You as the author are not required to only license
under GPL. This is in the GPL faq. While FSF (obviously) discourages license changes which would
take a piece of code proprietary, If you'd have
to switch to say BSD license and don't mind doing so then GPL cannot prevent you from doing so.
A decade ago RISC workstations (mostly running Unix,
although NT started coming online soon after) had a
far better price/performance on floating point
compute cycles than Intel x86. If you needed to solve problems beyond the capacity of RISC processors (then 20-50Mflops) for $10-20K you needed to move to Cray (vector) processors (200+Mflops / CPU) with
costs running in the range of a million $US per
cpu.
Unfortunately crush analysis (and structural FEA in general) doesn't scale all that well on multi
cpu systems (at 8 CPUs you might see 4x speedup).
Thus if you really needed run this sort of solution (in less than a couple of months compute
time) it was worth the cost of the Cray.
5 years ago while Xeon began to eclipse RISC in
showing the best price / performance numbers, Unix/RISC (usually using super-scalar architectures) systems scaled to the point of
replacing the vector based systems and proprietary Unix clustering has been solid and
scalable to an order of magnitude beyond what
you could think of doing with NT/W2k.
Today the 'sweet spot' in price/performance is
definitely intel Xeon and as most of this code
has been developed on Unix, Linux is easy to do,
and Linux clustering is certainly good enough to
manage these tasks.
Nothing against stationwagons, I have an '86 volvo that I'm very happy with. of course it's also easy to work on.
The ducati on the otherhand is a little hard to maintain, but worth the effort.
For my $0.02 Debian looks like the worst of both worlds, arcane to manage, and not very fast, I prefer and useobsd & slackware. Ymmv
of course, sane people who I respect use and like deb.
Ahh so you know enough about antitrust law and
the judiciary to comment on the competence of a
senior judge?
I found Jackson's article to be persuasive and as
intelligent as I had come to expect based on his
performance in the MS trial:
The judiciary is in many ways the most secretive of the three branches of the federal government. It is not subject to the Freedom of Information Act or any other so-called "sunshine" statute. Judicial disciplinary proceedings are conducted in private. Although the judicial system professes to display its decisional processes "on the public record," its most important decisions are made behind closed doors, whether by judges or juries. Law clerks and supporting staff are sworn to secrecy. There are remarkably few "leaks," and no whistleblowers. A veteran journalist once told me that "we know more about how the CIA operates than we do about you."
Jackson demonstrated having a firm grasp on the
technical and economic issues at hand. This is strongly at odds with my general experinence. It is most ofen among lawyers and MBA-types that I get the "Microsoft is just a good thing, look at them they're so successfull" attitude.
I'm glad this one indeed is willing to step a little outside the box, and from all I could see of the following 'remedy' proceedings the current
judge will be following Jackson's logic, if not the specifics of his recommended solution.
and healthy OSS-based businesses as evidence in support of this article's observations:-).
I've been a user and author of free software for
a dozen years and I remember when this existed
only within BSD and as the gnu tools for use in
mostly proprietary OS environments. Today of course
free software (in the oss, gnu and bsd senses) run in a wider variety of patforms and even have predominance in a few key areas. That's all good.
That some granted patents are flawed is not news,
and these mistakes are expensive to fix and imx
do not represent the average situation. In fact
the various patent offices do provide consistent and valid interpretation of applied patents, and in any case are a reality which I don't see changing any time soon.
The statement "Information want's to be free" is an observation, not a valid challenge of either the constitutional or statutory basis of the IP legal framework.
And exactly who do you think is going to provide the investment in IP if there's no way to recover
the R&D expenses?
Nearly all patents are (and have been for a long time) developed in corporate or.edu research labs and are automatically assigned by
employment contract. The reason: very few engineers
have the funds for the research, let alone the legal
expense of defending IP.
As IP is (for the forseeable future) an asset, how is restricting transfer / assignment any different from restricting transfer of any other
asset or property.
All of my (5) (medical device) patents are assigned to a (former) employer. That's ok I was
well paid for the time I spent developing these
inventions.
Sure Disney and RIAA are pushing the politicians
into some stupid extensions, that does not render
the entire concept wrong (or as easy to change as
the denizens of/. seem to wish.
Watercooling the cpu and eliminating the fan
results in very little cooling for all the other
parts of this cute little box. The HDD espeically
will suffer reduced life expectancy.
