... it is the social (read human political) barriers. How often do you get organisations which have a disjoint culture agreeing to "share" resources? Scientists are no different as there's only a limited pot of government funds... ask the Department of *ENERGY* why they're doing genome research. Corporate cultures makes it difficult to merge, just putting together a backbone and lots of honey pots (CPU resources) doesn't automatically lead to a automagic collaboration of trust. Grid computing doesn't address the social issues... will my work be safe?, can I get a fair cut of the machine, will commercialisation contaminate standards, etc...
My point is that it takes a while for *HUMAN* systems to adjust to new technology waves. I would point out that in the early 1900s, factories were driven by belt-pulleys and machines (lathes/drills/press/etc) were contained in small 3-story buildings. Once electric motors got small enough and eliminated the physical requirement of being mechanically linked to the power source, then we could suddenly build whole acres of assembly plants and skyscrapers.
I see a necessary transition for software... someone in a distant/. post noted that the GPL promoted a wierd form of trust... because you knew the viral nature would eventually force publishing of any improvements, you had some confidence that the effort you put into developing software would (potentially) be amplified giving you improved down the track. The Sun Community Source License (SCSL) and Microsoft End-User License (MSFU) don't exactly inspire the same confidence and level of trust.
Currently TeraGrids are the beowulf of ASPs... but nothing different from a fancy queuing system. Other systems such as Globus are seriously researched but writing apps is still difficult. As for Microsofts.BET, it is stilll an unknown factor (and RPCs over low-latency internet doesn't exactly promote radically new killer apps). What does it require for a radically new level of trust (integrity, availability, confidentiality) to engineer the new killer apps? Chucking money at hardware without solving the human issues seem a little like an indirect government subsidy to the chip companies to me.
... is hard in that fact that you're so disadvantaged by the terrain (urban, buildings, limited LoS). Groups like SAS train extensively and I would wonder seriously about the sanity of any general sending in a bunch of grunts based on these games. Not that I doubt the army are lacking at simulation (they do excellent distributed interactive simulations with thousands of vehicles) but fundamentally because the retail gaming market is selling entertainment which is a direct conflict with building survival skills (paranoia rules!). Sure you can have semi-fun things like counter-strike but the kill-at-a-click mentality is not something I would want to see in the military... it's bad enough having police rough up civilians but asking army people to desensitise themselves to the point of being trigger happy is not my idea of peace mission which is what modern warfare is about... low-intensity pacification/de-arming missions (e.g. Timor).
ObJoke... a few years ago when the US visited Australia for military exercise, they did some flight combat simulations and thought it would be a fun thing to buzz the local kangaroos. Imagine their surprise when after disappearing around the hills, they popped up and let loose a bunch of Stinger missiles! Apparently the developers reused the objects for human soldiers and just modified their movement parameters to emulate wildlife. I know Australia has the reputation have having crocodiles and the most poisonous snakes in the world but when wombats start shooting back, then I'll really start to worry.
... is to refactor VNC to multicast directly to a bunch of Linux frame-buffers (a la SunRay). If companies are insisting on per CPU licensing and refusing to offer floating licenses (think legacy apps) then by running it on a half-decent back-end server (with fast storage) you can amortise the cost of the software over a wider geographical region, as well as support multiple legacy versions. Of course, you better have a decent network first.
BTE, whatever has happened to embedding X into the web browser (X11R7? Broadway?) How come that's not being used to port some of the older X utilities across to work over the internet?
Why do both MS and Sony want to control the broadband bridgehead into the living room? Because they can then become the toll-booth onto the distribution of electronic services. It may surprise people but Sony has acquired themselves a bank and MS own a controlling stake in a cheque clearing-house. Much like phone companies have to subsidise handsets and stick customers with the long-term contracts, everyone is gunning for a slice of the electronic services that businesses are switching over... you don't buy airline tickets, you bid for a seat, insurance, superannuation, identity, membership of professional societies, job contracts, even social contacts (rolodex on steroids)... all these are basically electronic goods that people will be willing to acquire.
The problems is making someone else fork out the capital for infrastructure, the smart people identify the bottlenecks and position themselves where the traffic concentration makes it worthwhile to extract their tax/toll/vig.
Nothing changed from highway robbery days except who gets to collect the loot.
... in that they sorta borrow money long-term (equity) and lend it short-term (purchasing depreciating hardware). Now this is OK if people are idiots enough to pay obscene amounts (dot.con) like clueless venture capitalists but if you get the situation where all your customers disappear (dot-bomb), you are left with a nice little term called negative cash-flow.
Seriously, unless the costs hit marginal pricing level, you have to be very very good to make money in the deflatory environment that Moore laws produces (as can be seen by the dire straits of many PC box-pushers).
Conclusion... buy monopolies like underwater sea cables.
Oddly enough, the talk of press releases and social-karma (or corporate equivalent of avoiding brand tainting) seems to work in imposing social norms on recalcite companies. This is not to claim that the GPL is good or bad, but to point out that they are doing the modern equivalent of the church punishing obnoxious behaviour in the middle ages by parading perpetuators around in public with ugly masks and placards. The effectiveness might be questionable but there's no doubt that people's desire to conform (peer pressure) is a powerful psychological force (sometimes excessively so in teenagers) and it does have the advantage of being cheaper than lawsuits.
The problem is so far the main groups that can take advantage of open source are the relatively prosperous western countries. Enforcement of the GPL license is going to be harder in places which don't respect intellectual endeavours (Eastern pirates), much less international laws (Taliban, etc). Ultimately laws are self-imposed constraints... we don't murder people because we understand the consequences of arbitrary violence. The GPL, whether mindless ideology or social conviction, is no less powerful in that at least people in the hacker community recognise the benefits and are willing to follow the principles.
The interesting fact about a global software economy is that reputation becomes so much more important. When details are kept track of contributors in freshmeat, sourceforge, etc. Old fashioned social ties are reused to subtlely enforce trust. Would you start up a company or work with someone you know that doesn't recognise the legimacy of software licenses? Public naming and threats to lose "face" in front of peers may ultimately unveil all sins (programming or otherwise).
... someone once said that VCs want impossible goals. How many business opportunities that existing with triple digits compound growth, clear exit strategy, and quantifiable risk? Nobody wants to be first to bake but everyone wants a double helping of the successful projects. The very nature of investments (harking back to the British India company) is to create competitive/proprietary positions which means exclusion of some sort, whether knowledge or opportunities.
