... before we resort to Taylorism (the guy who invented productivity measurement and thus can be blamed for filling out ridiculous detailed task-sheets), I'd like to point out that experts and amatures have different mechanisms when interacting with computers. Studies from cognitive psychology shows that we are constantly reordering our environment much like a cook (assuming you're not the nukem couch potato) sequences the tools/bowls in sequence before starting work. We put umbrella stands near doors, grocery lists on fridges, and address books on phones. Basically we impose some sort of spatial-temporal orderordering to cut down the performance of the task to a series of unreflective actions, triggered by environmental cues. Hence the desire for people (especially those visually oriented) to customise their too frequently sub-optimal desktop (hands up those who use multiple deskviews!). On the other hand, experts have been shown to encapsulate knowledge with a much higher set of rules and create powerful keyboard expressions (think regular expressions rather than GUI). The prefer a deep, consistent and even orthogonal ruleset (multiple modes such as shift=extend, alt=reverse, etc) and woe betide anywho dares take away their favorite editor. The problem is that too many applications have evolved independently which causes cognitive friction (the Mac is probably the best at ensuring conformity) and skins are not the real answer. In fact a recent talk a Linux Conference Australia pointed out the absurbity of putting the full-screen next to the close window button. The problem is that we just don't pay enough attention to useability, going for fweeping creaturism (see FLTK) instead, and window dressing will not solve the problem. Computers are complicated and expecting an internet appliance without some significant simplification (aka labotomisation) is like hoping a high-school kid to understand nuclear physics (see chain reaction go boom).
Productivity would be enhanced by letting a free-market reign by individual choice of development tools, independent (no more benchcrafting!) evaluation of maintenance costs (the biggest part of any software budget) and a better understanding of cognitive limitations (including why marketing people should not be given release control of things like clippy).
... as versus the Craft of Programming. Perhaps one interpretation of the situation is that people view the world in different ways. RMS was incensed that a closed system prevented modification whereas the corporation probably thought of it was preserving their investment (by limiting competitors). Much as in the same way that you can view a brick as either a contructive building block or alternatively as a radical social statement (by chucking through windows), you can use (or abuse) code in similar ways. If you take the ESR viewpoint that true hackerdom (creative aspect) is scratching an itch, then it conflicts with the programming as purchased service (business logic embodied in codified format). Without some form of control (via a form of pricing mechanism) in limiting free-riders, the second form is economically valueless which sorta kills the need for programmers (and their high salaries). The crafting of such systems now drives an increasing segment of the economy which means less tolerance for purely artistic endeavours (e.g. emulators for the sake of showing off which in turns shake the vertically integrated hardware-software model).
Free resources (as stated by Lessig) are effectively entities which have yet to be subscribed an economic value through elaborate technical transformations. Oil was basically a inflamable seepage from the ground until industrialists figured out a way to convert it into kerosene (which saved the whales from being hunted to death) and subsequent catalytic conversion into many synthetic downstream products. Similarly many bits of code have yet to find their niche and Lessig makes the valid point that if their original creators don't have much of a clue, then excessive control means they will never be adapted for something more useful (oak->java, tcl->expect, etc).
... hmmm is there an off-line forum where these items are discussed? I'm not an assembly level hacker but I would like to explore some ideas on how you can retrofit non-linear memory systems into a von-neumann machine. For example, if you have a FPGA sitting in SRAM slot (I need to dig out that HK research paper), then you could have a deliberately dirty page as a guard... then when you get that cache miss... try to analyse the last set of instructions and work out if a different memory access pattern (ie alter the algorithm residing on the FPGA) might be better.
... there's already some interesting ideas coming out of the VR3 project... offhand I can think of that Snow ABI which considers building apps in a different way to be more memory thrifty... unlike a PC with virtual memory, a PDA is severely constrained with no guarantee that a wireless connection will be available. Some of the ideas could be extrapolated... for example, if you have a transmeta chip, could the ABI refactor themselves in memory (ie reorder libraries to drop non-used portions?) What about mechanisms to detect dead code or where the memory/code hierarchy changes (think reconfigurable chips hibernating in spare memory slots as one HK uni research group published).
I mean, we evolved from the dwarf binary format to elf... perhaps we need to consider next generation hardware advances to.... ummm... create the hobbit ABI. Think reconfigurable. think non-linear memory, think small embedded devices that can join together in a single complex task, think auto-optimisers/refactors a la JIT.
I think this is the begining of the trend of the hardening of the biological sciences (as in becoming infused with mathematics... and compsci algorithms in this field are essentially discrete maths). Unfortunately that means many people seeking easy course credits are going to be hit with a shock to the programming/maths skills as biology realigns to more complex techniques in shifting from wet to dry labs.
For people wishing to advance professionally, O'Reilly is now introducing their Bioinformatics Technology Conference (http://conferences.oreilly.com/biocon/) in fact at this moment in Arizona. Also there is the Int. Soc. for Computational Biology (http://www.iscb.org) which organise regular bioinformatics meetings. Open-source tools are a regular part of this.
... given the fact that molecular biology is becoming capital intensive (look at the multi-million dollar synchrotrons, protein-chips, sequencers, etc) the public system just cannot fund everyone. The acceptance of private funding means constraints as companies are not in the charity business. If you look at the impact of the medical system, you'd notice similar structural shifts as GPs are merged into medical centres clustered around MRI/X-ray/capital machines. A similar activity is happening in academia with greater infrastructure forcing consolidation into smaller clusters (e.g. take a look at the San Diego biotech cluster). The explanation is simple, better equipment equals higher throughput and thus productivity. Given that bioscience is basically a search through a multi-dimensional space of all combinations of proteins, you can see why the group that covers more area has a greater probably of discovering something interesting.
On the other hand, public science has the implicit assumption of peer-review... even industry recognise that they cannot delay publications more than a certain amount (6-9 months??). Hence the excuse that they are preserving the publication track record of their apprentices is a bit of a cop-out. Basically they are saying yes but wait x months for some underling to publish. The problem comes in the rush to produce results, people ignore the fact that they have to be *reproduceable* results (otherwise by definition is it not science). There have been stories of groups losing original material so the only claim to fame is by squinting at a graph and hoping that the parameters they choose for their analytical process (trust me... any mathematical analysis has zillions of twiddle factors) are not a fluke.
So either we go back to the slow but certain government funded research or accept that private incentives will create only temporary information asymmetry. The private market for knowledge is rather unformed at the moment as there are no clear guidelines as to what are acceptable practicese.
... in restoring competition? One of the assumptions behind anti-trust is that business practices valid for small agressive companies are not legal for monopolies. But imposing behaviour constraints seems to be like pissing in the wind... basically if the officers are not legally responsible for their corporate culture (e.g. FUD sales tactics) then there is no incentive to alter their mindset.