Back in the 1960's, the U.S. State Department used to warn U.S. citizens of the dangers of travelling to the USSR, siting that the KGB (secret police) could do whatever they wanted to them
What's news in this? break the law in a foreign
state and you're potentially in very big trouble, and under the local rules.
If a US citizen leaves the working compounds in Saudi Arabia and runs afoul of the authorities the penalty for stealing is the loss of a hand.
These boys were stupid enough to come to the US
and demonstrate their 'prowess' and got arrested.
Nation states laws only apply within thier jurisdiction, what's more, usually foreign policy is clear on these matters. It's not illegal under
US law for companies to practice monopoly methods
outside the US, never has been and probably never will be.
What's more, this is recognized. So while I have no idea how Russia feels about this, I'm dubious on any concerns that the FBI was breaking Russian law. FBI may have been somewhat unethical in sniffing the suspect's passwords but those are the rules of law enforcement: if you can trick the perp into doing something stupid you win. Once the passwords were in hand no cracking was needed to gain access.
Finaly this is how it's done in the 'real world'.
and just as there are places where the US State Departement can't get a US citizen out of trouble, just so, even if Russia wanted these
guys back (which I doubt) I rather expect they understand the score.
This is similar to how Chris Tresco was nabbed,
Dod was infiltrated, evidence gathered and a
bunch of kids who thought they were all 'leet
discovered otherwise.
As it happens I've met Chris - he did a linux server config for me several years back, nice enough guy (but not all that sharp with linux)
Ok, 'lilofree' came out of a beef in 2001!?? old
news imho
I work on OPN irc daily, lilo's requests for funds
(which admittedly fall on deaf ears here) hit my
server window, and are pretty unobtrusive.
Contention is surely the nature of IRC, snotty OPs dripping attitude,/banning folks who don't kowtow. The OPN net is well run imo in that it
actively discourages that. Of course as an ircnet
which is nominally intended to aid opensource collaboration, that's as it should be.
I don't really care about politics around Mr. Levin. As long as this ircnet is better run than
most (admittedly damnation by faint praise), I'm
happy to continue using it. From what I can see the rants are just more IRC noise and certainly if Debian doesn't want advertising, all they need
do is move.
And no surprise that the Debianistas would be the
ones making noises. Same crowd that's been picking a fight over GPL with Linux -- Pffft!
I say Debian - good riddance, just as they can
now use the long-awaited Hurd if Linux isn't ideologically pure enough. Personally I'm happy to live in a real world and deal with real people, who sometimes don't live up to my expectations. I can always choose to move elswhere, no cause I can see for making a big flap.
As ever,/.'s run an article that has some potential for raising the GNU vs Linux (vs BSD vs Unix...) spectre. Yawn...
If the linux kernel had been released using the BSD license it would still be popular Why do I mention this? The point is that *how* the project is run has a lot more to do with how the development is managed than the details of the license. BSD *could* just as easily be run as a bazaar and I've seen plenty of GPL-licensed projects whither on the vine because for one reason or another the founders don't allow adequate control to the 'troops'. And OSS troops have a way of voting with their feet when projects don't meet their needs.
Many of todays license 'zealots' are seriously misinformed I have read opinions on/. that BSD is somehow beholden to GNU because because "most of BSD is GPL code" Of course this is nonsense, yes lots of GPL code runs on BSD and some (e.g. gcc) is indeed part of the *BSD variants. However people who actually use the BSD's have probably observed that most of the OS tools are not the GPL variants; ls, cat, more... are all quite different on BSD vs Linux. Hell, OpenBSD (the only current BSD I use) provides csh, and it's/bin/sh is a POSIX shell, not bash.
For all the "idealism" RMS and FSF are practical in what they do Myth would have us believe that Linus is the king of practical and RMS will sacrifice no practicality on the alter of the GPL. I'm sorry but facts do not bear this out.
In '93 when I chose to push a little corporate $$ at FSF I purchased tapes for GCC, emacs, X11 and various GNU utilities. FSF was as happy to accept their $150 fee for X11 as the GNU pieces. Later, when Linux had quite enlivened the opensource arena RMS (in '98 I think) decided that the X license was bad (I think this is the same timeframe as he started in on the GNU/Linux rants).
I think it's notable that RMS/fsf have redoubled their efforts on the Linux kernel at the time that Hurd is finaly ready for prime time. I wish them well in yet another attempt to balkanize the landscape, I'm sure the Debian folks at least will be on the bandwagon...Ho-hum...