Unfortunately VCs are the only people willing to invest in high risk (read unknown to them) speculative ventures. Banks are basically pawn-brokers and bean-counters, they only risk their money on assets which have a ready secondary market. But unfortunately there's none for failed (or half-finished) ideas which leads to a fair amount of cluelessness. I've just come from a dinner where someone said that the only reason a "VC" invested in their company was that they read in Red Herring that nanotechnology was going to be "big" and they thought a name like Nano-xxxx (name disguised to hide the guilty) was related... wait until someone tell them they bought 40% share of a electron microscope:-). What VCs continually forget is that they are investing in people, not business plans.
Oh well... at least every engineer has got someone else to blame for the stress:-).
... consisting of $100 for hardware and $400 for the MS software. In this situation, you might as well accept that the software is an embedded device anyway... in which case people will just buy multiple units and switch between them using wireless ethernet... watch for MS entering the home networking in a big way, especially if they can license their.NET + XML extensions directly into the router/concentrator/broadband bridge and isolate Intel/Cisco in the process (remember that silicon is just frozen software).
The question is how come nobody has shown up the OEM license agreement a la the Halloween leak? I'm sure anti-competitive bodies around the world would be more than interested in looking at the exclusionary covenants and tying agreements.
The claim that international rogue commercial elements (corporations of concern) is irrelevant is a little disingeneous as the real point is that any place which requires large-scale infrastructure will usually be located in a developed country (techs who operate multi-million dollar backbones and data centres don't get born in Siberia) which usually is reachable from other developed countries, either legally, cutting cables or last resort of nuking them back to the stone age. Economically this is the basic barrier to entry which doesn't work in the early stages of a technology paradigm shift (did the telcos see the ISPs taking off?). If someone combines the equivalent of napster with individual wireless (as compared to cell-towers), are you going to have the BSA confiscating every antenna or PDA they see?
The second claim of the network being resilient is actually a two edged sword, bringing social intrusion of foreign laws, spam and generally lower signal-noise ratios. As more devices/software interact in increasigly complicated ways, predictability is going to appeal to consumers which means the path of least resistence will be followed. As other people have discovered, frankly there's not much interest in communicating with clueless dweebs outside professional or social circles which means that ultimately the human network is self-limiting. I believe the statistic is that we can keep track of ~200 odd names/faces.
The claim that the hackers is irrelevant is only true if you consider life from a VC point of view... if you're betting on a proprietary technology and are willing to put the marketing dollars behind it, then at best hackers are a semi-persistant nuisance (as evidenced by the European satellite TV). However, studies have also shown that economic growth is highly dependent on exogeneous factors, primarily technology. Hackers, as free agents, can scratch their own itches as they have both the talent (probably) and time (hopefully) creating new applications where people didn't realise they would want that product/service/etc. Did anyone have a focus group to discover Hypercard or Visicalc? Would Counterstrike have been supported in a corporate lab? Would the next RMS introduce a philosophy or Linus-to-be implement casually an idea which changes how you live? While the individual footprint of hackers may be neglibible, the feet of a thousand penguins can lead to surprising destinations.
People keep on forgetting that public media is in the attention business... it is in their economic interest to sensationalise news in order to flog those accompanying ads (a bit of a problem for CNN when there's no major wars going on). As such, manufactured fluff (ie press-releases) is easier to regurgitate than any in-depth research or second-hand opinions (syndicated columns). Historically public media was part and parcel of the lecture circuit (aka rubber chicken show) where you would invite real authorities and experts to come in and air their thoughts in a proper interview. However, two general trends mitigate against this... the increasing complexity of real-world issues (anyone who thinks Middle-East is a simple case of good-guys/bad-guys is in deep trouble) which limits the potential audiences interested in understanding the issues, and the move to tabloid style audience capture which tends to confuse celebrity with fame. Why should sports-heros and actresses (apart from the convenience of recycling pre-existing studio contracts) be ask to comment on areas way outside their domain? Why should TV shows get people from the entertainment field to present business news (and you wonder why the stock market is irrational). Unfortunately those with real knowhow gained from the school of hard knocks tend to be people who charge for their services... would you want a surgeon who has never practised on patients before so why are we willing to listen to highly filtered news passed along by talking heads? It's becoming nothing more than a massive Chinese whisper in a global cocktail party.
Fortunately we have some countervailing examples... the/. ask XYZ is a particularly good way for the plebs to touch some of the people involved in the thick of things.
LL
The economic models which are the equivalent of yelling "fire" as loudly as possible to rush people to newsfeeds are creating some really bad incentives... ultimately people have to realise there is a cost in misleading/diluted information (e.g. did anyone notice that the bard-wire concentration camp story that helped sparked the Balkans intervention and sundry bombings was filmed on the inside looking out?).
Hmmm, and here I was thinking that the SV1 was basically a cluster of J90s (admittedly with souped up processors... lost track of whether they called them the S+ or SE now) and some rather beefy I/O. If you're looking at raw vector grunt, then the NEC SX series is rather impressive though supplies may not have resumed after that anti-dumping action was lifted. Cray has not really produced a top-end vector machine since their T90s and with the Japanese hell bent on their Whole Earth Simulator project (40 Tflops), I don't really see the US catching up anytime. And no, a beowulf of Itaniums don't count unless the problem is embarassingly parallel and your compiler cooperates.
Anyway, now that Cray has been purchased by Tera (the guys who developed that highly threaded CPU) it will be interesting to see their technical direction. In terms of processor development, theirs is the only vaguely interesting CPU that has reached the semi-commercialisation stage.
As with any system, there are trade-offs. The PSX2 graphics chip (EmotionEngine) has a number of features which require careful programming to exploit, especially given the (relatively) limited main memory (32M). The question is to what extent you expect to compromise your API's growth in capability/portability to support gee-whiz features. The dual of this question is of course that with a cross-platform API, you tend to end up with something equally mediocre across all systems, potentially leading to a catch-up mentality in the marketing perception which is particularly fatal in the computer gaming sector which continually relies on new features to draw in users. How do you expect to handle these compromises without alienating too many of the stakeholders (developers, manufacturers, users, etc)?
