling word-of-mouth effects as quantum waves but in practical terms, a patent system designed for manufacturing is just not suited for creating new services. Imposing a slow moving legislative process (and there are good reasons why people prefer time to think over complex issues) over a fast mutating technology market is like controlling hot air with a piece of string. I would argue that if Amazon really thought their idea was so hot, they should create the software and license it to other players and let them reap the benefits. If the idea is so simple that people can implement it without paying then it must not have been that novel or distinctive in the first place. Now what are worthy things that deserve protection? I would probably nominate file formats (ie anything stored in permanent form) as in return for making it non-proprietary, the community can award the firm some rights such as naming and/or leadership/respect role. Here one can contrast the reputation of Adobe (Postscript -> PDF) vs SGI (GL -> OpenGL) in their efforts to foster the wider market. IMHO Adobe had a chance to compete with Microsoft in defining desktop document standards (ie interface to printers, especially industrial ones) but blew it while SGI has gone whole hog and freed up their GLX and Performer APIs.
Unfortunately companies can measure #patents but not intangibles like goodwill but which is more important in the long-term?. Unfortunately (or maybe deliberately so) consumer education for complex technical stuff like computers has still a long way to go when people can still confuse MHZ (effectively the RPM of the CPU) with the total software/hardware package (ie motor + comfort + millage). Perhaps consumer education is one area where Linux can be more effective, especially in warning against the long-term hard that frivilous patents can cause.
... take a look at this article. This highlights the contrast between the traditional bread earners and the so called new IT revolution. One needs to question what is wealth creation or even the definition of profit. The capitalisation of a company is nothing more than the expected discounted income stream plus expected growth. Hence if a company is valued at say $1 billion, shareholders expect a company to earn at least that much in profits (or revenues in the case of.com fluff) over a period when it is deemed to have a competitive edge. For some strange reason the assumption is that share price is roughly analogous to dominance in market share (ie the number of customers you "own") which seems a little presumptious considering that with the reduction in transaction costs, there may be multiple paths/sources to obtaining the same good or service. Thus as soon as a new competitor arrives, the share price drops to indicate a dilution of the market. Why would an IT cartel have such a high "value" compared to a farm which produces so called essentials. Basically it relates to the laws of supply and demand and people mistaking a trajectory which they assume continues forever. Domain name registrations will slow down once every business and person has an account (or we run out of recognisable letters). Once the marginal costs reduce down to the actual operating costs of renewing a site, then you will see some serious cash-flow hiccups along with brutal competition to keep others from poaching your customers. and then maybe it might be smart switching your vaporcash into a real farm. IT companies only have growth potential so long as their products result in improved productivity and capabilities. Once the stress and pain becomes too much (as in the case of a not-so-popular OS) then people start resisting. The real competition to software is not other players but your historical software and customer habits. People tend to underestimate the risk premiums in share prices and until a crash or two happens to kill off the ignorant, rationality won't return to the share-market. Also, it would help if politicans kept their meddling hands out of the pot.
Companies are not in the charity business. They do not spend millions on R&D and then "give" it away for free. My guess in looking at where H&P sees its future profits is the ultimate conversion from analog disconnected instruments to a fully digital freeform communications net. If you go to a hospital, you'd see zillions of equipment, most working on old standards (can we say serial lines?) or proprietary interfaces and thus lack the transparency and interface standards for them to work seamlessly. If we make this assumption, then we can see a reverse chain of logic connecting the GPL and their market. Hence the push for e-speak as a technology platform to try and gain a competitive advantage in the next generational upgrade. Hence they need a pool of developers and experience base (system integrators) so they can flog their hardware. Hence the GPL of the initial toolset and APIs to try and draw attention away from the Wintel/Java platforms. The smart theory being if you put out the honeypot, the worker bees will come.
The computer industry is rapidly following the development of the early car industry with distinct feature sets (cars, trucks, etc as well as the associated fallout and consolidation). My general impression of the analogies
IBM - corporate market - big iron + Java connectivity Sun - mid-sized corporate market - medium iron + Java HP - medical + manufacturing industries - instruments + e-speak SGI - scientific market - big/medium iron + OpenSource Apple - education, prosumer market - cute simple boxes Palm/Nokkia - wireless market - handsets + WAP etc Wintel - anything and everything
What the big companies are doing is trying to build up the component manufactuers and affiliates and then sell the finished branded product. Sure, you could assemble your own hand-tuned custom car today, but there's a reason why people go off to rummage around the car-yard instead of mucking around with the parts. I expect something similar for computers when the hardware/software/wetware complexity reaches a point such that hackers have to invest in a postgraduate (a la medicine) course just to understand the silly things. For your interest, the Australian Computer Society has pushed to obtain a professional recognition status for IT. Which means that they can now "exclude" non-qualified people from practising. Expect specialist IT salaries to keep on rising as they become the new lawyers/doctrs/dentists to the information infrastructure (with matching fees).
They must have invented this treadmill just for the rat race.
to identify "original" impressions? I'm thinking of the case of digital cameras/camcorders where you need to use the results as evidence in say a court and you wish to verify that any resulting image is the true and faithful record rather than a digital marked up version. The abuse of technology to fake evidence, influence a constitutent or pervert the course of justice, either deliberately or by chance) can be a danger to a society which increasingly requires a informed evaluation of rather complex issues. Already artists think nothingabout touching up their works. Given the increasingly use of synthetic imagery, how much further will it go before we trust anything we see or hear? Perhaps this will follow the case of rubies where the artificial ones are so perfect, the real collectors items are those with natural flaws which are difficult to fake. But with digital media which is infinitely malleable, how can one tell the difference between reality and augmented? Think of the increasing use of artificial substitutes for money (book tokens, phone cards, gift vouchers, etc).... how easily can these be faked, especially in digital form? As a fiat money, the dollar bill represents nothing except a promise backed by the trust and faith of the people for a future claim on some resource, good or service. Laws and technology may be fine, but they are no substitute for personal trust and eyeballing the system to make sure there are no hidden gotchas.
is actually public opinion. Corporations find it very hard to sell things if they piss off (much less kill) their customers (though the tobacco companies has had a good run). Now everyone is entitled to social theorising but I think it is worthwhile injecting some cold hard reality into the talk. Large governments will always lord over corporations because the legal system is a codification of social conduct. I would argue that distortions are more likely to be created by the law of unintended consequences (e.g. the patent system) and that public corporations can only operate within the bounds of legal and thus social acceptability (though the law may lag the technology). Once someone screws up in a major way, then legislation will be passed so quick that you head will spin. Good example is the shooting in Australia a few years ago which galvanised the public (and thus the government) to ban automatic rifles, not to mention the expropriation of the guy's estate to reimburse the victims (a rather dubious precedent no matter how well-meaning). Similiarly with all the gee-whiz biotech advances that pundits are speculating on. It may well be that there is no market for gene-therapy once the risks are assessed. How many other markets have failed to eventuate despite the posturing of the big players (Farenhit, NetPC, etc)?
As for the money aspect, you have to spend money to get it back in the future. People forget that the costs are ultimately passed onto the consumer so all the big sums of money being thrown at private research will eventually end up in your medical bill for products/services at prices that the market will bear (otherwise it will just be droped as unprofitable). This you can thank the good old FDA for controlling the clinical tests and thus creating inelastic markets. On the other hand, the public system might be inefficient at research but at least it is more likely to address real social concerns rather than fee-paying cosmetic augmentation stuff.