Now it's time for the ad-hominem ?? I'm sure RMS is a fine guy, and there are things about him that I find admirable. However, I truly find the tune he dances to to be all about RMS. Over the years he's used whatever tactic seems best to push his personal agenda. Honestly if I want to play power games I'd far prefer to do it in a context of leathersex
or BDSM, which is a helluva lot more fun way to play power games imo.
I'm hardly about to start paying $50 or $100 a year for an email address when I can get from Microsoft or others for free.
Iff you're willing to have your email, web pages etc plastered with advertising, then by all means go get it free. I don't use macs so mac.com would not be the first thing I'd go to but as near as I can tell they're offering a decent service. Many moons ago I used the e-world service that was basically a mac-centric clone of AOL (same software, for all I know same network and services). IMX Apple did manage to provide a better (larger) signal:noise environment.
The boom year+1/2 of internet-hype surely led a lot of folks to expect they could get services free on the net, and the fact that most of the businesses offering these services were underwriting operations with checks written by investors (i.e. diluting shareholder's equity) meant that *all* services had to be offered for free in order to get customers.
The flip side of this existential coin of course was that the users data was being collected, on the theory that fine-grained tracking/profiling would create lucrative new abilities to target customers.
I for one quickly tire of emails from yahoo et-al subscribers plastered with spam trailers. Mac.com addresses don't have these, so if they're now having to charge for it, then those users will get to make a choice between a relatively higher quality service and annoying people like me who absolutely abhor commercial adds in private emails.
In motorcycling we say 'if you have a $50 head, by all means use a $50 helmet'. If your web pages / content / email doesn't look worse for a commercial trailer over which you have no editorial control then a free+advertising service is the the thing you want.
he would actually be pretty interesting to meet...but I bet he [B Gates] still knows his stuff.
Trying not to drop into an ad-hominem cheapshot
The last project that I know of which Gates authored was a ROM Basic. The Basic interpreters which followed for various early microcomputers were written by his assciates at MSFT. Of course whether or not Bill still writes code I have no actual knowlege, but nothing I've read from MSFT suggests that he acts in any capacity but architect / vision-leader.
I think this 'aura' of a brilliant coder plus his wealth is exactly the primary MSFT strategic advantage. I know dozens of lawyers, MBA's, executives who seem to beleive the following:
This guy (company) is fabuloulsy successful so their product must be just wonderful and: He's this really brilliant programmer / geek and that's the basis of it all
And because these folks haven't got a tech background they're basically taking it on faith. honestly it's insidious, I've seen an entire company (very big one) in a different business say 'wow that's great, lets emulate it... ohh and yes lets also go with MS in the Data Center! [doh!]. (They fired an MIS director and then a CIO who couldn't make this fine strategy actually work in practice.)
Now what *is* true about Bill (IMO) is that he's really bright (and that his early commercial coding was largely in either assembler or on DEC PDP / Vaxen used for emulation of 8080/z90/x86 systems). Where to my knowlege he applies this is strategy and architecture, and if I don't like his choices, I'm the first to admit they've been effective (if underhanded and illegal) in the market.
Second, for 2 decades MSFT aggressively hired the very best and brightest CS grads. A freind who teaches in one of the better university CS departments observed this and on that basis only started investing in MSFT. That was a very good investment strategy for him:-).
Today I think even the financial types are beginning to realize that some of this is smoke & mirrors. I think the combination of unreasonable licensing changes and the slap on the wrist they just got from SEC are just the sort of thing that these people pay attention to.
Microsoft has always been brought more or less kicking and screaming into standard technologies. netbui vs tcp/ip; WINS vs DNS; NT Domains vs Kerberos|LDAP. Often they have implemented open technology (DCE) in the internals, just not making these the preferred API's.
Of course the whole time I and other opensource types have been looking on and saying *yikes* you want to put this cruft in an enterprise??! MSFT is highly feature driven and lusers love features. Nowhere near enough coders (or architects) work to the priciple that the least code that will do the job is usually the best solution.
Througout, Gates has pushed Basic as the language of choice [shrug]. Gates I don't really want to meet, his original partner Paul Allen, also a billionaire who has said "Blame me for having to type the backslash"... he doesn't want to meet me in an alley:-)
I will say that I'm glad Gates is focussing on technology again..NET has promise, and the mono initiative will make it open. His foundation is also giving big money in important areas of medical research and he cares about the right stuff, (e.g. HIV/AIDS).