... then think what artists could do with this starting from finger painting up to conducting. It does look like it's different from the usual touch-screen (no pizoelectric scanning) which makes you ask the question how can you program it to recognise distinctive gestures? Does the software have a learning component which is unique for each individual? Can you adapt this for a biometric device? What about its spatial resolution (they claim 50-200 Hz), is it high enough that you can use a stylus for very fine-grained work? If you put lego blocks and mapped them onto software components, can you create rapid layouts by shifting them around? What about multiple people, could it segment 3-4 different hands for multi-user input?
There are some intriguing possibilities if you get away from the impression that it is a keybaord, and look at it as a generic input surface.
... and pray tell me who is going to analyse those Terabytes of data that your multimillion satellite has collected? And no, this is not a SETI@home project as image processing is data-intensive and the datapath and memory hierarchy within a PC is not exactly well tuned for these high throughput systems.
The NSA has similar issues in that irregardless of Echalon, they still need skilled analysts to interpret the information, computer filters notwithstanding. Could you perform a vegetation cover auto-correlation with the spatial extent and connectivity of basin drainage? If so, volunteer your computer and expertise.
Some posters have asked why companies would want to have their boxes back. My suspicion is that like all good potential monopolists, they want to destroy the secondary market. To help explain the economics, bear with me as I go through some MBAthink.
Basically the real value of any capital good is the price that the *next* buyer is willing to pay, ie the resale value. It doesn't matter if the house you bought at auction cost you $1M if the next person (given a free choice) is only willing to pay $500K. This disconnect between (sunk) cost and (future) value has been the downfall of many speculators (cough*dot con shares*cough).
Now for a good/service/title to accurately priced, there must be enough instances around so that potential future buyers can evaluate their utility and alternative factors of consumption. Thus for something like land, it could be zoned for residential, business, recreation, mixed, etc and different people would value each according to their needs. The secondary commercial leasing market allows much more flexilibility than just risking a big lump sum on a potentially unsuitable structure.
Another reason the secondary market arises when primary purchasors want liquidity, ie they find that they wish to redeploy capital (e.g. second mortgage for startup business). So even if you bought that house but can't make much use of it due to excessing work hours, you can lease it out. The biggest example is the share market which was originally a mechanism for the investment banks / underwriters to offload risk of being caught with an investment that was going downhill (after of course getting the best profits for themselves due to insider information, etc). Secondary markets are important in the sense that if they are large enough, they actually give very good price signals (cf efficient market hypothesis, random walks, etc).
Now how does this apply to the consumer services model? Basically the problem (from the business point of view) of computers (hardware) is that they can be used for anything and everything (software). Unfortunately the issue with any new technology is convincing users that they have a need for that junk... err... status symbol (cough*Palm*cough). Thus to convince people to adopt (google bridging the chasm), they tend to create loss leaders or subsidised on-ramps (cf bundled plans for mobile phones) to convince the potential users that the price is really affordable. Now as any hacker with half a clue realises, any electronic device with an unsecure interface (cough*cuecat*cough), can be refactored into something more practical and thus useful. So the combination of refactorable loss-leaders and a savvy user population tends to create parasitic markets where you cannabalise sales from one segment into a lower-margin one. Hence the desire to eliminate competition for themselves by offering closed devices, dumbing the user base (by hiding stuff or increasing complexity), as well as the standard retail devices of deliberately having a short shelf-life and guaranteed obsolescence.
Now the incentives for destroying the secondary market is that people don't have a chance to properly value/price the alternative uses and that liquidity is removed increasing barriers to competitors, allowing the corporation to get away with a higher priced primary market (and the all important profits, options and return to shareholders). This can be seen most clearly in things like the technical textbook market where the Doctrine of First Sale means that publishers prefer coming out with new editions (and incompatibilities... err erata) rather than allowing students to resell their textbooks to other hardup students. IBM was the classic case of a monopolist that deliberately leased their gear to prevent a secondary market (google Amdahl/IBM marketing practices). If you read any standard MBA handbook, you realise the mind-boggling sales tactics at work which is naturally anthema to any self-respecting engineer who wants to look under the hood before buying... alas engineers are not a majority of the sheeple population.
Now applying this impromptu MBA lesson shows that from a business point of view, it is easier to sell future goods/services if you can eliminate the infrastructure for any competitors (the so-called deep entry moat). Hence convincing the customer that *the firm* owns the gear (despite using the retail system) and thus can exercise control over recall (naturally at their convenience). Obviously with electronic stuff this is a problem because people think of it as a manufactured *good* (aka appliance) rather than on-going *service* (maintenance). Hence the serious pricing problems since when you outright buy a good, you usually do so if the price (present value) is less than the value of all the future services you expect out of it. This disconnect is going to cause a lot of sustainability problems in the long-term as the accounting rules for capital items and software don't really reflect the real cost of services (software stability, interoperability quality, service level), just like inefficient market for medical pricing results in invisibile costs such as long waiting lists.
At least with open source you know exactly what you are getting up-front, the right that your contribution will be reflected and amplified in future iterations and refinements of the software. This is not the case with commercial providers that vary the terms of usage at their whim (see the Sexual Practices of Licenses at http://slashdot.org/comments.pl?sid=01/06/21/18102 58&cid=66).
Part of the problem is that there are just way too many variants of licenses, whether OSS or otherwise. Part of the problem is that what browse-through license are is that they are actually a service level agreement e.g. I guarantee such a level of performance, provided you stay within the nominated activity envelope. Given the inherent complexities of computers plus the hair-splitting legal mumbo-jumbo, along with zilch consumer education and you basically get a system which treats license as no more than a disclaimer. With software, it is somewhat easier as you can split them into classes (GNU, BSD, Mozilla) with specific instances (Artistic, etc) with stated variations of terms (e.g. Alaladin is GNU-like but with temporal phase-shift, etc). Unbfortunately the internet world has not really standardised on an equivalent unless you count the MSFU (see the Sexual Practices of Licenses at http://slashdot.org/comments.pl?sid=01/06/21/18102 58&cid=66) which changes every week.
How well can you define a service? Using RDF you can probably specify the functional aspects but how do you nominate remedies for down-time, contacts for contingencies, independent auditing of claims (99% availabilty, etc). Ultimately you're probably see network access like a utility once the standard performance metrics are defined (MByte/sec up/down, latency (up/down) disruption distribution function, etc), and software quality is improved to such a degree that you can provide customer guarantees. But I don't see this happening until there's more localised competition rather than the big media titans duking it out.