People are always concerned about technology, you go back 10, 20, 50, 100 years and you'd find similar stories about cars, movie projectors, electricity, whats-not. In another 100 years it might be nanotech and mini-blackholes. The point about an open society is that any discoveries can be discussed and feedback applied to moderate excesses. Here one must be careful of all the things you read as opinions are not facts (e.g. the misinformation about Serb concentration camps). The voice of reason and some critical thinking will go a long way towards reducing the fears people have about biotech. At this state biocomputing is so early that it's like we haven't invented the equivalent of the transitor yet, much less understand the many different intracellular processes in great detail. However, if and when innovative applications appear, some simple moral questions to ask are would you be able to sell the concept to your mother? If the development background was splashed on the national front page, would people still be interested (one reason why artifical substitutes have been found to replace animal testing)? And most importantly, could you look at yourself in the mirror in the morning if you were the inventor? People are not stupid and trying to force your beliefs or dogma onto others is a sure way to create enemies and annoy friends.
Wasn't this topic discussed, oh, about a year ago and the/. community thrashed the concept about? Perhaps it's symptomatic of episoidal topics that get recycled every season or so (or a reflection of the continual growth of the/. readership) but perhaps it would be wise to include in the related links a link to the last closely relevant discussion (threshold=4) so that discussion can continue off the higher base rather than going through a process of reeducating as to the background. Also why not have a link to any forward (ie future) discussion as well? For people with limited time as any long-timer here is likely to be flatout employed on mission critical tasks (or burnt out) so short-circuiting the normal flamers and lamers and thus would get to the meat of the discussion more quickly. It would also give a better historical/future record of the crosslinked discussion thread allowing less informed souls to realise and hopefully appreciate the momentum of the topic. We get so caught up with day-to-day emergencies that often we don't appreciate that solutions have been thought out previously or smarter minds hae already considered the problem./. is more than a newspaper or e-journal and it behooves the editors (and the pocketbooks of the onwers) to make/. a unique and distinctive experience.
Brain drain is a rather emotional laden concept. However, it might be pointed out that there is no movement of people unless there is an empty position for them to shift into. The question should then be asked as to how come the US can't generate enough American talent to fufill these jobs? Given that professors at US unis get paid significant amounts of money, at least more than the average non-US uni, I would hope they give their students the skills to fufill the needs of the economy. If this is not the case and all the foreigners are taking the high paying jobs, where does that leave the average Joe? The fallacy in this type of thinking is limiting your thinking to just the US economy and assuming that to raise US standard of living, you have to depress the rest of the world. Economics is not a zero sum game.
However, it would also help if the Fed could get their money supply and debtunder control so that salaries and purchasing power parity figures for global comparison are not distorted. Even a starting Silicon Valley income can be net negative once you factor in teh cost of living.
I believe it was Peter Drucker that said the role of a business is to create a need. You might consider the example of the early automative wars when safety belts was non-existant. It was only due to one persistent soul who dragged the car manufactuers kicking and screaming over the costs and hassles of installation. From their point of view, it would have been seem as scare-mongering and market fracturing. People forget that the market is a dynamic process, in essense you are trying to shape and alter the desires of consumers. One can only look at Amazon, their job is flogging books/CDs/widgets but their business is altering (and ultimately controlling) consumer habits. In this environment, especially intellectual property which is inherently limited by the wetware (ability of our brains to acceptand understand the product), FUD is a key weapon. Don't buy x because it is not the "standard". Use y because it will make your company more competitive. If you don't invest in z, you will disappear where x,y,z can be substituted for e-commerce, i-retailing, or the buzzword du jour. In other words, FUD can be a way of motivating your market to take certain action (perferably in transferring money from their pockets to yours).
The software market is slowly being transformed and the real fight is over where the balance of power (and thus control over spending) goes. Between the IP manufacturers and bundlers or the consultants and service people. FUD is a powerful weapon, as it can be used to deliberately suppress a competitor's stock price and thus create an easy take-over target using overpriced script as technology currency. Is it good or bad? Depends on whether you are the predator or the prey. Certainly Wall Street brokerage firms are not complaining. Fear and greed are two powerful emotions that can motivate individuals and FUD feeds on these elements. While from an engineering perspective, the technical advantages may be obvious, unfortunately life is not like that and good-enough tends to rule the day. While Linux anti-FUD may be a counteracting force, at the end of the day it is just the same thing in a different guise in trying to convince the market that your vision is a superior model despite the billions of capital lined up with proprietary solutions. As for the truth, whether benchmarking or quality testing, does anybody really care anymore? And this is the sad part.
It is still an open question what role the junk DNA, technically called introns plays in organism development. Unlike the simple unicelluar critters (prokaryotes) such as bateria, all higher level organisms (eukaryotes) have these long non-coding sequences which have been retained across evolutionary generations despite the extra energy/space required. The whole area is akin to the physists search for all the various subatomic particles after the cracking of the atom. We can see the bits and pieces, we can assemble the various sequences, but there's no unifying standard model of how or why. With Nobel prizes and new killer apps in the air, it is not surprising that universities and institutes are throwing money into the research.
The 19th centure might have been the dominance of physics and engineering but there's a lot of speculation and anticipation (especially by the empty hands of the biologists and zoologists) that the next century will be their turn at the gravy train:-). Fun times ahead.
People have pointed out the technical disadvantages of WAP, particularly as a specification controlled by a "cartel". Well, technical standards are nice, but not necessary for commercial success. I suppose it is a bit naive to expect these manufacturers to give up the control of standards (see, Microsoft has shown how a pure software business model can really work). Businesses are in the process of making money, and if artificial scarcity drives up the entry barriers, so much the better (ever wonder why airlines offer frequent flyer points instead of just lowering the ticket-price?). Whether it is high switching costs, bundled tools, or the ability to impose rules that enforce the status quo, Linux hackers would be foolish to expect the business world to let go of a potentially lucrative cash cow. Without the ability to exclude, it would be impossible to create a monetary differential or perceived value. Heck, if they could meter the air you breathe (which they can already do for gas, bandwidth, electricity, and water) they would. Conditioning the consumer is a prime goal as it is easier to convince people to continue paying on a time-based charge a la mobile phone charges rather than the freebies people expect from the internet (even though the underlying technology is identical). If you haven't got a way of lining up the sheep... err... consumers to be fleeced, then you might as well not be in business.
As for Linux, it still remains to be seen whether giving away your technology and time can create sustainable business models.
Despite the best attempts of media to convince the unwashed masses otherwise, there is a distinct difference between celebrity and fame. Part of it is to do with your time horizon and any distinctive or unique claim to a place in history. While most current pop stars or sports heroes will be forgotten in another decade or five, certain names (at least in the hacker sphere) will still be remembered (RS, Linus, ERS, etc) for changing the way we think and view the world. In a sense, fame is somewhat (initially) conversely opposite to popularity (ie raise a ruckus), where people (whether conciously or not) take a principled stand in their view and stick to their convictions. Unlike the standard 30 soundbites, your actions, words and deeds are recorded for prosperity in/. and other sources. Hence how you act, think and reason can be judged by your peers long after the event without much chance of professional spin-doctoring. The truth might hurt, but hopefully it forges the character a little bit stronger and in a positive way. So although there is a similar effect to the media spotlight, I would trust that people take it in the same spirit as peer-review and supply positive criticism and feedback rather than throwing brickbats (unless it's well deserved stupidity). The only real answer to idiots is to ignore them.
Too bad none of the politicians have started a truth in media platform:-(.
Bruce Perens wrote Besides, I don't have Ulitmate Karma on Slashdot. Sig11's is higher than mine (I just happened to look at his user record a while back), probably other people as well.
Well, IMHO the absolute karma becomes rather meaningless after a while. I think what is more important is the relative karma/post ratio, ie some sort of median of how often your score goes up relative to the postings you make per week. I suspect people goes through a number of stages, initially posting lots to get the +1 bonus. Then more strategically (e.g. very early on in the thread to get noticed early) to race against the competition in running up your karma. Then I suppose the boredom stage sets in or your time runs out when more important things intrude in your life like an SO.