It may be that Peru has actually accepted just software and not support services. Software support is quite expensive, and software without support is generally not very valuable.
As the article I read states "software, cash and consulting services", I imagine MSFT will be providing support.
It doesn't always work out that way, I read years ago that in the '70s W.R. Hearst 'gave' a dismantled monastery to the city of San Francisco (probably reaping a nice tax benefit). However as
he neglected to provide any funds to assemble the stones, they sat in Golden Gate Park, rain washed off the identifying marks (which goes where) and thus today there's just a pile of stones.
Anyhow, in the automotive industry, my uderstanding is that most of the *profit* is in 'support', auto manufacturerers typically sell the vehicle at or near their manufacturing cost and make their money on parts and service.
One thing I don't (and don't expect to see for awhile) is provisions for requiring computing systems to be 'secured'(sic). There's a fairly large tug-of-war on the 'Net these days between those who are responsible for maintaining production systems and people who run poorly secured systems which are routinely used for attack by the whoever chooses to use 'em.
In most states of the US (and most developed nations) you are not allowed to operate an automobile without maintaining basic safety (and emissions) equipment. I expect sometime in the near future similar requirements may be made of systems connected to the internet.
Today the conversations may look like: ISP: Your system is being used for attack by an intruder, if you don't take it offline and get it fixed we will enforce our AUP and take you offline. customer1: Ooops, sorry ok we'll spend the $$ / time to fix it customer2:YOU CAN'T DO THAT.. I pay for this service and I'm not responsible / can't afford to fix it... ISP: CLICK
Today, while it's feasible to keep systems patched / audited for a reasonable level of safety, many (most?) orgainizations don't have the skillset / funds allocated to keep their systems secure against even the 'kiddies, let alone a determined attacker. That's gonna have to change IMO either thru systems that are harder to break into in the first place or better practices.
Some of the provisions of this bill are also simple clarifications of existing statutes.
For instance see the provision: Specify that an existing ban on the "advertisement" of any device that is used primarily for surreptitious electronic surveillance applies to online ads. -- apparently while it's illegal to advertise wiretapping equipment in print, this will extend the restriction to online ads also.
This explains why I've been seeing the adds and spame for keyboard keystroke recorders (shame on you thinkgeek!) and packet sniffers to protect (spy on) your kids or spouse.
I originally fielded snort (1.7x) on a home-built k7 server running slack 7 / kernel 2.2.13, in that setting it
would segfault when local traffic began to push the 100 mbit level.
More recently I've run short 1.8x on a dual P3 w/ ECC Ram & hardware RAID for a year now, first on kernel 2.2.19, and more recently 2.4.17.
Not a single failure in 12+ months of either snort or the kernel. I suspect that the combination of a marginal backplane and IDE disk, coupled with the older kerenel probably caused the first machine's problems.
the fact that most people don't use Linux means that the value of using Linux is less than the cost of using linux.
The cost analysis was done based on linux, however most of the code analysed in fact is for things that run on other platforms, and much of which was in development for years before linux 0.9 hit the 'Net.
So the measure of value based on who uses Linux includes everyone who uses linux-hosted apache servers. The more general case includes everyone who accesses servers that depend on (Perl, BIND, sendmail, mysql.... etc) or were/are developed using (X11, CVS, bitkeeper, emacs, gcc.... etc)
The economic value isn't small. That much I'm pretty certain of, just how big, well it works for me, I'll leave the analysis to the economists.
Where have I heard this before?
on
Microsoft Freon
·
· Score: 4, Insightful
Though it is unclear whether such a product will ever be built, its core concept appears to have the backing of Microsoft Chairman Bill Gates, who wrote in an internal memorandum in January
So much for other companies trying to raise funding to develop products in this market.
Oft-observed MS behavior:
see a new technology;
(sidebar: contact nascent developers of the new tecnology and express interest in 'partnership', get a feel for their approach);
write press release announcing newest MS brainstorm, including vagues statements about timing;
suggest you go have a look at LSM - Linux Security Module... see the.mp3 of the lsm presentation at this weeks kernel summit at lsm discussion. LSM creates hooks in the kernel functions which are security relevant (about 150) and can mitigate access to a couple dozen kenel data structures.
security model is too coarse grained:... move towards ACLs, for example in NSA's SE Linux, as well as LIDS
Actually SELinux does not implement ACL's, but rather Type Enforcement. It also has potential (and experimental impementation) to implement MLS or other security policies / methods.