... apart from the horse racing, the pokies, the cricket matches, the.... etc, etc...
There's already a sizeable grey economy despite the red tape of the Grab Snatch and Take (GST) tax. The average Australian doesn't take much notice of what goes on in Canberra and I seriously doubt whether a fiat dictate on internet gambling is going to make much of an impact. Gambling is really a stupidity tax on people who can't do maths so I suspect this latest triump of parliamentary wisdom is really due to the politicians hating competition.
Correction... the scientific publishing model is that research results be available at the marginal cost of distribution. Freelance writers who live purely on the basis of their commentary/articles do not fall into this category as their papers are not ads for their real output (reproduceable theoretical/experimental/computable results). As such they have a legitimate claim on their sole output.
The real problem is that there is too much rubbish out there (hey given copyrights of 90 years it's easier to recycle than to create) and it is hard to distinguish the cruft from the craft.
The most recent calculations show
that the annual cost to consumers and taxpayers of its 29 member countries' support for agriculture and horticulture amounts to US$361 billion. Such a large sum is difficult to grasp, but it is large enough to pay for a first class, round-the-world air ticket for each of the 56 million cows in the 29-member OECD?s dairy herd, and to give each cow a further US$1450 spending money for her stopovers in the US, Europe and Asia.
Thus while there may be laudable social objectives in keeping Scottish farmers around, there is a serious economic cost which makes you wonder why don't they just give the money directly to funding a Silicon Glen.
There's some interesting speculative thoughts about computers and their impact on law... one of the interesting claims found in <A HREF="http://www.erights.org/smart-contracts/index.html">www.erights.org</A> is that we are due for an era of prosperity (hah!) but only if we can bury copyright (as we know it). Given that the whole point of copyright is to restrict exchange (creating semi-artificial scarcity), the GNU approach of copy-left is already half a step along this path. Now I don't have the experience to judge whether the rest of the claims of utopia (capability computing, crypotographic protocols, smart contracts) will automatically lead to a brave GNU world but it is intriguing. Many of our concepts of "wealth" are likely to change over the next decade. In the pre-industrial age it was ownership of land, in industrial revolution it was access to resources (cough*colonies*cough), in the post-industrial economy a liquid capital market. Perhaps in the knowledge economy it will be computer-enforced contracts... I promise not to create a BSOD when undertaking action XYZ. How much is such a promise worth? If you can attach a negative value to lost productivity, will Free software actually be cost-effective? This is going to be a really interesting area of research... the ability to guarantee a software promise.
<P>
LL
... I would refer you to this article which discusses structured environments. Basically to be efficient at something, e.g. preparing a dish, we line up all the instruments in a particular order to help cue us when a particular task needs to be done. Extending this into a semi-spatial (if ESR can get the right relationship of in-on-up-etc representing the true connectivity between kernel modules) setting would help people orientate themselves and get to work faster (so you can get back to that QUAKE game). Just like we would put a letter by the door, or shopping list by the car keys, each cue triggers associated memories and reminds us of specific actions that need to be done. Sure you can have a linear check-list like the space shuttle pilots but anything computing-wise is so variable that a more flexible arrangement is desirable. I've been looking into something similar for the make processing for reproduceable documents and it is not as easy (or trivial) as people think (at least to get right). For example, color assignment... do you map this property to time-of-last-modification (ie heat colors) or to likely hazard (red=stop, orange=hazard, yellow=caution, etc). If you dig far enough, you eventually realise it is actually a text variant of scientific visualisation but in the qualitative domain. There are a number of theories on how you allocate the properties based on human cognitive functions (see Lloyd w.r.t. OpenDX and fiber bundles)... but there are a lot of issues remaining such as what works for 5-6 objects doesn't work for thousands.
Like all ideas, let the users decide and if it works, it will be included into the meme-pool. Imagine TuX, the penguin avatar trundling around kernel space looking for fish or and fixing security leaks. At the worse, it will provide a few minutes of amusement.
LL
Generally the principles under which students contribute work falls into the
1) copyright to thesis belongs to student (you write it, you own it)... I presume the code is attached as an appendix
2) any restrictions (e.g. on publication) are time0limited (e.g. Pharmaceutical companies usually don't hold up publication more than 6 months, 18months is very extreme)
3) any examiner may be required to provide acknowledgement of confidentiality
The bottom line is that a supervisor, if they are smart, sets out ahead of time a written agreement stating exactly what is expected as participation of a project. If you have moral qualms about fueling the military industrial complex, that's the time to bow out gracefully and ask to be reassigned.
OTOH, universities can ask for the right to an outcome, for example rights to use that material for teaching purposes. The nasty situation is when a tutor has leeway in slection of assignments and gains a personal financial benefit (e.g. external party consulting) but universities usually have groundrules on this (e.g. no more than 20% of time, or waiver on consulting outcomes). Obviously some universities are more careful than others and there have been some stories about early career researchers being basically fleeced of their efforts. Students, because they tend to be doing bulk/mass assignments are usually working a known problem in which case there is no real intellectual advances and it would be very hard to assert IP rights except as a performance (rearrange widgets using this script).
As always, reading the rules before signing on the dotted line as there's always the small print to stumble over.
Currently there are a number of XML billing initiatives such as XMLPay, VisaXML, etc... How do you see these trends influencing the design and development of GNUcash... in particular do you think that your users are only looking for a simple view/format/verify client (think IMAP server + XML extensions) while GNUcash moves towards being the agent of record (aka a specialised ASP?) Of you do see GNUcash as a drop-in replacement for existing monolithic accounting packages? In summary, given that the market for electronic businesses and purchases (greater use of international currency, direct attachment to electronic settlement, near real-time risk management) is going to change, how do you think GNUcash should respond?
I'm surprised more people don't use it... it's really useful for managing multiple versions of libraries and different applications which depend on them. Unforuntately when you start hitting C++ class interdependencies or zillions of versions it starts being tricky... which is where namespace managment and scoping rules become more important.
And the DIY home decorating industry has been a very profitable source of income for those handymen able to come in and fix up your mess... unfortunately the jumble of components nad scripts is underneath the hood hidden behind a gee-whiz veneer for too many people to care.
Untile people appreicate professionalism and artistry, expect the marketing hype to dominate.