Perhaps its it time to consider identifying where the competitive pressure has failed and why it is os. Competition refers to more than the guy trying to do unto thee what you're trying to do unto them. It also includes substitute goods/services, threat of entry, pricing power of suppliers, and pricing power of buyers. Here we see specific tactics that should be considered as hindering competition
- denial of interoperability to avoid subsitution
- rapid change of APIs to impose high conversion costs to smaller entrants (cough.NET cough)
- extend and embrace to extinguish markets created by more gung-ho startups
- distribution contracts that forbid secondary sales
The last component is a subtle one as it disguises a depreciating service (license) as a phsycial good (CPU). If people were free to sell their MS software license (substitute Linux instead say) on their OEM box then the market price for MS software would be established by the free market which would estimate the half-life of a specific version of their OS/app. Unfortunately a free market does not serve the needs of a monopolist as people can then have an alternative to compare the cost of other features (such as lack of security).
So what can the legal system do to help restore competition to the software market? I would recommend requirements that licenses/terms of usage be written in clear language rather than legalese, that any promises (spam-free mail) be backed up by some form of performance bond, and that termination/opt-out clauses be subject to scrutiny by fair trading groups.
Ahhh... the good ol' days of font wars between Adobe and MS. I believe that one unintended side-effect (at least I hope so otherwise it would indicate some really devious business thinking) was that by making acroread ubiquitous, they effectively controlled the pathway to the printer. All their publishing products can suppress printing/cut&pasting and those rules are enforced by acroread. Hence by giving away their viewers, they effectively created a distribution channel for PDF files that required comparatively expensive authoriing tools... it is a heck of lot easier selling 10,000 $500 packages than trying to flog 1,000,000 $5 viewers.
The puzzling thing is why they don't produce unix versions of their latest acroread versions. This opens up a gap in their product range which open-source viewers could potentially evolve to bypass any protection scheme.
And combine this with a template generator such as Lout (http://www.ptc.spbu.ru/~uwe/lout/) for the permanent stuff and you effectively get documentation on demand. Users will type in questions,queries,quotes (of source code) and once the issue is resolved (think of it as a dynamic form of literate programming) can be rendered to hard copy for permanent archiving/case-study/how2, etc...
What is the most efficient market for selling stuff... ?
If you study the nobel prize winner of a few years ago, you'd discover that dutch auctions can be theoretically proven to be the most efficient price discovery process.
Guess who's implementing auctions in a massive way?
The same group that's expanding from collectibles to cars to sports gear to CDs... (and I'm not talking Amazon here)
Already large-scale companies are dumping overstocked or out-dated goods on eBay... there's no reason why they can't commoditise other mass market non-perishable bundles (think personal care kits, think entire kitchenware, think car personalisation). The biggest barrier to adoption has always been social... the better micetrap doesn't work beyond a certain point. It's all about distribution, reliability, after-sales service, etc...
OK that's the long-term killer for ads. Now what about service organisations (ie offering something other than tangible goods). The service is about finding someone to do something that you can't do yourself... this is a directory problem. Already you see personal advertising in newspapers disappearing due to specialised employment agencies, help-desks,... I see this trend also happening to banner-ads. People will go to specific trading sites which persists reputations rather than wanting to be inundated with services they don't need at the moment.
So... the other ads that are left are the branding which are image/style based... frankly those wish to be associated with an experience and reminding the user that you're responsible for a over-limit bandwidth bill and a waste of time is not good karma. Coke and Pepsi sponsor rock concerts not sport statistics.
In summary, unless there are some fundamental problems with my observations, I would say that ads as we know them (banner, etc) will become ineffective due to going under the personal threshold of normal perception. Go to a rural place and you'd really notice the *ABSENCE* of billboards. Instead you will be mor eproduct placements... movie trailers which use products that normal people can identify with and feel part of the crowd. I can talk about social alienation in cities which lead people to identify with their professional peers rather than neighbours but this is a geek-site not socio-economic trend analysis.
In summary, IMHO pure ad-driven renue models will fail. It might have worked for the radio-broadcasting industry which requires continuous listening but unless something radical happens to social perception of the internet, the ability to jump-click outside a walled domain, and the fundamental cost-structure (ads=bandwidth=costs) I don't see them being viable.
Of course the 64 million dollar question is what is a viable business model which all the VCs would give their souls (or unmortgaged remainder thereof) to discover.
... that was benchmarketed [1] on/. a while ago... (I vaguely recall a lot of noise comparing features vs iPAQ). Are the apps cross-source-comptible? I mean if the claim is that open-source creates innovation then by theory we should see a lot of competitive entries into the PDA market as the cost of entry is so much lower (create a compiler-farm and voila instant apps).
LL
[1] Some people are complaining about acronyms/jargon so I'll just clarify for newcomers. Benchmarks are side-by-side comparisons of features/functions/performance using quantifiable tests/numbers. Benchmarketing is the selective choice of a subset for simplistic bragging rights (e.g. spec-rating) which can be taken to extremes by benchcrafting (google on Mindcraft and Linux) which is an artificial benchmark designed to distort public opinion.
I believe certain countries specifically protect the livelihood of employees who call attention to illegal, improper or immoral activities? Often there is a public perception shift (e.g. slavery is now considered abhorant) but the initial persecution for people speaking out can be quite intimidating. Given that certain companies (again hersay) have a policy of not committing any sensitive information to permanent records like email, the only way is to encourage individuals to step forward and speak the "truth". Creating legal precedents which hinder the principle does not support the public good. Is shareholder value so important that you have to silence critics (both internal and external)?
Given the lack of corporate-employee loyalty, shouldn't prospective workers have a chance to study both points of view? Trial by litigation (as the modern form of trial by combat) seems a rather inefficient way of resolving such angst.
... would the USENET archives qualify? As compared with say MS development network which is the equivalent knowledgebase? Hate, love or indifferent, you cannot deny that MS has had a major influence on the growth of the PC sector and a large part of this success is their fanatical devotion to their developers (please no jokes about if you got them by the balls, their wallets will follow). USENET is a nice snapshot but is it something purposeful?
I was just musing the other day about what would be the 7 wonders of the digital world... personally I would consider the first choice to be the Guttenburg Project which is low-key but represents the unflinching efforts of many many experts and volunteers. Given the dissipation of the social contract w.r.t. modern copyright laws, the scope and vision of the originators can only be admired.
Sure, GNU/Linux could be nominated but I'm a little ambivient about it as the impact is mainly social (due to GPL and the contributors' belief in libre software). As a technical piece of work, is it on the same relative scale as the ancient wonders were in their heyday? We are talking global uniqueness, recognised by a wide population segment, and something difficult to duplicate here.