What type enforcement gets you is the ability to create highly fine-grained security controls, so that the program and user-security-context have privilege to execute critical functions and that privilege can be removed from the root user.
One of the debian SELinux implementers placed an SELinux system on the 'net with root-password / ssh access advertised. This is not a proof of safety, but in fact noone succeeded in escalating privilege.
As it looks like LSM is on track to be in kernel 2.6, at least the way is presently paved.
The perl motto (there's more than one way to do it) applies to computing generally, and to the idea of what's the 'selling point'.
As it happens I have a couple of posts today about the what goes on in the OS market, so I'll just link Exchange on nt 3.1? & lessons for OSS and (long) Is Linux Dead. And yes, the market will hand irrelevancy to systems which don't adapt. If GPL does not adapt it will follow the same path, and by my read it has:
Beyond all that, free is not just GPL. FSF used to distribute X11 from the X-Consortium at ~$150 / tape Sometime later RMS decided the X11 license was 'bad'. Today, (perhaps with Debian/Hurd as his ace in the hole?) RMS is trying to push Linux to a strict (activist?) interpretation of GPL.
When I look a the history I think Stallman for all his principles exhibits pragmatism in his actions which is so often attributed to Linus. Linus made it clear long ago that he was not going to give FSF/Deb the blank cheque that many GPL developers do licensing under "gpl-current or whatever later version"
I daresay Linus drew a line in the sand saying "2.0 and no later version of GPL", and I bet if he hadn't we'd be looking at GPL-V3 today.
To be clear I'm not trying to knock either approach. I happen to have a bit more sympathy for Linus's views but that doesn't invalidate the strengths and accrued benefits of other approaches.
Re:read article / which uses 'interesting' stats
on
Is Linux Dead?
·
· Score: 2
Umm.... the MSNBC article states that MS leads in the server market. While a simple statistic of licence or server counts might imply this the investment or total capacity measure would be a very different number. NT / Win2k still don't scale as well as Linux which in turn is still a far cry from what's possible with Unix.
Here's my letter to the MSNBC author suggesting different interpretation / conclusions from his analysis.
Dear Mr Schoen,
A few notes on your column regarding the absence of Linux presence
(at a recent tech expo in nyc? the context / event was not clearly
named).
First I'd like to note different versions of a few of the statistics
you present.
The statement that Windows (NT/2k/XP) servers are the largest server
installed base: That is probably true if you look at number of licenses
or CPUs. However if you look at a more meaningful number like total
computing or transaction capacity running under Unix / Linux I think
the number rather changes.
Win32 has it's best penetration in small organization and departmental
applications. Windows does not scale well past 16 cpu SMP or clusters
beyond a few dozen nodes. Therefor virtually all very large servers run Unix,
which runs in SMP beyond 64 processors and can aggregate a thousand or
more processors into supercomputers or large clusters.
Linux still lags behind the vendor-based unix platforms (IBM, HP, Sun),
but is already considerably more capable of high-end scaling than MS's
offerings, and is penetrating that market at a pace which reflects it's
strengths (and weaknesses).
It is not news that Microsoft has aggressivly targetted both Unix and
Linux in the data center. And for all I can see they are not winning
anywhere near as often as they would like. IBM has won a number of
very large accounts with AIX / Linux in direct competition with MS.
This is pointed to in your own statement "Linux server sales jumped by
more than 50 percent to $400 million, with IBM leading"
Part of the reason for that is that many Unix shops are able to deploy
less-expensive Linux servers where priorly they were locked into
proprietary RISC hardware running vendor-based Unix. However they *had*
invested in platforms with fundamentally open programming interfaces.
The point here is that these savings are exactly realizable by organizations
which had initially invested in open-platform investments. Merril Lynch
recently announced a conversion of internal applications from Unix to Linux,
citing the lower cost of maintaining applications within the opensource
framework of Linux.
The key point here is that while porting applications between proprietary
Unix versions (e.g. HP, IBM, Sun) is relatively expensive compared to
standardizing on Linux, Windows simply doesn't play in that equation
in a meaninful way. Microsoft is working very hard to counter this
equation, but the economics don't work in their favor in the data center.