Slashdot Mirror
... it is the social (read human political) barriers. How often do you get organisations which have a disjoint culture agreeing to "share" resources? Scientists are no different as there's only a limited pot of government funds ... ask the Department of *ENERGY* why they're doing genome research. Corporate cultures makes it difficult to merge, just putting together a backbone and lots of honey pots (CPU resources) doesn't automatically lead to a automagic collaboration of trust. Grid computing doesn't address the social issues ... will my work be safe?, can I get a fair cut of the machine, will commercialisation contaminate standards, etc ...
... someone in a distant /. post noted that the GPL promoted a wierd form of trust ... because you knew the viral nature would eventually force publishing of any improvements, you had some confidence that the effort you put into developing software would (potentially) be amplified giving you improved down the track. The Sun Community Source License (SCSL) and Microsoft End-User License (MSFU) don't exactly inspire the same confidence and level of trust.
... but nothing different from a fancy queuing system. Other systems such as Globus are seriously researched but writing apps is still difficult. As for Microsofts .BET, it is stilll an unknown factor (and RPCs over low-latency internet doesn't exactly promote radically new killer apps). What does it require for a radically new level of trust (integrity, availability, confidentiality) to engineer the new killer apps? Chucking money at hardware without solving the human issues seem a little like an indirect government subsidy to the chip companies to me.
My point is that it takes a while for *HUMAN* systems to adjust to new technology waves. I would point out that in the early 1900s, factories were driven by belt-pulleys and machines (lathes/drills/press/etc) were contained in small 3-story buildings. Once electric motors got small enough and eliminated the physical requirement of being mechanically linked to the power source, then we could suddenly build whole acres of assembly plants and skyscrapers.
I see a necessary transition for software
Currently TeraGrids are the beowulf of ASPs
LL
... is hard in that fact that you're so disadvantaged by the terrain (urban, buildings, limited LoS). Groups like SAS train extensively and I would wonder seriously about the sanity of any general sending in a bunch of grunts based on these games. Not that I doubt the army are lacking at simulation (they do excellent distributed interactive simulations with thousands of vehicles) but fundamentally because the retail gaming market is selling entertainment which is a direct conflict with building survival skills (paranoia rules!). Sure you can have semi-fun things like counter-strike but the kill-at-a-click mentality is not something I would want to see in the military ... it's bad enough having police rough up civilians but asking army people to desensitise themselves to the point of being trigger happy is not my idea of peace mission which is what modern warfare is about ... low-intensity pacification/de-arming missions (e.g. Timor).
... a few years ago when the US visited Australia for military exercise, they did some flight combat simulations and thought it would be a fun thing to buzz the local kangaroos. Imagine their surprise when after disappearing around the hills, they popped up and let loose a bunch of Stinger missiles! Apparently the developers reused the objects for human soldiers and just modified their movement parameters to emulate wildlife. I know Australia has the reputation have having crocodiles and the most poisonous snakes in the world but when wombats start shooting back, then I'll really start to worry.
ObJoke
LL
... is to refactor VNC to multicast directly to a bunch of Linux frame-buffers (a la SunRay). If companies are insisting on per CPU licensing and refusing to offer floating licenses (think legacy apps) then by running it on a half-decent back-end server (with fast storage) you can amortise the cost of the software over a wider geographical region, as well as support multiple legacy versions. Of course, you better have a decent network first.
BTE, whatever has happened to embedding X into the web browser (X11R7? Broadway?) How come that's not being used to port some of the older X utilities across to work over the internet?
LL
Why do both MS and Sony want to control the broadband bridgehead into the living room? Because they can then become the toll-booth onto the distribution of electronic services. It may surprise people but Sony has acquired themselves a bank and MS own a controlling stake in a cheque clearing-house. Much like phone companies have to subsidise handsets and stick customers with the long-term contracts, everyone is gunning for a slice of the electronic services that businesses are switching over ... you don't buy airline tickets, you bid for a seat, insurance, superannuation, identity, membership of professional societies, job contracts, even social contacts (rolodex on steroids) ... all these are basically electronic goods that people will be willing to acquire.
The problems is making someone else fork out the capital for infrastructure, the smart people identify the bottlenecks and position themselves where the traffic concentration makes it worthwhile to extract their tax/toll/vig.
Nothing changed from highway robbery days except who gets to collect the loot.
LL
... in that they sorta borrow money long-term (equity) and lend it short-term (purchasing depreciating hardware). Now this is OK if people are idiots enough to pay obscene amounts (dot.con) like clueless venture capitalists but if you get the situation where all your customers disappear (dot-bomb), you are left with a nice little term called negative cash-flow.
... buy monopolies like underwater sea cables.
Seriously, unless the costs hit marginal pricing level, you have to be very very good to make money in the deflatory environment that Moore laws produces (as can be seen by the dire straits of many PC box-pushers).
Conclusion
LL
Oddly enough, the talk of press releases and social-karma (or corporate equivalent of avoiding brand tainting) seems to work in imposing social norms on recalcite companies. This is not to claim that the GPL is good or bad, but to point out that they are doing the modern equivalent of the church punishing obnoxious behaviour in the middle ages by parading perpetuators around in public with ugly masks and placards. The effectiveness might be questionable but there's no doubt that people's desire to conform (peer pressure) is a powerful psychological force (sometimes excessively so in teenagers) and it does have the advantage of being cheaper than lawsuits.
... we don't murder people because we understand the consequences of arbitrary violence. The GPL, whether mindless ideology or social conviction, is no less powerful in that at least people in the hacker community recognise the benefits and are willing to follow the principles.
The problem is so far the main groups that can take advantage of open source are the relatively prosperous western countries. Enforcement of the GPL license is going to be harder in places which don't respect intellectual endeavours (Eastern pirates), much less international laws (Taliban, etc). Ultimately laws are self-imposed constraints
The interesting fact about a global software economy is that reputation becomes so much more important. When details are kept track of contributors in freshmeat, sourceforge, etc. Old fashioned social ties are reused to subtlely enforce trust. Would you start up a company or work with someone you know that doesn't recognise the legimacy of software licenses? Public naming and threats to lose "face" in front of peers may ultimately unveil all sins (programming or otherwise).
LL
... someone once said that VCs want impossible goals. How many business opportunities that existing with triple digits compound growth, clear exit strategy, and quantifiable risk? Nobody wants to be first to bake but everyone wants a double helping of the successful projects. The very nature of investments (harking back to the British India company) is to create competitive/proprietary positions which means exclusion of some sort, whether knowledge or opportunities.