... I'm just wondering whether the right social factors are in place for a broad-based uptake of XML. The real advance of the web IMHO is not w3c/AOL/etc but the IETF with its RFCs... this allows companies, consortiums and even consumers to describe a protocol, and let the free market compete to provide implementations. Witness BXXP (or now renamed beepcore) which has Java, TCL, Ruby, etc libraries as well as C/C++ reference.
Now look at the same issue with DTDs. The fundamental constraints are that they are industry specific, tied to specific domain knowledge and it takes a lot of to-ing and fro-ing to come to consensus. Once a DTD is officially published, then modifying/varying it can be difficult. For example, I wanted to modify the XBEL (XML bookmark exchange) to add in a deprecate-by date field but it would be incompatible with all the existing implementations. Trying to work with large industry standards would be even more cumbersome as there are so many entrenched interests (just look at the proliferation of billing-based XML). How do large groups resolve the negotiations and compatibility issues (and talking about sub-schemas is another can-of-worms), not to mention the ontological definitions of any specialised language corpus (witness the biological community trying to define items in a rapidly moving field). XML might be the currency of computer originated messages but who guarantees its inter-convertibility (present and future)? For a XML based application to work seamlessly, cooperative structures that span multiple groups probably need to be established. While corporates would be more than willing to set forward, general mistrust based on past misbehaviour does not auger well for wide-spread uptake of the technology. How do you know whether that biz-talk or TLA of-the-day is not secretly sending out sensitive information? And if technical guys are dubious about security, how can you expect consumers to embrace something which is beyond their understanding and has little immediate benefits (XML IMHO is more useful for computer-computer transactions than computer-human).
The real revolution with XML will be social, not technological.
The process of taking a basic patent and then lodging in the patent office pipeline a lot of iterative "improvements" can cause problems. In legal parlance, these "submarine patents" are kept under wraps until products they cover come to market, and then create a legal minefield as the patent holder seeks licensing.
I would note that perhaps we should set specific hurdles for software patents... at a bare minimum it should be demonstrate that there is a period of continuous (or near continuous) improvement/development/implementation. Ie, use it or lose it. If the idea is so great and thus worth the expense of patenting it, then why are they sitting on it? Waiting for someone else to do the hard work of debugging and actually convincing the rest of the market to adopt it (the hardest problem in todays software world) and then taking a slice of the action is nothing more than parasitic behaviour.
Perhaps another reform should be the more innovative a software technique is, then the longer the period where it could be held by the originator.
... which is being proposed for outback Australia. According to some of the designs, a log-spiral configuration could extend more than 2000 kilometres (!!) which pretty much spans the whole continent. About 40% of the lens will be concentrated in a single kilometre grid. The interesting feature is that they hope this will be a software telescope with insane amounts of DSPs to help correlate the different frequency radio signals and generate composite snapshots. Browse around that site for some interesting reads!
When the behaviour is simple combinataion of philic/trophic responses then the analog model works due to the relatively gross scale of operation (this is intended for kids). Given a small enough sampling speed, this analog behaviour can be mimicked by digital means (effectively the discretisation and solution of PDEs). The advantage of a digital model is that (if programmed correctly) is based on a Turing machine which is much much more flexible. That means you can create new inputs/responses rather than the set of prewired functions.
Man... sounds like the electreonic equivalent of a retrovirus (for those without a biological background, these are virus-like critters that modify small segments of a host cell's DNA, usually by inserting themselves at inconventient spots). These are often associated with STDs and tend to be very difficult to eliminate.
You have to understand the corporate mentality... all cable operators (which given their telco background is probably their CTO's wet dream) have their share price valued at $XXX per customer which reflects the long-term monetary value they are able to "extract". Now given the competitive landscape of many alternatives (satellite, wireless, 802.11 etc), the only way they can "increase value" (aka screw the customer) is to charge gatekeeper fees... hence the gratuitous crack of your system, reversionist defacing of your bookmarks (to their premium advertisers of course) and rebranding of your system (ie treat you as a free billboard for any visitors). The cost might be zero (for the CD) but the value for you as an individual is definitely negative... I'm surprised that this can actually get through the equivalent of better business bureaus because it appears to be bordering on the verge of deceptive advertising (scaring you of deletorious effects if you don't use it).
I certainly hope someone from the consumer report is taking note of these type of activities....
The golden rule used to be "whoever has the gold makes the rule" but I would observe from current machinations the golden collorary "he who writes the rules, defines the gold".
The reason... think what the internet does.. every single piece of information whether written in the past or immediate future (think trailers) is now immediately available. It's like a thirsty man in a desert being swept away by a flash flood. All the historical economic models based on a content/distribution model is now completely invalidated. Historically media studios could release stuff at different price/time points (movies, videos, cable reruns, etc) with the nice kicker that a popular franchise can be remastered with relatively little marginal cost.
Now suddenly anyone (with a modicum of hacker skill) can bypass their time/space-controls (cough DVD-region-coding), the TiVo is just one small example. Suddenly all their media libraries is implicity devalued as they can't withdraw "obsolescent" titles. The First Sale doctrine means that anyone can resell their "original" copy which creates competition for their newest overhyped gee-whiz production. Hence their incentive, nay long-term economic survival, in pushing Digital Rights Managment (aka service selectivity/variability) by stealth (submarine legislation) or by wealth (trial by litigation).
Of course, they don't always have much of a clue (cough*CueCat*cough) so they have to rely on the tech experts to provide them with the tools to control/segment the entertainment market. Which means that unless you have a tech department under your belt like AOL, they are held over the barrel by the likes of Microsoft who have their own ambitions of being the broadband toll-keepers.
Economics alway always been about scarcity (whoever dies with the biggest toys wins) but the internet inverts all that into a surplus. The "gift culture" that ESR mentions is thus anathema to any self-respecting aspiring monopolist as infinite replication/distribution of information-based products limits their market of gullible fools.
It will be an interesting decade as all these economic forces resolve themselves.
The concept is that the market price should be close to what people are willing to pay (fair value). Airlines specifically market fisrt-class as "better". Just like they put on extras onto DVDs to make them more attractive than video tapes. Now if game publishing houses are too stupid to deploy the same tactics (e.g. software easter eggs for English market, etc) then they are not addressing the equity problem (why customers get disturbed with paying varying prices). We have "social" discounts such as veteran benefits, student fares, etc... if you extrapolate this on a global basis, why should you feel insulted if a company decides to cross-subsides countries with lower GDP? You see this selectivity everywhere, from having chip manufacturers test MHz from the same batch and down-grade those with unstable charateristics to having software/firmware governers that step-down the performance (a favorite for IBM mainframes).