It is simply not meaningful today to compare market penetration from Desktop
to Enterprise in either tech or economic terms. The market environments of
servers at application, database, web and department are all markedly
different. MS has excellentpenetration at the departmental level and of
course the Desktop, but they have yet to translate this into serious
penetration in any of the other three.
Those of us who do adopt linux on the desktop (I have been using Unix for
professional / technical / server computation for 10 years) continue to
marvel at the low reliability expectations of Win32 desktop users. Yes things
are a lot better than win 3.1 / 95 / 98 days when a networked workstation
could not be counted on to stay up for a working day.
NT and Win2k have improved on that record, however they are still a far sight
behind Linux in reliability. I run Linux on a late-model laptop (which is
probably Linux's toughest challenge, given that virtually no vendors support
Linux on these systems, due to the low market share / returns to investment).
My experience remains that I have better stability than I get with Win2k Pro
Sure I'm not the average user. My work is either technical or programming and
the tools that work best for me are from the same domain as the servers which
I run, deploy and support. And I'm not religious about this. As long as
end-users have an easier time with proprietary systems they will (and should)
stick with them.
However I also know that when the New YorkTimes runs anarticle on the relative
compatibility, ease of use and low cost ($50 supported) of Open Office
(On win32 or Linux) compared with $600+ for MS Office. I think it's fair to say
that Opensource is beginning to competing with MS on the desktop.
http://www.nytimes.com/2002/06/20/technology/circu its/20STAT.html
Many observers equate Open Source software with Linux, or the Free Software
Foundation, or GPL. In fact the concept of OSS is probably older than Unix but
Unix and the original Arpanet / Internet were the substantive proving grounds
for these concepts. Today this has paid off in an infrastructure potent enough
that it can develop cheap, reliable open systems that also run in the closed
platforms.
How the race will play out remains to be seen, however if you are going to write
about the subject I think it's best to look at all the elements of the market
not just the ones which bolster your preferred outcome.
I couldn't agree more, and I was unaware of the steel and agriculture issues, but trade negotiations in those areas are also pretty standard stuff, you'll not I also indicated my concern: >> They still have opportunities to snatch defeat from the waiting hands of victory,
haven't even touched upon what this can do to China
Asia today trades more with itself than the west. That is a sign of a long-term shift to a more healthy economy. The only thing I think we can count on is the outcomes aren't easy to guage / guess. The world economy is a complex beast and will pretty well do what it wants. Most of the world understands this today (which is why OPEC hasn't tried to push oil prices to where they hurt the overall economy as happened in the '70s)
The scariest news is that we'll probably come out of this even bigger
I wouldn't bet on that (either way), Asia has good opportunities to realize real growth, but anyhow I'm not an economist / don't play one on tv:-). Dunno what's coming, but I expect it to be interesting
kernel developers to Debian: put up or shut up
on
Kernel Summit Wrapup
·
· Score: 3, Insightful
my own favorite quotes from the discussion: - audio (approximately)
(Linus speaking): moving this (binary drivers with which Stallman / deb take issue) into user space is a sign of mental disorder.... we are clear from a copyright standpoint... linux has intentionally taken a non-rabid standpoint... as I've shown with my use of bitkeeper I don't care about black and white people.
[issues about firmware && binary modules]
(Alan Cox?) The kernel developers do not have energy to sit down and determine a clear set of rules... Debian has an endless supply of people who have nothing better to do than study legal issues....
[Linus points out that actual GPL violating files get addressed in ca 24 hr timeframe]
The conclusion was to send a message back to the Debian users to "put up or shut up"
I'm sure RMS will have a press release out later this week.
The comparison to the auto industry doesn't work. Back when there were 50 auto companies
there were a lot fewer people...
I suggest you look up 'Kondratiev cycle technology'. Jay Forrester (inventor of magnetic-core memory), studied this at MIT/Sloan school and determined that the a long-term economic cycle develops due to the 'self-ordering' nature of capital equipment.
Basically Forrester's group found evidence for the following feedback loop: Early in the deployment of any technology there is a scarcity of capital. Capital equipment is expensive, and early investments involve high degrees of risk accompanied by high profits in a given technology sector. That in turn brings investment in the businesses developing this capital. However, a large part of this capital is used in the development of the capital itself (i.e. IT tends to need advanced hardware and software to develop the bleeding-edge new hardware and software for actual end-use).