... wait until someone tell them they bought 40% share of a electron microscope :-). What VCs continually forget is that they are investing in people, not business plans.
... at least every engineer has got someone else to blame for the stress :-).
Unfortunately VCs are the only people willing to invest in high risk (read unknown to them) speculative ventures. Banks are basically pawn-brokers and bean-counters, they only risk their money on assets which have a ready secondary market. But unfortunately there's none for failed (or half-finished) ideas which leads to a fair amount of cluelessness. I've just come from a dinner where someone said that the only reason a "VC" invested in their company was that they read in Red Herring that nanotechnology was going to be "big" and they thought a name like Nano-xxxx (name disguised to hide the guilty) was related
Oh well
LL
... consisting of $100 for hardware and $400 for the MS software. In this situation, you might as well accept that the software is an embedded device anyway ... in which case people will just buy multiple units and switch between them using wireless ethernet ... watch for MS entering the home networking in a big way, especially if they can license their .NET + XML extensions directly into the router/concentrator/broadband bridge and isolate Intel/Cisco in the process (remember that silicon is just frozen software).
The question is how come nobody has shown up the OEM license agreement a la the Halloween leak? I'm sure anti-competitive bodies around the world would be more than interested in looking at the exclusionary covenants and tying agreements.
LL
The claim that international rogue commercial elements (corporations of concern) is irrelevant is a little disingeneous as the real point is that any place which requires large-scale infrastructure will usually be located in a developed country (techs who operate multi-million dollar backbones and data centres don't get born in Siberia) which usually is reachable from other developed countries, either legally, cutting cables or last resort of nuking them back to the stone age. Economically this is the basic barrier to entry which doesn't work in the early stages of a technology paradigm shift (did the telcos see the ISPs taking off?). If someone combines the equivalent of napster with individual wireless (as compared to cell-towers), are you going to have the BSA confiscating every antenna or PDA they see?
... if you're betting on a proprietary technology and are willing to put the marketing dollars behind it, then at best hackers are a semi-persistant nuisance (as evidenced by the European satellite TV). However, studies have also shown that economic growth is highly dependent on exogeneous factors, primarily technology. Hackers, as free agents, can scratch their own itches as they have both the talent (probably) and time (hopefully) creating new applications where people didn't realise they would want that product/service/etc. Did anyone have a focus group to discover Hypercard or Visicalc? Would Counterstrike have been supported in a corporate lab? Would the next RMS introduce a philosophy or Linus-to-be implement casually an idea which changes how you live? While the individual footprint of hackers may be neglibible, the feet of a thousand penguins can lead to surprising destinations.
The second claim of the network being resilient is actually a two edged sword, bringing social intrusion of foreign laws, spam and generally lower signal-noise ratios. As more devices/software interact in increasigly complicated ways, predictability is going to appeal to consumers which means the path of least resistence will be followed. As other people have discovered, frankly there's not much interest in communicating with clueless dweebs outside professional or social circles which means that ultimately the human network is self-limiting. I believe the statistic is that we can keep track of ~200 odd names/faces.
The claim that the hackers is irrelevant is only true if you consider life from a VC point of view
LL
People keep on forgetting that public media is in the attention business ... it is in their economic interest to sensationalise news in order to flog those accompanying ads (a bit of a problem for CNN when there's no major wars going on). As such, manufactured fluff (ie press-releases) is easier to regurgitate than any in-depth research or second-hand opinions (syndicated columns). Historically public media was part and parcel of the lecture circuit (aka rubber chicken show) where you would invite real authorities and experts to come in and air their thoughts in a proper interview. However, two general trends mitigate against this ... the increasing complexity of real-world issues (anyone who thinks Middle-East is a simple case of good-guys/bad-guys is in deep trouble) which limits the potential audiences interested in understanding the issues, and the move to tabloid style audience capture which tends to confuse celebrity with fame. Why should sports-heros and actresses (apart from the convenience of recycling pre-existing studio contracts) be ask to comment on areas way outside their domain? Why should TV shows get people from the entertainment field to present business news (and you wonder why the stock market is irrational). Unfortunately those with real knowhow gained from the school of hard knocks tend to be people who charge for their services ... would you want a surgeon who has never practised on patients before so why are we willing to listen to highly filtered news passed along by talking heads? It's becoming nothing more than a massive Chinese whisper in a global cocktail party.
... the /. ask XYZ is a particularly good way for the plebs to touch some of the people involved in the thick of things.
... ultimately people have to realise there is a cost in misleading/diluted information (e.g. did anyone notice that the bard-wire concentration camp story that helped sparked the Balkans intervention and sundry bombings was filmed on the inside looking out?).
Fortunately we have some countervailing examples
LL
The economic models which are the equivalent of yelling "fire" as loudly as possible to rush people to newsfeeds are creating some really bad incentives
Hmmm, and here I was thinking that the SV1 was basically a cluster of J90s (admittedly with souped up processors ... lost track of whether they called them the S+ or SE now) and some rather beefy I/O. If you're looking at raw vector grunt, then the NEC SX series is rather impressive though supplies may not have resumed after that anti-dumping action was lifted. Cray has not really produced a top-end vector machine since their T90s and with the Japanese hell bent on their Whole Earth Simulator project (40 Tflops), I don't really see the US catching up anytime. And no, a beowulf of Itaniums don't count unless the problem is embarassingly parallel and your compiler cooperates.
Anyway, now that Cray has been purchased by Tera (the guys who developed that highly threaded CPU) it will be interesting to see their technical direction. In terms of processor development, theirs is the only vaguely interesting CPU that has reached the semi-commercialisation stage.
LL
As with any system, there are trade-offs. The PSX2 graphics chip (EmotionEngine) has a number of features which require careful programming to exploit, especially given the (relatively) limited main memory (32M). The question is to what extent you expect to compromise your API's growth in capability/portability to support gee-whiz features. The dual of this question is of course that with a cross-platform API, you tend to end up with something equally mediocre across all systems, potentially leading to a catch-up mentality in the marketing perception which is particularly fatal in the computer gaming sector which continually relies on new features to draw in users. How do you expect to handle these compromises without alienating too many of the stakeholders (developers, manufacturers, users, etc)?