The concept of discriminatilon ranks with many people, but how do you separate those willing to pay for convience/priority even though the product/service is exactly the same? There ain't no such thing as a free lunch.
LL
The cost of leisure ...
on
Sony vs Modchips
·
· Score: 5, Interesting
... the problem with the entertainment industry is that it is often tied to disposable income (if you don't watch TV you're not going to die regardless of what kids think). As such there is serious competition for our attention ranging from walking in parks (NY muggings excepted) to window-shopping in malls (a legitimate form of entertainment as shown by theme parks taking this philosophy to extremes such as Disney). Groups such as Sony have to come up with ever more inventive ways of parting you from your money... err catching your attention and delivering amusement. This problem is exasperated by the fact that different countries value leisure differently. A third world sweat shop worker just simply has better things to buy (like education for their kids) than light entertainment. Hence global companies cannot charge the same price for the same item (CD) in different countries. Hence their desire for market segmentation tools such as multi-zoning.
Now is this considered fair? Places like Australia don't believe so as their competitive watchdog recently ruled that multi-zoing was anti-competitive as it hindered parallel importing (is source CD from other countries). On the other hand companies argue that it is like passenger classes in planes, first-class still get there at the same time as cattle-class but pay significantly more. Many companies (esp software/pharmaceuticals) use the high prices of their products in 1st world countries to cross-subsidise less developed markets. Given the increasing connectivity of world trade this is becoming increasingly difficult.
Computers with digital rights management (aka service variability) is one mechanism to enforce this market segmentation, especially if it can be enforced through fixed/controlled end-points (cough*Xbox*cough). This is why companies hate mod-chipers and related products (satellite decodes, overclockers, etc) as it allows individuals to exploit the artificial price differential between 1st/3rd world pricing strategies. The end-result is a technological arms race (embedded ids, self-destruct, registrations, etc) in order to maintain this separation between high-margin customers and more marginal users. A person collecting warez for bragging rights is *NOT* willing to pay the same recommended price as someone looking to kill time by renting an evening game.
Anyone who thinks a company is going to destroy their global economic model just to please a small (but vocal) group of (from their point of view) "parasites". A large enough business entity can tolerate a small percentage of free-riders but is likely to come down hard on any systematic or organised threats to their business provided they can distance themselves from any media-fallout (cough*Adobe*cough)... up to the point of lobbying legislators (cough*DCMA*cough) to exterminate what they view as inappropriate economic conduct.
Fortunately the free market (e.g. open source movement) has a little influence in moderating the extreme behaviour of the more pervasive global corporations.
.... reminds me of a regional finance company that specialises in infrastructure funding (things like ring-roads, traffic bypasses, fibre, etc). They built a high-speed traffic bypass asnd started charging tolls. However, they found out residents were still using local roads so they did a deal with the state authority repsonsible for public roads to seal them off.
The point is that they did the conomic analysis and showed that there was a net savings in petrol consumption and driver's time. But in order for them to pay for building the new infrastructure, they had to convince motorists that there was no longer a free-ride. The problem is that motorists only saw the daily toll charges and not the weekly savings in petrol/time. Not to mention that very little advance warning was given to changing the road access. You can probably guess the PR fallout resulting from this:-).
Sooner or later a similar scenario will happen with communications networks as they reach the limits of scaiability and in order to transition to a more efficient/lower cost/reduced maintenance system, they need to convince people of the benefits of switching This usually requires changing their usage patterns. Unfortunately aggressive telco upstarts don't always have the diplomatic skills to address customer's expectations. Pricing is a particularly sensitive point as there may be incredible customer acquisition costs or hidden cross-subsidies that distort the cost structure.
There has probably been some over-investments in network infrastructure that the current recession is revealing. As Warrne Buffett says, it's only when the tide goes out that you see who's swimming naked. Companies that pass the buck (literally) for their corporate mistakes are going to have a hard time keeping onto their customer base and will have to either swallow the losses (and shock horror forfeit the CEO bonuses/options) or else try and merge to gain monopoly pricing power and justify their executive packages. While some people may decry the double-sided nature of telcos and Wall Street, hopefully the survivors will be more sensitive to their users's needs.
... and biofeedback controllers are now entering commercialisation phase... take a look at BioForce which was voted the best gaming technology in the LA Electronic Entertainment Expo. This uses low intensity electric pulses to provide external stimulii... just the thing for convincing Quake Gods with dellusions of invulnerability to switch to Counterstrike instead:-). This recently won an Entrepreneur Prize with the specific aim of commercialising it so expect to see some shocking developments in the haptics arena in the not so distant future.
... as to customer service as Cringely points out. The skills for financial engineering are quite different from that of altering purchasing habits (which the dot cons were promising to do). The traditional success of the Silicon Valley guys were in funding hardware that took advantage of Moore's Law. I heard stories that the East Coast bankers got jealous of the success and decided to muscle in. Unfortunately, they lacked the experience (discipline?) of their West Coast peers and saturated the market (there is a shortage of *GOOD* ideas that are also *NET CASHFLOW POSITIVE*) and consequently a lot of dumb money.
What makes people return to a place for their next purchase? Whether it is servicing a need, comfort for affliction or just plain satisfaction, it takes a special clarity of thought to get a good feel for what people want (and are willing to pay) and from there build up a sustainable operation.
VC's probably overstate their talents but at least they're willing to risk other people's money in ways that banks can't or won't (and if anyone who started up a business on a credit card can attest it is nerve-wracking).
Now if people could only come up with some clever Open Source Funding models.
... there's been some rather interesting work being done on analysing markets based on assymetric and incomplete information (in fact I beleive this years Nobel prize in economics was on precise that... selling used cars as lemons). The problem with software (especially closed) is that there is zero information about the quality of a *new* piece of software. Until you get to use it (and discover + avoid things which BSOD it), it basically has zero utility.
So how do software companies actually generate cash-flow? They have to "prove" their fitness in other ways (ie signal to the potential buyers that their code solves your future needs). Whether it is giving out a first freebie (QUAKE) or carpet bombing the user base (AOL), somehow the vendor needs to demonstrate that they provide something worth the "value" others place on it.
Now, the curious thing is should open source be such a signal? It might not be the "monopoly" in the traditional sense, but more a meritopoly... ie through the school of hard knocks, it has demonstrated that it is the "best" at a particular niche (given economic constraints) and which case it is better non competiting against it. Look at stuff such as Aladin ghostscript, or the publishing templates DocBook.