The biggest problem is that because you can create unlimited karma, the whole process gets out of control (can we say unbounded feedback loop?) and the whole system becomes inflated beyond recognition. To paraphrase Peter Drucker's saying that governments only do two things well, wage war and inflate the currency, I'd toss in the contention that/.ers only do two things well, initiate flamewars and inflate their karma:-). Don't know what the solution is, perhaps introduce a new rule that once your karma reaches 100 (tribal elder status?) you can moderate at any time and simultaneously post iff you're willing to sacrifice your existing karma. The alternative is to make sure that karma moderations on average sum to zero (or 1) (more difficult) so that people have an equal chance of going up/down and that the total karma doesn't tend to infinity and thus become overinflated (and thus devalued).
One can point to the trend in furniture/kitchen in that the components (sinks/cupboards) are becoming commodities so furniture stores are turning into lifestyle/design/fitouts where they assemble a package tailored towards the customer's desires. Perhaps car dealers will shift towards a similar philosophy in ordering on-line, but offering mag-wheels, gee-whiz mobile-phone/stereo/DVD, beer cooler, leather seats, etc.... Heck it'd be just like designing a portable room:-). Everyone is going to know the price so the only way to add value is to offer services (insurance, repairs etc) that make a one-stop shop with physical visits worthwhile. If you're just moving wheels, then you will disappear.
As for retail, the basic stuff will be commoditised and automated but there will still be need for people to suggest alternatives and provide an admiring audience. Perhaps the days of the megamalls will disappear?
I'm surprised people here are so ignorant about the realities of the marketplace.
Facts
1) If a company doesn't have a competitive advantage, then sooner or later it will lose marketshare. The NT based Intel line (idea of the previous CEO) was targetting a niche which has been eroded by Dell's just-in-time assembly and creating a gap until HDTV standards become mandatory. SGI is being smart cutting their losses when they can. Unlike the disposable mentality that Wintel is trying to foster, SGI machines are designed to last for a long time (10 years+) which if your amortise the higher upfront costs, is actually cheaper in the long run.
2) SGI still has 300 odd engineers working on Irix and a growing pool of Linux developers to progressively shift their core competencies (multiprocessor design, low-latency memory subsystems) into value-added components. 10% of the market is seriously interested in high performance vs good enough.
3) Processors take a long time to die. Their MIPS line still has got excellent cost per application performance and decent data bandwidth (running a gigahertz Intel chip with IDE drives to surf the internet sounds a little fishy). As an embedded processor, the x86 is dwarfed by sales of MIPS cores cross-licensed to multiple vendors and their high-end stuff (R16000, etc) goes out to another 7 years.
4) The world market has been very very lucky in the Asian crisis has depressed global supply to such an extent that firms are willing to sell subsystems at near cost just to keep the cashflow up. In my opinion this has distorted the market and once things start bouncing back (as with the recent hike in memory prices) then their offerings will be more competitive.
5) Intel is becoming more like a venture capitalist than chip maker by throwing advanced designs at anything that looks, walks and talks like a high-growth startup. SGI is sorta doing the same thing by developing high-end, then flowing the tech to NDIVA (graphics), Cray (memory subsystem) and MIPS (CPU design).
6) Forget hardware, the value is now concentrated on the services. Even IBM are finding it difficult to move their big iron. Ultimately you will see companies becoming coordinators assembling components and software bundles to targetting specific market categories and capturing benefits from after-sales support and in the life-time money stream. You'd might be surprised how little car manufacturers make from sell the initial wheels compared with gouring consumers with repairs and add on knic-knacs.
7) Linux is shifting the competitive landscape in that every processors has an equal chance as the source is available to all players. Intel is recognising this and is dumping money to maintain dominance. If IBM can offer a PPC reference design, then perhaps SGI could do something similar for their MIPS, OpenGL and other graphics libraries (ie OpenAPI but retain implementation IP). Don't rule out the Koreas with their Alpha licensing and the Japanese with MIPS variants (e.g. Sony Playstation2).
8) Lighten up, markets go up and down in the short term. If anyone recalls the dumping Apple was going through several years ago, SGI is going through a similar patch. With a new CEO on board and some good Linux buzz, they have opportunities to catch the next wave. From a professional point of view, it would be interesting to see if their boxes could be adapted to keep the same fast memory subsystem but accept MIPS, Cray, Intel or Alpha processor node cards and just absolutely dominate the SMP and Beowulf market.
9) High end graphics is tough with very few firms having the capability to tackle complex real-time graphics and simulation. Today's multiplayer games are a shadow of hosting real-time thousands of players. If SGI could shift some of their talent to exploit the gaming niche, then they could gain more revenue streams from selling their IP. Admittedly they do have problems with delayed supply of third party boards (sigh... the XIO bus still beats the wet dreams of FutureIO and is here now) but then they do certify and guarentee the performance of the ones that do make the effort. If SGI was smart, they'd license the XIO technology to foster a vibrant alternative.
In short, SGI may be down at the moment but certainly not out. You don't have to be big so long as you can be profitable in your own little market niche, afterall BMW and Rolls-Royce still survive despite the Fords, Hondas, etc of the world. If people stop fixating on pure clock speed and look at real-world performance (e.g. broadband systems) they might be pleasantly surprised.
... when you don't have the industrial-financial base to support it. It takes a heck of capital, both physical (plants, machinery, materials) and human (know-how, skills, training) to turn the ideas (stolen or otherwise) into real actual working machines. The US had to go nearly $6 trillion into debt to fund the Cold War (accumulation of deficits from Korean, Vietnam and various anti-communist insurgencies) and even then it was a rather close thing that they didn't go bankrupt before Russia did. And selling off Mickey Mouse hats and movies to the rest of the world is not going to pay this off either.
If you look at China (how many people here have actually visited that place?) you'd still find mud houses (strangely enough sitting next to a modern satellite equiped apartment block) in central Beijing. Their energy needs are still powered by high suphur content coal (one reason why interior cities are so polluted). And too many of their really good technical people leave and/or are snatched up by Western firms. Despite the war-mongering vote chasing by Bush, China is still pretty much an agarian society with much agriculture still being doing by hand. People forget that the US war machine is funded by taxes (40% of GDP?) from realatively high incomes (by world standards) that goes to pay for all those nicy shiny missiles (which have to be replaced after shooting up the reminants of Serbian infrastructure) and associated pork-barrelling. In fact some people would claim that the US economy is still on a war footing.
It is only now that resources are being released back into the civilian sector that you see such advances in communications (CDMA was originally a military application) and software. China's level of technical skill is comparatively poor and while they may be a world power by 2050, at the moment it is not exactly in a state to wave any sticks. IMHO, the US has more worries about trying to keep moral high and designing attractive careers for its own military personnel than worrying about any other army on the earth which they could beat hansomely (given enough warning) except the Isrealiis who have a (justifiably) paranoid and professional army. In other words, if you're Chinese and have a hankering for violence, the best idea is to give up now, emigrate to the US and learn how to play Quake.:-)
mjacksol wrote Yes, but it is precisely this licensing model that is the problem, and it is Microsoft that had this stroke of evil genius. If I buy product X, which is essentially identical to product X+, but crippled to sell it cheaper, I feel ripped off.
Welcome to the world of market segmentation. Do you feel ripped off when you get that special discount air ticket that allows you to book holidays 4 weeks in advance? But surely (shock, horror) you get to your destination just as quickly as a first class passenger? The point is that the airlines deliberately add extra constraints and conditions to price match the demand of people and their willingness to pay. In effect, you pay extra for greater degrees of freedom (but don't tell the other passengers this:-) ).