LL
... then think what artists could do with this starting from finger painting up to conducting. It does look like it's different from the usual touch-screen (no pizoelectric scanning) which makes you ask the question how can you program it to recognise distinctive gestures? Does the software have a learning component which is unique for each individual? Can you adapt this for a biometric device? What about its spatial resolution (they claim 50-200 Hz), is it high enough that you can use a stylus for very fine-grained work? If you put lego blocks and mapped them onto software components, can you create rapid layouts by shifting them around? What about multiple people, could it segment 3-4 different hands for multi-user input?
There are some intriguing possibilities if you get away from the impression that it is a keybaord, and look at it as a generic input surface.
LL
... and pray tell me who is going to analyse those Terabytes of data that your multimillion satellite has collected? And no, this is not a SETI@home project as image processing is data-intensive and the datapath and memory hierarchy within a PC is not exactly well tuned for these high throughput systems.
The NSA has similar issues in that irregardless of Echalon, they still need skilled analysts to interpret the information, computer filters notwithstanding. Could you perform a vegetation cover auto-correlation with the spatial extent and connectivity of basin drainage? If so, volunteer your computer and expertise.
LL
Some posters have asked why companies would want to have their boxes back. My suspicion is that like all good potential monopolists, they want to destroy the secondary market. To help explain the economics, bear with me as I go through some MBAthink.
... err ... status symbol (cough*Palm*cough). Thus to convince people to adopt (google bridging the chasm), they tend to create loss leaders or subsidised on-ramps (cf bundled plans for mobile phones) to convince the potential users that the price is really affordable. Now as any hacker with half a clue realises, any electronic device with an unsecure interface (cough*cuecat*cough), can be refactored into something more practical and thus useful. So the combination of refactorable loss-leaders and a savvy user population tends to create parasitic markets where you cannabalise sales from one segment into a lower-margin one. Hence the desire to eliminate competition for themselves by offering closed devices, dumbing the user base (by hiding stuff or increasing complexity), as well as the standard retail devices of deliberately having a short shelf-life and guaranteed obsolescence.
... err erata) rather than allowing students to resell their textbooks to other hardup students. IBM was the classic case of a monopolist that deliberately leased their gear to prevent a secondary market (google Amdahl/IBM marketing practices). If you read any standard MBA handbook, you realise the mind-boggling sales tactics at work which is naturally anthema to any self-respecting engineer who wants to look under the hood before buying ... alas engineers are not a majority of the sheeple population.
2 58&cid=66).
Basically the real value of any capital good is the price that the *next* buyer is willing to pay, ie the resale value. It doesn't matter if the house you bought at auction cost you $1M if the next person (given a free choice) is only willing to pay $500K. This disconnect between (sunk) cost and (future) value has been the downfall of many speculators (cough*dot con shares*cough).
Now for a good/service/title to accurately priced, there must be enough instances around so that potential future buyers can evaluate their utility and alternative factors of consumption. Thus for something like land, it could be zoned for residential, business, recreation, mixed, etc and different people would value each according to their needs. The secondary commercial leasing market allows much more flexilibility than just risking a big lump sum on a potentially unsuitable structure.
Another reason the secondary market arises when primary purchasors want liquidity, ie they find that they wish to redeploy capital (e.g. second mortgage for startup business). So even if you bought that house but can't make much use of it due to excessing work hours, you can lease it out. The biggest example is the share market which was originally a mechanism for the investment banks / underwriters to offload risk of being caught with an investment that was going downhill (after of course getting the best profits for themselves due to insider information, etc). Secondary markets are important in the sense that if they are large enough, they actually give very good price signals (cf efficient market hypothesis, random walks, etc).
Now how does this apply to the consumer services model? Basically the problem (from the business point of view) of computers (hardware) is that they can be used for anything and everything (software). Unfortunately the issue with any new technology is convincing users that they have a need for that junk
Now the incentives for destroying the secondary market is that people don't have a chance to properly value/price the alternative uses and that liquidity is removed increasing barriers to competitors, allowing the corporation to get away with a higher priced primary market (and the all important profits, options and return to shareholders). This can be seen most clearly in things like the technical textbook market where the Doctrine of First Sale means that publishers prefer coming out with new editions (and incompatibilities
Now applying this impromptu MBA lesson shows that from a business point of view, it is easier to sell future goods/services if you can eliminate the infrastructure for any competitors (the so-called deep entry moat). Hence convincing the customer that *the firm* owns the gear (despite using the retail system) and thus can exercise control over recall (naturally at their convenience). Obviously with electronic stuff this is a problem because people think of it as a manufactured *good* (aka appliance) rather than on-going *service* (maintenance). Hence the serious pricing problems since when you outright buy a good, you usually do so if the price (present value) is less than the value of all the future services you expect out of it. This disconnect is going to cause a lot of sustainability problems in the long-term as the accounting rules for capital items and software don't really reflect the real cost of services (software stability, interoperability quality, service level), just like inefficient market for medical pricing results in invisibile costs such as long waiting lists.
At least with open source you know exactly what you are getting up-front, the right that your contribution will be reflected and amplified in future iterations and refinements of the software. This is not the case with commercial providers that vary the terms of usage at their whim (see the Sexual Practices of Licenses at http://slashdot.org/comments.pl?sid=01/06/21/1810
LL
Part of the problem is that there are just way too many variants of licenses, whether OSS or otherwise. Part of the problem is that what browse-through license are is that they are actually a service level agreement e.g. I guarantee such a level of performance, provided you stay within the nominated activity envelope. Given the inherent complexities of computers plus the hair-splitting legal mumbo-jumbo, along with zilch consumer education and you basically get a system which treats license as no more than a disclaimer. With software, it is somewhat easier as you can split them into classes (GNU, BSD, Mozilla) with specific instances (Artistic, etc) with stated variations of terms (e.g. Alaladin is GNU-like but with temporal phase-shift, etc). Unbfortunately the internet world has not really standardised on an equivalent unless you count the MSFU (see the Sexual Practices of Licenses at http://slashdot.org/comments.pl?sid=01/06/21/18102 58&cid=66) which changes every week.
How well can you define a service? Using RDF you can probably specify the functional aspects but how do you nominate remedies for down-time, contacts for contingencies, independent auditing of claims (99% availabilty, etc). Ultimately you're probably see network access like a utility once the standard performance metrics are defined (MByte/sec up/down, latency (up/down) disruption distribution function, etc), and software quality is improved to such a degree that you can provide customer guarantees. But I don't see this happening until there's more localised competition rather than the big media titans duking it out.