To actually get bigger than a sustainable niche, you basically have to bullshit like crazy... (cough*intel*cough with claims that its high-speed chips were essential for the internet ignoring the minor issue of bandwidth/latency). Unfotunately marketing is a sunk cost... once you've spent your advertising budget, you can't recover it. Hence for firms who suddenly find they've overspent, the only way they can make back their money is to either eliminate competitors or else or cut back on the quality/support. And guess what, if you're the only solution, then you automagically create a self-fufilling prophency that your product is the "best" in its category and therefore all the technology followers (ie conservative foggies who don't trust anything less that v3.0) are forced to purchase it irregardless of the absolute quality (stability, security, mtbf, etc).
However, with open source this approach doesn't work as you cannot eliminate a competitor that cannot be bought off. As such, a rational player would instantly quit the market segment as soon as they can tell the open-source version exceeds their market share. Hence by definition, OpenSource becomes the new monopoly...
Slashdot Mirror
... before we resort to Taylorism (the guy who invented productivity measurement and thus can be blamed for filling out ridiculous detailed task-sheets), I'd like to point out that experts and amatures have different mechanisms when interacting with computers. Studies from cognitive psychology shows that we are constantly reordering our environment much like a cook (assuming you're not the nukem couch potato) sequences the tools/bowls in sequence before starting work. We put umbrella stands near doors, grocery lists on fridges, and address books on phones. Basically we impose some sort of spatial-temporal orderordering to cut down the
performance of the task to a series of unreflective actions, triggered by environmental cues. Hence the desire for people (especially those visually oriented) to customise their too frequently sub-optimal desktop (hands up those who use multiple deskviews!). On the other hand, experts have been shown to encapsulate knowledge with a much higher set of rules and create powerful keyboard expressions (think regular expressions rather than GUI). The prefer a deep, consistent and even orthogonal ruleset (multiple modes such as shift=extend, alt=reverse, etc) and woe betide anywho dares take away their favorite editor. The problem is that too many applications have evolved independently which causes cognitive friction (the Mac is probably the best at ensuring conformity) and skins are not the real answer. In fact a recent talk a Linux Conference Australia pointed out the absurbity of putting the full-screen next to the close window button. The problem is that we just don't pay enough attention to useability, going for fweeping creaturism (see FLTK) instead, and window dressing will not solve the problem. Computers are complicated and expecting an internet appliance without some significant simplification (aka labotomisation) is like hoping a high-school kid to understand nuclear physics (see chain reaction go boom).
Productivity would be enhanced by letting a free-market reign by individual choice of development tools, independent (no more benchcrafting!) evaluation of maintenance costs (the biggest part of any software budget) and a better understanding of cognitive limitations (including why marketing people should not be given release control of things like clippy).
LL
... as versus the Craft of Programming. Perhaps one interpretation of the situation is that people view the world in different ways. RMS was incensed that a closed system prevented modification whereas the corporation probably thought of it was preserving their investment (by limiting competitors). Much as in the same way that you can view a brick as either a contructive building block or alternatively as a radical social statement (by chucking through windows), you can use (or abuse) code in similar ways. If you take the ESR viewpoint that true hackerdom (creative aspect) is scratching an itch, then it conflicts with the programming as purchased service (business logic embodied in codified format). Without some form of control (via a form of pricing mechanism) in limiting free-riders, the second form is economically valueless which sorta kills the need for programmers (and their high salaries). The crafting of such systems now drives an increasing segment of the economy which means less tolerance for purely artistic endeavours (e.g. emulators for the sake of showing off which in turns shake the vertically integrated hardware-software model).
Free resources (as stated by Lessig) are effectively entities which have yet to be subscribed an economic value through elaborate technical transformations. Oil was basically a inflamable seepage from the ground until industrialists figured out a way to convert it into kerosene (which saved the whales from being hunted to death) and subsequent catalytic conversion into many synthetic downstream products. Similarly many bits of code have yet to find their niche and Lessig makes the valid point that if their original creators don't have much of a clue, then excessive control means they will never be adapted for something more useful (oak->java, tcl->expect, etc).
LL
LL
... there's already some interesting ideas coming out of the VR3 project ... offhand I can think of that Snow ABI which considers building apps in a different way to be more memory thrifty ... unlike a PC with virtual memory, a PDA is severely constrained with no guarantee that a wireless connection will be available. Some of the ideas could be extrapolated ... for example, if you have a transmeta chip, could the ABI refactor themselves in memory (ie reorder libraries to drop non-used portions?) What about mechanisms to detect dead code or where the memory/code hierarchy changes (think reconfigurable chips hibernating in spare memory slots as one HK uni research group published).
... perhaps we need to consider next generation hardware advances to .... ummm ... create the hobbit ABI. Think reconfigurable. think non-linear memory, think small embedded devices that can join together in a single complex task, think auto-optimisers/refactors a la JIT.
I mean, we evolved from the dwarf binary format to elf
LL
I think this is the begining of the trend of the hardening of the biological sciences (as in becoming infused with mathematics ... and compsci algorithms in this field are essentially discrete maths). Unfortunately that means many people seeking easy course credits are going to be hit with a shock to the programming/maths skills as biology realigns to more complex techniques in shifting from wet to dry labs.
For people wishing to advance professionally, O'Reilly is now introducing their Bioinformatics Technology Conference (http://conferences.oreilly.com/biocon/) in fact at this moment in Arizona. Also there is the Int. Soc. for Computational Biology (http://www.iscb.org) which organise regular bioinformatics meetings. Open-source tools are a regular part of this.
LL
... given the fact that molecular biology is becoming capital intensive (look at the multi-million dollar synchrotrons, protein-chips, sequencers, etc) the public system just cannot fund everyone. The acceptance of private funding means constraints as companies are not in the charity business. If you look at the impact of the medical system, you'd notice similar structural shifts as GPs are merged into medical centres clustered around MRI/X-ray/capital machines. A similar activity is happening in academia with greater infrastructure forcing consolidation into smaller clusters (e.g. take a look at the San Diego biotech cluster). The explanation is simple, better equipment equals higher throughput and thus productivity. Given that bioscience is basically a search through a multi-dimensional space of all combinations of proteins, you can see why the group that covers more area has a greater probably of discovering something interesting.
... even industry recognise that they cannot delay publications more than a certain amount (6-9 months??). Hence the excuse that they are preserving the publication track record of their apprentices is a bit of a cop-out. Basically they are saying yes but wait x months for some underling to publish. The problem comes in the rush to produce results, people ignore the fact that they have to be *reproduceable* results (otherwise by definition is it not science). There have been stories of groups losing original material so the only claim to fame is by squinting at a graph and hoping that the parameters they choose for their analytical process (trust me ... any mathematical analysis has zillions of twiddle factors) are not a fluke.