Now MS applies this principle to software and you think it is outrageous? I believe IBM had something similar with mainframe machines which were physically the same internally but had their clock speed governed by a hidden switch? If you didn't know about this switch, would you complain about paying less for a slower machine? Ignorance is bliss sometimes.
Differential pricing is not new and sooner or later, somebody will come up with an equivalent for the internet (e.g. jump to the head of the queue if willing to pay a premium). It would be very very interesting to see the pricing calculations and models MS (or any other software company) uses to determine market demand. Any ex-marketing guys out there?
Oversoul wrote So, back to what I was saying, if people in the tech industry are principled, ethical, clear-headed folks, item #4 (price out the talent) falls. The monster strangle-hold companies (like microsoft) have their knees cut out without the gray matter.
Your sentiments are very noble and I hope you get your chance to demonstrate your commitment. At this stage, I am reminded of the old saying, "If you're not a socialist by 20, you've got no heart. If you're not a capitalist by 40, you've got no brain".
Business is in the business of accumulating capital (ie wealth) and maximising its growth, usually by exchanging a slice of it (like from pension funds) for your time/labor/ideas. Good, bad or indifferent, the system has been structured to act that way through enforced laws and ingrained habits. The bottom line is that if you don't make enough to cover your own salary (or at least what the general market thinks you're worth working for another job), much less make a profit, then you (or your company) will get bought out and absorbed by a more efficient organisation. The only way OpenSource path can prove superior is only if it is more efficient at achieving business solutions or satisfying consumer desires. The evidence is still out on this but all is not lost as some studies show that cooperative behaviour creates long-term benefits (but as they say, in the long term we're all dead and giving away the fruits of your labor is not a obvious solution to paying bills the next day).
By the way, be prepared for the fact that some (or many) companies demand that you sign over every piece of intellectual value you create (even off-premises). Protecting yourself through foreknowledge is the only way to avoid becoming an economic slave to the system. Correct me if I'm wrong but some people complain that with the current starting salaries and cost of living in Silicon Valley, they are actually earning negative net income.
There's a reason why they call it the business jungle out there. Best of luck in your studies and in finding a worthwhile job when you get out.
Slashdot Mirror
Is US becoming the first nation in the world to be dominated by options wielding technocrats? Or are the elite merely waiting for their life-extending biotech rah-rah while
- 50% of the world are still impverished
- a large majority don't even have a phone
- computers are useless because they can't read
- baby boomers waiting for the massive transfer of wealth and anticipate living off the tax-sweat of the next generation of the young
- can't afford health care much less the exotic drugs the pharmaceuticals charge to recoup R&D (plus hefty margin) costs (nothing like a captive market)
- still waiting for the US to pay off its $6 trillion dollar debt while addicting third world nations and various corrupt governments to a consumer lifestyle they can't afford
Yes, it's nice getting 6 figure salaries designing the next smart sweatshop sneaker and worrying about biobabble. I'd like to point out a little newspaper article that caught my eye when a reader ask her son (serviing in the East Timor Peace Keeping Forces) what the people over there would appreciate as a present from her and he replied that for the Timorese, Xmas is a sacred period for celebrating with kin and giving thanks for their delivery, not the credit-draining consumerism exercise it is here in the US. For what merit is technology without the moral sense to apply it wisely? Too often we see the glitter of a holy grail without realising the price. DDT, nuclear research, exploration spread exotic weeds, monocultures, derivative based capital flows, all had consequences beyond those intended by their creators. I just hope the Internet makes the human race as a whole a lot more prepared for the next technology wave than historical economic evidence. In particular read up on the past just to see how economic development has been derived from the struggles of various self-interests (of which OpenSource is yet another saga). Read up on books like Carl Sagan's "The Demand Haunted World" just to realise how much the average citizen is fascinated by superstition (its scary when more people believe in a live Elvis than a dead God). Will biotech be any different? Hopefully we will have developed a better sense of moral ethics by then to guide our decision making.
LL
I certainly hope Red Hat have done their market research. Despite the rather gung-ho image projected by the State Department, China is still predominantly an agricultural based economy (average GDP/capita $3.6K, CIA factbook). Hence if they buy computers, it is likely to be cheap knock-down clones from the Taiwanese or white boxes assembled from components sourced inland. Getting the average person to spend a third of their (average) yearly income on a luxury good (to them) seems rather extreme. Microsoft have decided to go for an internet appliance and I beleive the home red team is also flogging their own TV hookup box. While the coastal regions like HK and Shanghai may be good prospects for sales, given that you can get pirated CDs/VideoDiscs/DVDs for near cost of media (if you even look foreign, they'd come up to you and openly offer to bargin titles) I'd wonder about the chances of any foreign firm trying to sell full-priced equipment. It is a really tough environment to do business, just read AsiaMoney archives on the pitfalls of western capitalism meeting eastern greed (not a pretty sight). I wound rank Japan, Singapore, Taiwan, India, Thailand, Korea higher up in the priority list as those countries have got good educational systems and would benefit more from computer automation and there are likely to be more multinationals there who could benefit from Linux support. It will be interesting to see what the WTO agreement will do to the ISP industry in that region though.
LL
In this world, you get paid for
... errr ... venturers say, their job is to load up with bear on the best bets), guess which renumeration scheme is favored? Human nature is perverse in that if you make a task look easy (think of an experienced TV repairman coming in, taking one look and replacing the exact part that failed), you tend to underappreciate it. You can see the "effortless" basket scoring or the clever hack but how many people realise the years or decades of training that led to it?
a) time
b) task
c) talent
Time spent is aasy to measure, task-based performance requires some management competence (hah) in defining specifications and quality, and creative talent is more a wild-card in that you can't always predict the outcome.
Now in a risk adverse commercial environment (despite what capital vultures
LL
Looks like the gaming engine world is starting on the slippery slope to the degenerate stage of nit-picking over the merits of each gaming engine. However, I would like to point out that it would be a worthwhile endeavour for the gaming companies to define a compatible API/script language/whatever so that other people can develop smarter AIs/bots/whatever that are portable across multiple gaming generations. One particular advantage I can think of is that it allows more persistant player development in that you can build up a character (let's face it, people can become attached to a persona) and allow them to retain them across the different releases. Plus I can see possibilities opening up in terms of new wrinkles.
Speaking of new wrinkles, after observing the recent Quake client look-ahead crack, I've always wondered what tactics would work on a precog (effectively what a client with perfect future knowledge is).
LL
Zico wrote
It's not like SGIs stopped being good computers. If you already own the SGIs, it sounds like the more expensive option is buying computers to replace them.
I'd second that thought are suggest that people start looking at the total costs. IMHO hardware only tends to be 10-15% of the final IT bill. I wish there were some formal studies to quantify the issue (instead of the usual benchcraft or benchmarketing) but my rule of thumb is that for every $10 you spend
- $1 goes to the base hardware
- $2 for storage/connectivity/peripherals
- $3 for software/operationas/maintenance
- $4 for user training and ongoing help
Now, you may think that your high up-front capital costs can be reduced but then have you really looked at what it encapsulates? From working with the Octane, I know it comes bundled with a mass of multimedia software that you'd need to support separately. I confess a bias in that I really dig their modular physical layout. Anyone who've disassembled an O2 will know what I'm talking about. And a lot of work and R&D has gone into their software to make the I/O one of the best (admittedly you have to have knowledgeable people in tuning stuff). I would say that for certain types of jobs, SGI can produce some really balanced and well-engineered architectures.