LL
... apart from the horse racing, the pokies, the cricket matches, the .... etc, etc ...
There's already a sizeable grey economy despite the red tape of the Grab Snatch and Take (GST) tax. The average Australian doesn't take much notice of what goes on in Canberra and I seriously doubt whether a fiat dictate on internet gambling is going to make much of an impact. Gambling is really a stupidity tax on people who can't do maths so I suspect this latest triump of parliamentary wisdom is really due to the politicians hating competition.
LL
Correction ... the scientific publishing model is that research results be available at the marginal cost of distribution. Freelance writers who live purely on the basis of their commentary/articles do not fall into this category as their papers are not ads for their real output (reproduceable theoretical/experimental/computable results). As such they have a legitimate claim on their sole output.
The real problem is that there is too much rubbish out there (hey given copyrights of 90 years it's easier to recycle than to create) and it is hard to distinguish the cruft from the craft.
LL
The most recent calculations show that the annual cost to consumers and taxpayers of its 29 member countries' support for agriculture and horticulture amounts to US$361 billion. Such a large sum is difficult to grasp, but it is large enough to pay for a first class, round-the-world air ticket for each of the 56 million cows in the 29-member OECD?s dairy herd, and to give each cow a further US$1450 spending money for her stopovers in the US, Europe and Asia.
Thus while there may be laudable social objectives in keeping Scottish farmers around, there is a serious economic cost which makes you wonder why don't they just give the money directly to funding a Silicon Glen.
LL
There's some interesting speculative thoughts about computers and their impact on law ... one of the interesting claims found in <A HREF="http://www.erights.org/smart-contracts/index .html">www.erights.org</A> is that we are due for an era of prosperity (hah!) but only if we can bury copyright (as we know it). Given that the whole point of copyright is to restrict exchange (creating semi-artificial scarcity), the GNU approach of copy-left is already half a step along this path. Now I don't have the experience to judge whether the rest of the claims of utopia (capability computing, crypotographic protocols, smart contracts) will automatically lead to a brave GNU world but it is intriguing. Many of our concepts of "wealth" are likely to change over the next decade. In the pre-industrial age it was ownership of land, in industrial revolution it was access to resources (cough*colonies*cough), in the post-industrial economy a liquid capital market. Perhaps in the knowledge economy it will be computer-enforced contracts ... I promise not to create a BSOD when undertaking action XYZ. How much is such a promise worth? If you can attach a negative value to lost productivity, will Free software actually be cost-effective? This is going to be a really interesting area of research ... the ability to guarantee a software promise.
<P>
LL
... I would refer you to this article which discusses structured environments. Basically to be efficient at something, e.g. preparing a dish, we line up all the instruments in a particular order to help cue us when a particular task needs to be done. Extending this into a semi-spatial (if ESR can get the right relationship of in-on-up-etc representing the true connectivity between kernel modules) setting would help people orientate themselves and get to work faster (so you can get back to that QUAKE game). Just like we would put a letter by the door, or shopping list by the car keys, each cue triggers associated memories and reminds us of specific actions that need to be done. Sure you can have a linear check-list like the space shuttle pilots but anything computing-wise is so variable that a more flexible arrangement is desirable. I've been looking into something similar for the make processing for reproduceable documents and it is not as easy (or trivial) as people think (at least to get right). For example, color assignment ... do you map this property to time-of-last-modification (ie heat colors) or to likely hazard (red=stop, orange=hazard, yellow=caution, etc). If you dig far enough, you eventually realise it is actually a text variant of scientific visualisation but in the qualitative domain. There are a number of theories on how you allocate the properties based on human cognitive functions (see Lloyd w.r.t. OpenDX and fiber bundles) ... but there are a lot of issues remaining such as what works for 5-6 objects doesn't work for thousands.
Like all ideas, let the users decide and if it works, it will be included into the meme-pool. Imagine TuX, the penguin avatar trundling around kernel space looking for fish or and fixing security leaks. At the worse, it will provide a few minutes of amusement.
LL
Generally the principles under which students contribute work falls into the
... I presume the code is attached as an appendix
1) copyright to thesis belongs to student (you write it, you own it)
2) any restrictions (e.g. on publication) are time0limited (e.g. Pharmaceutical companies usually don't hold up publication more than 6 months, 18months is very extreme)
3) any examiner may be required to provide acknowledgement of confidentiality
The bottom line is that a supervisor, if they are smart, sets out ahead of time a written agreement stating exactly what is expected as participation of a project. If you have moral qualms about fueling the military industrial complex, that's the time to bow out gracefully and ask to be reassigned.
OTOH, universities can ask for the right to an outcome, for example rights to use that material for teaching purposes. The nasty situation is when a tutor has leeway in slection of assignments and gains a personal financial benefit (e.g. external party consulting) but universities usually have groundrules on this (e.g. no more than 20% of time, or waiver on consulting outcomes). Obviously some universities are more careful than others and there have been some stories about early career researchers being basically fleeced of their efforts. Students, because they tend to be doing bulk/mass assignments are usually working a known problem in which case there is no real intellectual advances and it would be very hard to assert IP rights except as a performance (rearrange widgets using this script).
As always, reading the rules before signing on the dotted line as there's always the small print to stumble over.
LL
Currently there are a number of XML billing initiatives such as XMLPay, VisaXML, etc... How do you see these trends influencing the design and development of GNUcash ... in particular do you think that your users are only looking for a simple view/format/verify client (think IMAP server + XML extensions) while GNUcash moves towards being the agent of record (aka a specialised ASP?) Of you do see GNUcash as a drop-in replacement for existing monolithic accounting packages? In summary, given that the market for electronic businesses and purchases (greater use of international currency, direct attachment to electronic settlement, near real-time risk management) is going to change, how do you think GNUcash should respond?
LL
I'm surprised more people don't use it ... it's really useful for managing multiple versions of libraries and different applications which depend on them. Unforuntately when you start hitting C++ class interdependencies or zillions of versions it starts being tricky ... which is where namespace managment and scoping rules become more important.
LL
And the DIY home decorating industry has been a very profitable source of income for those handymen able to come in and fix up your mess ... unfortunately the jumble of components nad scripts is underneath the hood hidden behind a gee-whiz veneer for too many people to care.
Untile people appreicate professionalism and artistry, expect the marketing hype to dominate.
LL