On the other hand, public science has the implicit assumption of peer-review
So either we go back to the slow but certain government funded research or accept that private incentives will create only temporary information asymmetry. The private market for knowledge is rather unformed at the moment as there are no clear guidelines as to what are acceptable practicese.
LL
... in restoring competition? One of the assumptions behind anti-trust is that business practices valid for small agressive companies are not legal for monopolies. But imposing behaviour constraints seems to be like pissing in the wind ... basically if the officers are not legally responsible for their corporate culture (e.g. FUD sales tactics) then there is no incentive to alter their mindset.
.NET cough)
Perhaps its it time to consider identifying where the competitive pressure has failed and why it is os. Competition refers to more than the guy trying to do unto thee what you're trying to do unto them. It also includes substitute goods/services, threat of entry, pricing power of suppliers, and pricing power of buyers. Here we see specific tactics that should be considered as hindering competition
- denial of interoperability to avoid subsitution
- rapid change of APIs to impose high conversion costs to smaller entrants (cough
- extend and embrace to extinguish markets created by more gung-ho startups
- distribution contracts that forbid secondary sales
The last component is a subtle one as it disguises a depreciating service (license) as a phsycial good (CPU). If people were free to sell their MS software license (substitute Linux instead say) on their OEM box then the market price for MS software would be established by the free market which would estimate the half-life of a specific version of their OS/app. Unfortunately a free market does not serve the needs of a monopolist as people can then have an alternative to compare the cost of other features (such as lack of security).
So what can the legal system do to help restore competition to the software market? I would recommend requirements that licenses/terms of usage be written in clear language rather than legalese, that any promises (spam-free mail) be backed up by some form of performance bond, and that termination/opt-out clauses be subject to scrutiny by fair trading groups.
LL
Ahhh ... the good ol' days of font wars between Adobe and MS. I believe that one unintended side-effect (at least I hope so otherwise it would indicate some really devious business thinking) was that by making acroread ubiquitous, they effectively controlled the pathway to the printer. All their publishing products can suppress printing/cut&pasting and those rules are enforced by acroread. Hence by giving away their viewers, they effectively created a distribution channel for PDF files that required comparatively expensive authoriing tools ... it is a heck of lot easier selling 10,000 $500 packages than trying to flog 1,000,000 $5 viewers.
The puzzling thing is why they don't produce unix versions of their latest acroread versions. This opens up a gap in their product range which open-source viewers could potentially evolve to bypass any protection scheme.
LL
And combine this with a template generator such as Lout (http://www.ptc.spbu.ru/~uwe/lout/) for the permanent stuff and you effectively get documentation on demand. Users will type in questions,queries,quotes (of source code) and once the issue is resolved (think of it as a dynamic form of literate programming) can be rendered to hard copy for permanent archiving/case-study/how2, etc ...
LL
Think people ... why do you have ads?
...
... ?
... (and I'm not talking Amazon here)
... there's no reason why they can't commoditise other mass market non-perishable bundles (think personal care kits, think entire kitchenware, think car personalisation). The biggest barrier to adoption has always been social ... the better micetrap doesn't work beyond a certain point. It's all about distribution, reliability, after-sales service, etc ...
... this is a directory problem. Already you see personal advertising in newspapers disappearing due to specialised employment agencies, help-desks, ... I see this trend also happening to banner-ads. People will go to specific trading sites which persists reputations rather than wanting to be inundated with services they don't need at the moment.
... the other ads that are left are the branding which are image/style based ... frankly those wish to be associated with an experience and reminding the user that you're responsible for a over-limit bandwidth bill and a waste of time is not good karma. Coke and Pepsi sponsor rock concerts not sport statistics.
... movie trailers which use products that normal people can identify with and feel part of the crowd. I can talk about social alienation in cities which lead people to identify with their professional peers rather than neighbours but this is a geek-site not socio-economic trend analysis.
To sell something
What is the most efficient market for selling stuff
If you study the nobel prize winner of a few years ago, you'd discover that dutch auctions can be theoretically proven to be the most efficient price discovery process.
Guess who's implementing auctions in a massive way?
The same group that's expanding from collectibles to cars to sports gear to CDs
Already large-scale companies are dumping overstocked or out-dated goods on eBay
OK that's the long-term killer for ads. Now what about service organisations (ie offering something other than tangible goods). The service is about finding someone to do something that you can't do yourself
So
In summary, unless there are some fundamental problems with my observations, I would say that ads as we know them (banner, etc) will become ineffective due to going under the personal threshold of normal perception. Go to a rural place and you'd really notice the *ABSENCE* of billboards. Instead you will be mor eproduct placements
In summary, IMHO pure ad-driven renue models will fail. It might have worked for the radio-broadcasting industry which requires continuous listening but unless something radical happens to social perception of the internet, the ability to jump-click outside a walled domain, and the fundamental cost-structure (ads=bandwidth=costs) I don't see them being viable.
Of course the 64 million dollar question is what is a viable business model which all the VCs would give their souls (or unmortgaged remainder thereof) to discover.
LL
... that was benchmarketed [1] on /. a while ago ... (I vaguely recall a lot of noise comparing features vs iPAQ). Are the apps cross-source-comptible? I mean if the claim is that open-source creates innovation then by theory we should see a lot of competitive entries into the PDA market as the cost of entry is so much lower (create a compiler-farm and voila instant apps).
LL
[1] Some people are complaining about acronyms/jargon so I'll just clarify for newcomers. Benchmarks are side-by-side comparisons of features/functions/performance using quantifiable tests/numbers. Benchmarketing is the selective choice of a subset for simplistic bragging rights (e.g. spec-rating) which can be taken to extremes by benchcrafting (google on Mindcraft and Linux) which is an artificial benchmark designed to distort public opinion.
... Whistleblower legislation?
I believe certain countries specifically protect the livelihood of employees who call attention to illegal, improper or immoral activities? Often there is a public perception shift (e.g. slavery is now considered abhorant) but the initial persecution for people speaking out can be quite intimidating. Given that certain companies (again hersay) have a policy of not committing any sensitive information to permanent records like email, the only way is to encourage individuals to step forward and speak the "truth". Creating legal precedents which hinder the principle does not support the public good. Is shareholder value so important that you have to silence critics (both internal and external)?
Given the lack of corporate-employee loyalty, shouldn't prospective workers have a chance to study both points of view? Trial by litigation (as the modern form of trial by combat) seems a rather inefficient way of resolving such angst.
LL
... would the USENET archives qualify? As compared with say MS development network which is the equivalent knowledgebase? Hate, love or indifferent, you cannot deny that MS has had a major influence on the growth of the PC sector and a large part of this success is their fanatical devotion to their developers (please no jokes about if you got them by the balls, their wallets will follow). USENET is a nice snapshot but is it something purposeful?