Now it may well be that thinking about your setup would gain better total costs. If you have a high turn-over of people then you'd be pouring money down a training black-hole unless you retain some institutional memory (or outsource). Also as the number of machine footprints expand, operational scalability issues come into play. Heck a decent sysadmin/programmer would cost $100K/year compared with the $300K/Octane/user which can be amortised over 5-6 years. An alternative may be getting a network of O2s for each user and an Origin 2100 (cheaper than their Origin 2000s) and using their new OpenGL Visualiser to overlay video across the network but concentrate processing within the SMP. Is the work you're doing continuous or bursty. Is time more critical than cost? Who is in charge of data integrity and associated risks? How do you respond to outrages and disruptions to routine? Are then plans for responding to threats and disasters? These should be the policy questions you should be looking at rather than whether the mouse is Microsoft or not.
Anyway, I'm not the best person to ask about configuration as every site and their requirements is different. You guys have got various labs (LLNL, LBL, etc) who do nothing except benchmark machines and some of the smartest code cutters in the world. Spend some money on a consultancy exercise instead of buy the popular hardware-du-jour. Nothing can replace some critical thinking about what setup mix is best for your needs. I'd point out that substituting souped up tricyles is no replacement for industrial strength trucks. Make sure you understand what is needed where as a diversified transport system is designed to provide a high throughput system without bottlenecks. Above all, don't skimp on the training otherwise all you'd end up doing is buying overpriced junk which will be obsolete in 1 year's time.
LL
Technology comes and goes but social advances are forever.
Why should OpenSource be profitable? Because ESR pointed out very clearly that software is a service, not a manufactured good and the bulk of programming time is spent in the maintenance phase. One big problem with corporatism is they have this habit of externalising negativities, what people commonly call privatising the profits but leting risks get lumped on the public. One obvious example is the tobacco industry which has caused ripple efects on the public health system. Hence the domination of a company in addicting the rest of the world on an endless cycle of unnecessary upgrades has got the rest of the industry feeling the impact of the technology arms race and stressed out and aggravated at the code quality. OpenSource can be seen as a natural reaction by the intermediate customers (who are not the hardware manufacturers) but the developers who want a stable and cheap platform in creating new internet services for their end-customers.
Now most people don't object to profits fairly gained and in fact if they are very successful they eventually dominate their market niche and become a natural monopoly. However, bad business practices can easily erode goodwill such as
Fundamentally a company operates in a network of trust (simply to avoid getting expensive lawyers in to guard every silly thing) and creating a toxic wasteground of relationships as a result of dubious activities, no matter how currently profitable in the short term, is fatal to long-term success. Thus it is not so much the profits that annoy people but the means of achieving those profits and the fact that larger organisations carry their fair share of the social consequences of their actions (e.g. oil companies with their environmental policies, manufacturing with their sweatshop practices, software companies with their founder's personality cults.
If OpenSource is successful as a profitable business model then it must indicate that it satisfies the necessary conditions for a sustainable industry. The point is that with the code being freely available, then the only distinguishing factor is the value of their associated services, reputation for sales support and means for attracting skiled staff with a combination of lifestyle and long-term mission/purpose. One obvious analogy is why do doctors volunteer to work for organisations such as Medici Sans Frontiers? Because they believe in the cause. Profits are an interesting metric of value but they can't measure everything.
LL
Fortran still has its strengths and with the new extensions (OpenMP, OO traits) can still be used. However, for teaching purposes, I would hazard that people would prefer coding in higher level languages such as MatLab. Sure Fortran may be that trusty comfortable hammer but when you've got an automated nailgun handy, it makes sense to shoot yourself in the foot faster :-). CPU cycles are getting cheaper whereas human development time is not. Debugging old style Fortran is still a pain even though it doesn't have the pointer aliasing problems of C. Also people at TriLabs have found that with the right approach, numerical libraries written in C++ can be just as good as Fortran. If you look at areas where some heavy number crunching is used, for example real-world physics in game engines, I suspect that C++ would predominate. Hence Fortran will still cruise on like the mainstay Cobol but the use of other languages (like Java and Python) will start to draw attention, especially to utilise commercial development environments.
:-).
The big bugbear is that Matlab compilers still don't optimise themselves very well for modern architectures with cache and/or hierarchical memory. If you're interested in the nitty gritty details of cache-line control and low-level latency then hand-tuned Fortran still has a place. As the Alpha processor has the highest raw floating point performane, it is a favorite of many academic and scientific grunt workstations. And if worst comes to worst, you can use the heat it generates as a coffee warmer
LL
The fundamental problem with patenting information is that you cannot exclude other people from having the same (or similar ideas) independently. Technology comes in waves and each wave of ideas builds on the previous infrastructure/memeware left by the previous. Creating a rigid legislative framework for a fluid medium is cause for trouble and only enriches only lawyers. There are a number of problems with the current system. The holy mantra of network effects (ie Metcalfe's Law that the utility of a network increases as the square of the number of participants) encourages companies to create Ponzi schemes to try and create the maximum customer base for their wares (can we say .con here?) on the hope that they can exclude any competitors (a la AOL instant messaging) and sell their customers to advertisers. Patents are just one weapon in this scheme as the speed of penetration translations to market share and any delaying tactics on your competition hurts their growth rates. There are actually some rather interesting theoretical work on modelling word-of-mouth effects as quantum waves but in practical terms, a patent system designed for manufacturing is just not suited for creating new services. Imposing a slow moving legislative process (and there are good reasons why people prefer time to think over complex issues) over a fast mutating technology market is like controlling hot air with a piece of string. I would argue that if Amazon really thought their idea was so hot, they should create the software and license it to other players and let them reap the benefits. If the idea is so simple that people can implement it without paying then it must not have been that novel or distinctive in the first place. Now what are worthy things that deserve protection? I would probably nominate file formats (ie anything stored in permanent form) as in return for making it non-proprietary, the community can award the firm some rights such as naming and/or leadership/respect role. Here one can contrast the reputation of Adobe (Postscript -> PDF) vs SGI (GL -> OpenGL) in their efforts to foster the wider market. IMHO Adobe had a chance to compete with Microsoft in defining desktop document standards (ie interface to printers, especially industrial ones) but blew it while SGI has gone whole hog and freed up their GLX and Performer APIs.
Unfortunately companies can measure #patents but not intangibles like goodwill but which is more important in the long-term?. Unfortunately (or maybe deliberately so) consumer education for complex technical stuff like computers has still a long way to go when people can still confuse MHZ (effectively the RPM of the CPU) with the total software/hardware package (ie motor + comfort + millage). Perhaps consumer education is one area where Linux can be more effective, especially in warning against the long-term hard that frivilous patents can cause.
LL
... take a look at this article. This highlights the contrast between the traditional bread earners and the so called new IT revolution. One needs to question what is wealth creation or even the definition of profit. The capitalisation of a company is nothing more than the expected discounted income stream plus expected growth. Hence if a company is valued at say $1 billion, shareholders expect a company to earn at least that much in profits (or revenues in the case of .com fluff) over a period when it is deemed to have a competitive edge. For some strange reason the assumption is that share price is roughly analogous to dominance in market share (ie the number of customers you "own") which seems a little presumptious considering that with the reduction in transaction costs, there may be multiple paths/sources to obtaining the same good or service. Thus as soon as a new competitor arrives, the share price drops to indicate a dilution of the market. Why would an IT cartel have such a high "value" compared to a farm which produces so called essentials. Basically it relates to the laws of supply and demand and people mistaking a trajectory which they assume continues forever. Domain name registrations will slow down once every business and person has an account (or we run out of recognisable letters). Once the marginal costs reduce down to the actual operating costs of renewing a site, then you will see some serious cash-flow hiccups along with brutal competition to keep others from poaching your customers. and then maybe it might be smart switching your vaporcash into a real farm. IT companies only have growth potential so long as their products result in improved productivity and capabilities. Once the stress and pain becomes too much (as in the case of a not-so-popular OS) then people start resisting. The real competition to software is not other players but your historical software and customer habits. People tend to underestimate the risk premiums in share prices and until a crash or two happens to kill off the ignorant, rationality won't return to the share-market. Also, it would help if politicans kept their meddling hands out of the pot.