... personally I would consider the first choice to be the Guttenburg Project which is low-key but represents the unflinching efforts of many many experts and volunteers. Given the dissipation of the social contract w.r.t. modern copyright laws, the scope and vision of the originators can only be admired.
I was just musing the other day about what would be the 7 wonders of the digital world
Sure, GNU/Linux could be nominated but I'm a little ambivient about it as the impact is mainly social (due to GPL and the contributors' belief in libre software). As a technical piece of work, is it on the same relative scale as the ancient wonders were in their heyday? We are talking global uniqueness, recognised by a wide population segment, and something difficult to duplicate here.
LL
... I'm just wondering whether the right social factors are in place for a broad-based uptake of XML. The real advance of the web IMHO is not w3c/AOL/etc but the IETF with its RFCs ... this allows companies, consortiums and even consumers to describe a protocol, and let the free market compete to provide implementations. Witness BXXP (or now renamed beepcore) which has Java, TCL, Ruby, etc libraries as well as C/C++ reference.
Now look at the same issue with DTDs. The fundamental constraints are that they are industry specific, tied to specific domain knowledge and it takes a lot of to-ing and fro-ing to come to consensus. Once a DTD is officially published, then modifying/varying it can be difficult. For example, I wanted to modify the XBEL (XML bookmark exchange) to add in a deprecate-by date field but it would be incompatible with all the existing implementations. Trying to work with large industry standards would be even more cumbersome as there are so many entrenched interests (just look at the proliferation of billing-based XML). How do large groups resolve the negotiations and compatibility issues (and talking about sub-schemas is another can-of-worms), not to mention the ontological definitions of any specialised language corpus (witness the biological community trying to define items in a rapidly moving field). XML might be the currency of computer originated messages but who guarantees its inter-convertibility (present and future)? For a XML based application to work seamlessly, cooperative structures that span multiple groups probably need to be established. While corporates would be more than willing to set forward, general mistrust based on past misbehaviour does not auger well for wide-spread uptake of the technology. How do you know whether that biz-talk or TLA of-the-day is not secretly sending out sensitive information? And if technical guys are dubious about security, how can you expect consumers to embrace something which is beyond their understanding and has little immediate benefits (XML IMHO is more useful for computer-computer transactions than computer-human).
The real revolution with XML will be social, not technological.
LL
The process of taking a basic patent and then lodging in the patent office pipeline a lot of iterative "improvements" can cause problems. In legal parlance, these "submarine patents" are kept under wraps until products they cover come to market, and then create a legal minefield as the patent holder seeks licensing.
... at a bare minimum it should be demonstrate that there is a period of continuous (or near continuous) improvement/development/implementation. Ie, use it or lose it. If the idea is so great and thus worth the expense of patenting it, then why are they sitting on it? Waiting for someone else to do the hard work of debugging and actually convincing the rest of the market to adopt it (the hardest problem in todays software world) and then taking a slice of the action is nothing more than parasitic behaviour.
I would note that perhaps we should set specific hurdles for software patents
Perhaps another reform should be the more innovative a software technique is, then the longer the period where it could be held by the originator.
LL
LL
When the behaviour is simple combinataion of philic/trophic responses then the analog model works due to the relatively gross scale of operation (this is intended for kids). Given a small enough sampling speed, this analog behaviour can be mimicked by digital means (effectively the discretisation and solution of PDEs). The advantage of a digital model is that (if programmed correctly) is based on a Turing machine which is much much more flexible. That means you can create new inputs/responses rather than the set of prewired functions.
... he who dies with the most toys wins ....
Still
LL
Man ... sounds like the electreonic equivalent of a retrovirus (for those without a biological background, these are virus-like critters that modify small segments of a host cell's DNA, usually by inserting themselves at inconventient spots). These are often associated with STDs and tend to be very difficult to eliminate.
... all cable operators (which given their telco background is probably their CTO's wet dream) have their share price valued at $XXX per customer which reflects the long-term monetary value they are able to "extract". Now given the competitive landscape of many alternatives (satellite, wireless, 802.11 etc), the only way they can "increase value" (aka screw the customer) is to charge gatekeeper fees ... hence the gratuitous crack of your system, reversionist defacing of your bookmarks (to their premium advertisers of course) and rebranding of your system (ie treat you as a free billboard for any visitors). The cost might be zero (for the CD) but the value for you as an individual is definitely negative ... I'm surprised that this can actually get through the equivalent of better business bureaus because it appears to be bordering on the verge of deceptive advertising (scaring you of deletorious effects if you don't use it).
....
...
You have to understand the corporate mentality
I certainly hope someone from the consumer report is taking note of these type of activities
Best wishes for a speedy recovery
LL
The golden rule used to be "whoever has the gold makes the rule" but I would observe from current machinations the golden collorary "he who writes the rules, defines the gold".
... think what the internet does .. every single piece of information whether written in the past or immediate future (think trailers) is now immediately available. It's like a thirsty man in a desert being swept away by a flash flood. All the historical economic models based on a content/distribution model is now completely invalidated. Historically media studios could release stuff at different price/time points (movies, videos, cable reruns, etc) with the nice kicker that a popular franchise can be remastered with relatively little marginal cost.
The reason
Now suddenly anyone (with a modicum of hacker skill) can bypass their time/space-controls (cough DVD-region-coding), the TiVo is just one small example. Suddenly all their media libraries is implicity devalued as they can't withdraw "obsolescent" titles. The First Sale doctrine means that anyone can resell their "original" copy which creates competition for their newest overhyped gee-whiz production. Hence their incentive, nay long-term economic survival, in pushing Digital Rights Managment (aka service selectivity/variability) by stealth (submarine legislation) or by wealth (trial by litigation).
Of course, they don't always have much of a clue (cough*CueCat*cough) so they have to rely on the tech experts to provide them with the tools to control/segment the entertainment market. Which means that unless you have a tech department under your belt like AOL, they are held over the barrel by the likes of Microsoft who have their own ambitions of being the broadband toll-keepers.
Economics alway always been about scarcity (whoever dies with the biggest toys wins) but the internet inverts all that into a surplus. The "gift culture" that ESR mentions is thus anathema to any self-respecting aspiring monopolist as infinite replication/distribution of information-based products limits their market of gullible fools.
It will be an interesting decade as all these economic forces resolve themselves.