LL
Companies are not in the charity business. They do not spend millions on R&D and then "give" it away for free. My guess in looking at where H&P sees its future profits is the ultimate conversion from analog disconnected instruments to a fully digital freeform communications net. If you go to a hospital, you'd see zillions of equipment, most working on old standards (can we say serial lines?) or proprietary interfaces and thus lack the transparency and interface standards for them to work seamlessly. If we make this assumption, then we can see a reverse chain of logic connecting the GPL and their market. Hence the push for e-speak as a technology platform to try and gain a competitive advantage in the next generational upgrade. Hence they need a pool of developers and experience base (system integrators) so they can flog their hardware. Hence the GPL of the initial toolset and APIs to try and draw attention away from the Wintel/Java platforms. The smart theory being if you put out the honeypot, the worker bees will come.
The computer industry is rapidly following the development of the early car industry with distinct feature sets (cars, trucks, etc as well as the associated fallout and consolidation). My general impression of the analogies
IBM - corporate market - big iron + Java connectivity
Sun - mid-sized corporate market - medium iron + Java
HP - medical + manufacturing industries - instruments + e-speak
SGI - scientific market - big/medium iron + OpenSource
Apple - education, prosumer market - cute simple boxes
Palm/Nokkia - wireless market - handsets + WAP etc
Wintel - anything and everything
What the big companies are doing is trying to build up the component manufactuers and affiliates and then sell the finished branded product. Sure, you could assemble your own hand-tuned custom car today, but there's a reason why people go off to rummage around the car-yard instead of mucking around with the parts. I expect something similar for computers when the hardware/software/wetware complexity reaches a point such that hackers have to invest in a postgraduate (a la medicine) course just to understand the silly things. For your interest, the Australian Computer Society has pushed to obtain a professional recognition status for IT. Which means that they can now "exclude" non-qualified people from practising. Expect specialist IT salaries to keep on rising as they become the new lawyers/doctrs/dentists to the information infrastructure (with matching fees).
They must have invented this treadmill just for the rat race.
LL
to identify "original" impressions? I'm thinking of the case of digital cameras/camcorders where you need to use the results as evidence in say a court and you wish to verify that any resulting image is the true and faithful record rather than a digital marked up version. The abuse of technology to fake evidence, influence a constitutent or pervert the course of justice, either deliberately or by chance) can be a danger to a society which increasingly requires a informed evaluation of rather complex issues. Already artists think nothingabout touching up their works. Given the increasingly use of synthetic imagery, how much further will it go before we trust anything we see or hear? Perhaps this will follow the case of rubies where the artificial ones are so perfect, the real collectors items are those with natural flaws which are difficult to fake. But with digital media which is infinitely malleable, how can one tell the difference between reality and augmented? Think of the increasing use of artificial substitutes for money (book tokens, phone cards, gift vouchers, etc) .... how easily can these be faked, especially in digital form? As a fiat money, the dollar bill represents nothing except a promise backed by the trust and faith of the people for a future claim on some resource, good or service. Laws and technology may be fine, but they are no substitute for personal trust and eyeballing the system to make sure there are no hidden gotchas.
LL
is actually public opinion. Corporations find it very hard to sell things if they piss off (much less kill) their customers (though the tobacco companies has had a good run). Now everyone is entitled to social theorising but I think it is worthwhile injecting some cold hard reality into the talk. Large governments will always lord over corporations because the legal system is a codification of social conduct. I would argue that distortions are more likely to be created by the law of unintended consequences (e.g. the patent system) and that public corporations can only operate within the bounds of legal and thus social acceptability (though the law may lag the technology). Once someone screws up in a major way, then legislation will be passed so quick that you head will spin. Good example is the shooting in Australia a few years ago which galvanised the public (and thus the government) to ban automatic rifles, not to mention the expropriation of the guy's estate to reimburse the victims (a rather dubious precedent no matter how well-meaning). Similiarly with all the gee-whiz biotech advances that pundits are speculating on. It may well be that there is no market for gene-therapy once the risks are assessed. How many other markets have failed to eventuate despite the posturing of the big players (Farenhit, NetPC, etc)?
As for the money aspect, you have to spend money to get it back in the future. People forget that the costs are ultimately passed onto the consumer so all the big sums of money being thrown at private research will eventually end up in your medical bill for products/services at prices that the market will bear (otherwise it will just be droped as unprofitable). This you can thank the good old FDA for controlling the clinical tests and thus creating inelastic markets. On the other hand, the public system might be inefficient at research but at least it is more likely to address real social concerns rather than fee-paying cosmetic augmentation stuff.
People are always concerned about technology, you go back 10, 20, 50, 100 years and you'd find similar stories about cars, movie projectors, electricity, whats-not. In another 100 years it might be nanotech and mini-blackholes. The point about an open society is that any discoveries can be discussed and feedback applied to moderate excesses. Here one must be careful of all the things you read as opinions are not facts (e.g. the misinformation about Serb concentration camps). The voice of reason and some critical thinking will go a long way towards reducing the fears people have about biotech. At this state biocomputing is so early that it's like we haven't invented the equivalent of the transitor yet, much less understand the many different intracellular processes in great detail. However, if and when innovative applications appear, some simple moral questions to ask are would you be able to sell the concept to your mother? If the development background was splashed on the national front page, would people still be interested (one reason why artifical substitutes have been found to replace animal testing)? And most importantly, could you look at yourself in the mirror in the morning if you were the inventor? People are not stupid and trying to force your beliefs or dogma onto others is a sure way to create enemies and annoy friends.
LL
Wasn't this topic discussed, oh, about a year ago and the /. community thrashed the concept about? Perhaps it's symptomatic of episoidal topics that get recycled every season or so (or a reflection of the continual growth of the /. readership) but perhaps it would be wise to include in the related links a link to the last closely relevant discussion (threshold=4) so that discussion can continue off the higher base rather than going through a process of reeducating as to the background. Also why not have a link to any forward (ie future) discussion as well? For people with limited time as any long-timer here is likely to be flatout employed on mission critical tasks (or burnt out) so short-circuiting the normal flamers and lamers and thus would get to the meat of the discussion more quickly. It would also give a better historical/future record of the crosslinked discussion thread allowing less informed souls to realise and hopefully appreciate the momentum of the topic. We get so caught up with day-to-day emergencies that often we don't appreciate that solutions have been thought out previously or smarter minds hae already considered the problem. /. is more than a newspaper or e-journal and it behooves the editors (and the pocketbooks of the onwers) to make /. a unique and distinctive experience.
LL
Brain drain is a rather emotional laden concept. However, it might be pointed out that there is no movement of people unless there is an empty position for them to shift into. The question should then be asked as to how come the US can't generate enough American talent to fufill these jobs? Given that professors at US unis get paid significant amounts of money, at least more than the average non-US uni, I would hope they give their students the skills to fufill the needs of the economy. If this is not the case and all the foreigners are taking the high paying jobs, where does that leave the average Joe? The fallacy in this type of thinking is limiting your thinking to just the US economy and assuming that to raise US standard of living, you have to depress the rest of the world. Economics is not a zero sum game.
However, it would also help if the Fed could get their money supply and debtunder control so that salaries and purchasing power parity figures for global comparison are not distorted. Even a starting Silicon Valley income can be net negative once you factor in teh cost of living.