LL
The concept is that the market price should be close to what people are willing to pay (fair value). Airlines specifically market fisrt-class as "better". Just like they put on extras onto DVDs to make them more attractive than video tapes. Now if game publishing houses are too stupid to deploy the same tactics (e.g. software easter eggs for English market, etc) then they are not addressing the equity problem (why customers get disturbed with paying varying prices). We have "social" discounts such as veteran benefits, student fares, etc ... if you extrapolate this on a global basis, why should you feel insulted if a company decides to cross-subsides countries with lower GDP? You see this selectivity everywhere, from having chip manufacturers test MHz from the same batch and down-grade those with unstable charateristics to having software/firmware governers that step-down the performance (a favorite for IBM mainframes).
The concept of discriminatilon ranks with many people, but how do you separate those willing to pay for convience/priority even though the product/service is exactly the same? There ain't no such thing as a free lunch.
LL
... the problem with the entertainment industry is that it is often tied to disposable income (if you don't watch TV you're not going to die regardless of what kids think). As such there is serious competition for our attention ranging from walking in parks (NY muggings excepted) to window-shopping in malls (a legitimate form of entertainment as shown by theme parks taking this philosophy to extremes such as Disney). Groups such as Sony have to come up with ever more inventive ways of parting you from your money ... err catching your attention and delivering amusement. This problem is exasperated by the fact that different countries value leisure differently. A third world sweat shop worker just simply has better things to buy (like education for their kids) than light entertainment. Hence global companies cannot charge the same price for the same item (CD) in different countries. Hence their desire for market segmentation tools such as multi-zoning.
... up to the point of lobbying legislators (cough*DCMA*cough) to exterminate what they view as inappropriate economic conduct.
Now is this considered fair? Places like Australia don't believe so as their competitive watchdog recently ruled that multi-zoing was anti-competitive as it hindered parallel importing (is source CD from other countries). On the other hand companies argue that it is like passenger classes in planes, first-class still get there at the same time as cattle-class but pay significantly more. Many companies (esp software/pharmaceuticals) use the high prices of their products in 1st world countries to cross-subsidise less developed markets. Given the increasing connectivity of world trade this is becoming increasingly difficult.
Computers with digital rights management (aka service variability) is one mechanism to enforce this market segmentation, especially if it can be enforced through fixed/controlled end-points (cough*Xbox*cough). This is why companies hate mod-chipers and related products (satellite decodes, overclockers, etc) as it allows individuals to exploit the artificial price differential between 1st/3rd world pricing strategies. The end-result is a technological arms race (embedded ids, self-destruct, registrations, etc) in order to maintain this separation between high-margin customers and more marginal users. A person collecting warez for bragging rights is *NOT* willing to pay the same recommended price as someone looking to kill time by renting an evening game.
Anyone who thinks a company is going to destroy their global economic model just to please a small (but vocal) group of (from their point of view) "parasites". A large enough business entity can tolerate a small percentage of free-riders but is likely to come down hard on any systematic or organised threats to their business provided they can distance themselves from any media-fallout (cough*Adobe*cough)
Fortunately the free market (e.g. open source movement) has a little influence in moderating the extreme behaviour of the more pervasive global corporations.
LL
.... reminds me of a regional finance company that specialises in infrastructure funding (things like ring-roads, traffic bypasses, fibre, etc). They built a high-speed traffic bypass asnd started charging tolls. However, they found out residents were still using local roads so they did a deal with the state authority repsonsible for public roads to seal them off.
:-).
The point is that they did the conomic analysis and showed that there was a net savings in petrol consumption and driver's time. But in order for them to pay for building the new infrastructure, they had to convince motorists that there was no longer a free-ride. The problem is that motorists only saw the daily toll charges and not the weekly savings in petrol/time. Not to mention that very little advance warning was given to changing the road access. You can probably guess the PR fallout resulting from this
Sooner or later a similar scenario will happen with communications networks as they reach the limits of scaiability and in order to transition to a more efficient/lower cost/reduced maintenance system, they need to convince people of the benefits of switching This usually requires changing their usage patterns. Unfortunately aggressive telco upstarts don't always have the diplomatic skills to address customer's expectations. Pricing is a particularly sensitive point as there may be incredible customer acquisition costs or hidden cross-subsidies that distort the cost structure.
There has probably been some over-investments in network infrastructure that the current recession is revealing. As Warrne Buffett says, it's only when the tide goes out that you see who's swimming naked. Companies that pass the buck (literally) for their corporate mistakes are going to have a hard time keeping onto their customer base and will have to either swallow the losses (and shock horror forfeit the CEO bonuses/options) or else try and merge to gain monopoly pricing power and justify their executive packages. While some people may decry the double-sided nature of telcos and Wall Street, hopefully the survivors will be more sensitive to their users's needs.
LL
LL
VC's probably overstate their talents but at least they're willing to risk other people's money in ways that banks can't or won't (and if anyone who started up a business on a credit card can attest it is nerve-wracking).
Now if people could only come up with some clever Open Source Funding models.
LL
... there's been some rather interesting work being done on analysing markets based on assymetric and incomplete information (in fact I beleive this years Nobel prize in economics was on precise that ... selling used cars as lemons). The problem with software (especially closed) is that there is zero information about the quality of a *new* piece of software. Until you get to use it (and discover + avoid things which BSOD it), it basically has zero utility.
... ie through the school of hard knocks, it has demonstrated that it is the "best" at a particular niche (given economic constraints) and which case it is better non competiting against it. Look at stuff such as Aladin ghostscript, or the publishing templates DocBook.
... (cough*intel*cough with claims that its high-speed chips were essential for the internet ignoring the minor issue of bandwidth/latency). Unfotunately marketing is a sunk cost ... once you've spent your advertising budget, you can't recover it. Hence for firms who suddenly find they've overspent, the only way they can make back their money is to either eliminate competitors or else or cut back on the quality/support. And guess what, if you're the only solution, then you automagically create a self-fufilling prophency that your product is the "best" in its category and therefore all the technology followers (ie conservative foggies who don't trust anything less that v3.0) are forced to purchase it irregardless of the absolute quality (stability, security, mtbf, etc).
...
So how do software companies actually generate cash-flow? They have to "prove" their fitness in other ways (ie signal to the potential buyers that their code solves your future needs). Whether it is giving out a first freebie (QUAKE) or carpet bombing the user base (AOL), somehow the vendor needs to demonstrate that they provide something worth the "value" others place on it.
Now, the curious thing is should open source be such a signal? It might not be the "monopoly" in the traditional sense, but more a meritopoly
To actually get bigger than a sustainable niche, you basically have to bullshit like crazy
However, with open source this approach doesn't work as you cannot eliminate a competitor that cannot be bought off. As such, a rational player would instantly quit the market segment as soon as they can tell the open-source version exceeds their market share. Hence by definition, OpenSource becomes the new monopoly
LL