LL
I believe it was Peter Drucker that said the role of a business is to create a need. You might consider the example of the early automative wars when safety belts was non-existant. It was only due to one persistent soul who dragged the car manufactuers kicking and screaming over the costs and hassles of installation. From their point of view, it would have been seem as scare-mongering and market fracturing. People forget that the market is a dynamic process, in essense you are trying to shape and alter the desires of consumers. One can only look at Amazon, their job is flogging books/CDs/widgets but their business is altering (and ultimately controlling) consumer habits. In this environment, especially intellectual property which is inherently limited by the wetware (ability of our brains to acceptand understand the product), FUD is a key weapon. Don't buy x because it is not the "standard". Use y because it will make your company more competitive. If you don't invest in z, you will disappear where x,y,z can be substituted for e-commerce, i-retailing, or the buzzword du jour. In other words, FUD can be a way of motivating your market to take certain action (perferably in transferring money from their pockets to yours).
The software market is slowly being transformed and the real fight is over where the balance of power (and thus control over spending) goes. Between the IP manufacturers and bundlers or the consultants and service people. FUD is a powerful weapon, as it can be used to deliberately suppress a competitor's stock price and thus create an easy take-over target using overpriced script as technology currency. Is it good or bad? Depends on whether you are the predator or the prey. Certainly Wall Street brokerage firms are not complaining. Fear and greed are two powerful emotions that can motivate individuals and FUD feeds on these elements. While from an engineering perspective, the technical advantages may be obvious, unfortunately life is not like that and good-enough tends to rule the day. While Linux anti-FUD may be a counteracting force, at the end of the day it is just the same thing in a different guise in trying to convince the market that your vision is a superior model despite the billions of capital lined up with proprietary solutions. As for the truth, whether benchmarking or quality testing, does anybody really care anymore? And this is the sad part.
LL
It is still an open question what role the junk DNA, technically called introns plays in organism development. Unlike the simple unicelluar critters (prokaryotes) such as bateria, all higher level organisms (eukaryotes) have these long non-coding sequences which have been retained across evolutionary generations despite the extra energy/space required. The whole area is akin to the physists search for all the various subatomic particles after the cracking of the atom. We can see the bits and pieces, we can assemble the various sequences, but there's no unifying standard model of how or why. With Nobel prizes and new killer apps in the air, it is not surprising that universities and institutes are throwing money into the research.
:-). Fun times ahead.
The 19th centure might have been the dominance of physics and engineering but there's a lot of speculation and anticipation (especially by the empty hands of the biologists and zoologists) that the next century will be their turn at the gravy train
LL
People have pointed out the technical disadvantages of WAP, particularly as a specification controlled by a "cartel". Well, technical standards are nice, but not necessary for commercial success. I suppose it is a bit naive to expect these manufacturers to give up the control of standards (see, Microsoft has shown how a pure software business model can really work). Businesses are in the process of making money, and if artificial scarcity drives up the entry barriers, so much the better (ever wonder why airlines offer frequent flyer points instead of just lowering the ticket-price?). Whether it is high switching costs, bundled tools, or the ability to impose rules that enforce the status quo, Linux hackers would be foolish to expect the business world to let go of a potentially lucrative cash cow. Without the ability to exclude, it would be impossible to create a monetary differential or perceived value. Heck, if they could meter the air you breathe (which they can already do for gas, bandwidth, electricity, and water) they would. Conditioning the consumer is a prime goal as it is easier to convince people to continue paying on a time-based charge a la mobile phone charges rather than the freebies people expect from the internet (even though the underlying technology is identical). If you haven't got a way of lining up the sheep ... err... consumers to be fleeced, then you might as well not be in business.
As for Linux, it still remains to be seen whether giving away your technology and time can create sustainable business models.
LL
Just to nitpick a little ...
/. and other sources. Hence how you act, think and reason can be judged by your peers long after the event without much chance of professional spin-doctoring. The truth might hurt, but hopefully it forges the character a little bit stronger and in a positive way. So although there is a similar effect to the media spotlight, I would trust that people take it in the same spirit as peer-review and supply positive criticism and feedback rather than throwing brickbats (unless it's well deserved stupidity). The only real answer to idiots is to ignore them.
:-(.
Despite the best attempts of media to convince the unwashed masses otherwise, there is a distinct difference between celebrity and fame. Part of it is to do with your time horizon and any distinctive or unique claim to a place in history. While most current pop stars or sports heroes will be forgotten in another decade or five, certain names (at least in the hacker sphere) will still be remembered (RS, Linus, ERS, etc) for changing the way we think and view the world. In a sense, fame is somewhat (initially) conversely opposite to popularity (ie raise a ruckus), where people (whether conciously or not) take a principled stand in their view and stick to their convictions. Unlike the standard 30 soundbites, your actions, words and deeds are recorded for prosperity in
Too bad none of the politicians have started a truth in media platform
LL
Bruce Perens wrote
/.ers only do two things well, initiate flamewars and inflate their karma :-). Don't know what the solution is, perhaps introduce a new rule that once your karma reaches 100 (tribal elder status?) you can moderate at any time and simultaneously post iff you're willing to sacrifice your existing karma. The alternative is to make sure that karma moderations on average sum to zero (or 1) (more difficult) so that people have an equal chance of going up/down and that the total karma doesn't tend to infinity and thus become overinflated (and thus devalued).
... the fun of designing a social currency.
Besides, I don't have Ulitmate Karma on Slashdot. Sig11's is higher than mine (I just happened to look at his user record a while back), probably other people as well.
Well, IMHO the absolute karma becomes rather meaningless after a while. I think what is more important is the relative karma/post ratio, ie some sort of median of how often your score goes up relative to the postings you make per week. I suspect people goes through a number of stages, initially posting lots to get the +1 bonus. Then more strategically (e.g. very early on in the thread to get noticed early) to race against the competition in running up your karma. Then I suppose the boredom stage sets in or your time runs out when more important things intrude in your life like an SO.
The biggest problem is that because you can create unlimited karma, the whole process gets out of control (can we say unbounded feedback loop?) and the whole system becomes inflated beyond recognition. To paraphrase Peter Drucker's saying that governments only do two things well, wage war and inflate the currency, I'd toss in the contention that
Ahhh
LL
One can point to the trend in furniture/kitchen in that the components (sinks/cupboards) are becoming commodities so furniture stores are turning into lifestyle/design/fitouts where they assemble a package tailored towards the customer's desires. Perhaps car dealers will shift towards a similar philosophy in ordering on-line, but offering mag-wheels, gee-whiz mobile-phone/stereo/DVD, beer cooler, leather seats, etc .... Heck it'd be just like designing a portable room :-). Everyone is going to know the price so the only way to add value is to offer services (insurance, repairs etc) that make a one-stop shop with physical visits worthwhile. If you're just moving wheels, then you will disappear.
As for retail, the basic stuff will be commoditised and automated but there will still be need for people to suggest alternatives and provide an admiring audience. Perhaps the days of the megamalls will disappear?
LL
I'm surprised people here are so ignorant about the realities of the marketplace.
... the XIO bus still beats the wet dreams of FutureIO and is here now) but then they do certify and guarentee the performance of the ones that do make the effort. If SGI was smart, they'd license the XIO technology to foster a vibrant alternative.
Facts
1) If a company doesn't have a competitive advantage, then sooner or later it will lose marketshare. The NT based Intel line (idea of the previous CEO) was targetting a niche which has been eroded by Dell's just-in-time assembly and creating a gap until HDTV standards become mandatory. SGI is being smart cutting their losses when they can. Unlike the disposable mentality that Wintel is trying to foster, SGI machines are designed to last for a long time (10 years+) which if your amortise the higher upfront costs, is actually cheaper in the long run.
2) SGI still has 300 odd engineers working on Irix and a growing pool of Linux developers to progressively shift their core competencies (multiprocessor design, low-latency memory subsystems) into value-added components. 10% of the market is seriously interested in high performance vs good enough.
3) Processors take a long time to die. Their MIPS line still has got excellent cost per application performance and decent data bandwidth (running a gigahertz Intel chip with IDE drives to surf the internet sounds a little fishy). As an embedded processor, the x86 is dwarfed by sales of MIPS cores cross-licensed to multiple vendors and their high-end stuff (R16000, etc) goes out to another 7 years.
4) The world market has been very very lucky in the Asian crisis has depressed global supply to such an extent that firms are willing to sell subsystems at near cost just to keep the cashflow up. In my opinion this has distorted the market and once things start bouncing back (as with the recent hike in memory prices) then their offerings will be more competitive.
5) Intel is becoming more like a venture capitalist than chip maker by throwing advanced designs at anything that looks, walks and talks like a high-growth startup. SGI is sorta doing the same thing by developing high-end, then flowing the tech to NDIVA (graphics), Cray (memory subsystem) and MIPS (CPU design).
6) Forget hardware, the value is now concentrated on the services. Even IBM are finding it difficult to move their big iron. Ultimately you will see companies becoming coordinators assembling components and software bundles to targetting specific market categories and capturing benefits from after-sales support and in the life-time money stream. You'd might be surprised how little car manufacturers make from sell the initial wheels compared with gouring consumers with repairs and add on knic-knacs.
7) Linux is shifting the competitive landscape in that every processors has an equal chance as the source is available to all players. Intel is recognising this and is dumping money to maintain dominance. If IBM can offer a PPC reference design, then perhaps SGI could do something similar for their MIPS, OpenGL and other graphics libraries (ie OpenAPI but retain implementation IP). Don't rule out the Koreas with their Alpha licensing and the Japanese with MIPS variants (e.g. Sony Playstation2).
8) Lighten up, markets go up and down in the short term. If anyone recalls the dumping Apple was going through several years ago, SGI is going through a similar patch. With a new CEO on board and some good Linux buzz, they have opportunities to catch the next wave. From a professional point of view, it would be interesting to see if their boxes could be adapted to keep the same fast memory subsystem but accept MIPS, Cray, Intel or Alpha processor node cards and just absolutely dominate the SMP and Beowulf market.
9) High end graphics is tough with very few firms having the capability to tackle complex real-time graphics and simulation. Today's multiplayer games are a shadow of hosting real-time thousands of players. If SGI could shift some of their talent to exploit the gaming niche, then they could gain more revenue streams from selling their IP. Admittedly they do have problems with delayed supply of third party boards (sigh
In short, SGI may be down at the moment but certainly not out. You don't have to be big so long as you can be profitable in your own little market niche, afterall BMW and Rolls-Royce still survive despite the Fords, Hondas, etc of the world. If people stop fixating on pure clock speed and look at real-world performance (e.g. broadband systems) they might be pleasantly surprised.
LL
... when you don't have the industrial-financial base to support it. It takes a heck of capital, both physical (plants, machinery, materials) and human (know-how, skills, training) to turn the ideas (stolen or otherwise) into real actual working machines. The US had to go nearly $6 trillion into debt to fund the Cold War (accumulation of deficits from Korean, Vietnam and various anti-communist insurgencies) and even then it was a rather close thing that they didn't go bankrupt before Russia did. And selling off Mickey Mouse hats and movies to the rest of the world is not going to pay this off either.
:-)
If you look at China (how many people here have actually visited that place?) you'd still find mud houses (strangely enough sitting next to a modern satellite equiped apartment block) in central Beijing. Their energy needs are still powered by high suphur content coal (one reason why interior cities are so polluted). And too many of their really good technical people leave and/or are snatched up by Western firms. Despite the war-mongering vote chasing by Bush, China is still pretty much an agarian society with much agriculture still being doing by hand. People forget that the US war machine is funded by taxes (40% of GDP?) from realatively high incomes (by world standards) that goes to pay for all those nicy shiny missiles (which have to be replaced after shooting up the reminants of Serbian infrastructure) and associated pork-barrelling. In fact some people would claim that the US economy is still on a war footing.
It is only now that resources are being released back into the civilian sector that you see such advances in communications (CDMA was originally a military application) and software. China's level of technical skill is comparatively poor and while they may be a world power by 2050, at the moment it is not exactly in a state to wave any sticks. IMHO, the US has more worries about trying to keep moral high and designing attractive careers for its own military personnel than worrying about any other army on the earth which they could beat hansomely (given enough warning) except the Isrealiis who have a (justifiably) paranoid and professional army. In other words, if you're Chinese and have a hankering for violence, the best idea is to give up now, emigrate to the US and learn how to play Quake.
LL
mjacksol wrote
:-) ).
Yes, but it is precisely this licensing model that is the problem, and it is Microsoft that had this stroke of evil genius. If I buy product X, which is essentially identical to product X+, but crippled to sell it cheaper, I feel ripped off.
Welcome to the world of market segmentation. Do you feel ripped off when you get that special discount air ticket that allows you to book holidays 4 weeks in advance? But surely (shock, horror) you get to your destination just as quickly as a first class passenger? The point is that the airlines deliberately add extra constraints and conditions to price match the demand of people and their willingness to pay. In effect, you pay extra for greater degrees of freedom (but don't tell the other passengers this
Now MS applies this principle to software and you think it is outrageous? I believe IBM had something similar with mainframe machines which were physically the same internally but had their clock speed governed by a hidden switch? If you didn't know about this switch, would you complain about paying less for a slower machine? Ignorance is bliss sometimes.
Differential pricing is not new and sooner or later, somebody will come up with an equivalent for the internet (e.g. jump to the head of the queue if willing to pay a premium). It would be very very interesting to see the pricing calculations and models MS (or any other software company) uses to determine market demand. Any ex-marketing guys out there?
LL
Oversoul wrote
So, back to what I was saying, if people in the tech industry are principled, ethical, clear-headed folks, item #4 (price out the talent) falls. The monster strangle-hold companies (like microsoft) have their knees cut out without the gray matter.
Your sentiments are very noble and I hope you get your chance to demonstrate your commitment. At this stage, I am reminded of the old saying, "If you're not a socialist by 20, you've got no heart. If you're not a capitalist by 40, you've got no brain".
Business is in the business of accumulating capital (ie wealth) and maximising its growth, usually by exchanging a slice of it (like from pension funds) for your time/labor/ideas. Good, bad or indifferent, the system has been structured to act that way through enforced laws and ingrained habits. The bottom line is that if you don't make enough to cover your own salary (or at least what the general market thinks you're worth working for another job), much less make a profit, then you (or your company) will get bought out and absorbed by a more efficient organisation. The only way OpenSource path can prove superior is only if it is more efficient at achieving business solutions or satisfying consumer desires. The evidence is still out on this but all is not lost as some studies show that cooperative behaviour creates long-term benefits (but as they say, in the long term we're all dead and giving away the fruits of your labor is not a obvious solution to paying bills the next day).
By the way, be prepared for the fact that some (or many) companies demand that you sign over every piece of intellectual value you create (even off-premises). Protecting yourself through foreknowledge is the only way to avoid becoming an economic slave to the system. Correct me if I'm wrong but some people complain that with the current starting salaries and cost of living in Silicon Valley, they are actually earning negative net income.
There's a reason why they call it the business jungle out there. Best of luck in your studies and in finding a worthwhile job when you get out.
LL