Thus the 'buildup phase' of new technology creates a high demand (for both the acutal equipment and the stock of the companies that make it). At some point, however the generation of this new (and expensive) equipment (or software) exceeds the actual (end-user) demand. When this happens the high profit margins that were being realized during the build-up phase disappear quite rapidly, the investment-value follows (crashing stock prices) and the investment money looks for other places.
See this article
Sound familiar? Whether or not you buy into the economic details, this is one of the behaviors seen in economics. The inflated acquisition prices mentioned are the direct result of this effect.
Because sure people make stupid mistakes even (especially?) with billion dollar transactions. But the funny thing about the stock market is that the money doesn't ever go away. Whenever someone loses in the market, someone else has
made a profit.
And yes it sucks when the players break the rules but especially on the financial rules the market punishes you very hard. I worked for a biotech firm that was growing well, showing solid net earnings ca $300m on $2b sales, a 30:1 p/e ratio. Our japan division was found to have been moving inventories to the tune of changing the sales #'s by $50m. This lie, accounting for only 2.5% caused a nearly 50% drop in stock price and a (justified) shareholders lawsuit.
Whenever someone fsck's with the rules of the game (fixing the books, insider trading, breaking anti-trust rules, whatever), real people get hurt and we have SEC, IRS etc to try and keep up with the process. MS imo is an excellent example of how a determined and unscrupulous competitor can harm while evading the systems controls(sic).
I'm just thanking my stars that (so far) the politicians havent fscked up like they did after the '29 stock market crash. The US enacted protectionist trade tarrifs which effectively were the first blow in killing off the *world* economy.
Post-sept-11'th fears and RIAA / DMCA idiocies aside, at least across a several bumpy decades our boneheads in Washington, the EU, etc at least so far have managed not to fsck up. They still have opportunities to snatch defeat from the waiting hands of victory, but so far it could be a whole lot worse I think.
Slashdot Mirror
You as the author are not required to only license under GPL. This is in the GPL faq. While FSF (obviously) discourages license changes which would take a piece of code proprietary, If you'd have to switch to say BSD license and don't mind doing so then GPL cannot prevent you from doing so.
A decade ago RISC workstations (mostly running Unix, although NT started coming online soon after) had a far better price/performance on floating point compute cycles than Intel x86. If you needed to solve problems beyond the capacity of RISC processors (then 20-50Mflops) for $10-20K you needed to move to Cray (vector) processors (200+Mflops / CPU) with costs running in the range of a million $US per cpu.
Unfortunately crush analysis (and structural FEA in general) doesn't scale all that well on multi cpu systems (at 8 CPUs you might see 4x speedup). Thus if you really needed run this sort of solution (in less than a couple of months compute time) it was worth the cost of the Cray.
5 years ago while Xeon began to eclipse RISC in showing the best price / performance numbers, Unix/RISC (usually using super-scalar architectures) systems scaled to the point of replacing the vector based systems and proprietary Unix clustering has been solid and scalable to an order of magnitude beyond what you could think of doing with NT/W2k.
Today the 'sweet spot' in price/performance is definitely intel Xeon and as most of this code has been developed on Unix, Linux is easy to do, and Linux clustering is certainly good enough to manage these tasks.
Of course it could be run on win32 but why?
Where you see FUD here is beyond me.
The ducati on the otherhand is a little hard to maintain, but worth the effort.
For my $0.02 Debian looks like the worst of both worlds, arcane to manage, and not very fast, I prefer and useobsd & slackware. Ymmv of course, sane people who I respect use and like deb.
I found Jackson's article to be persuasive and as intelligent as I had come to expect based on his performance in the MS trial:
Jackson demonstrated having a firm grasp on the technical and economic issues at hand. This is strongly at odds with my general experinence. It is most ofen among lawyers and MBA-types that I get the "Microsoft is just a good thing, look at them they're so successfull" attitude.I'm glad this one indeed is willing to step a little outside the box, and from all I could see of the following 'remedy' proceedings the current judge will be following Jackson's logic, if not the specifics of his recommended solution.
I've been a user and author of free software for a dozen years and I remember when this existed only within BSD and as the gnu tools for use in mostly proprietary OS environments. Today of course free software (in the oss, gnu and bsd senses) run in a wider variety of patforms and even have predominance in a few key areas. That's all good.
That some granted patents are flawed is not news, and these mistakes are expensive to fix and imx do not represent the average situation. In fact the various patent offices do provide consistent and valid interpretation of applied patents, and in any case are a reality which I don't see changing any time soon.
The statement "Information want's to be free" is an observation, not a valid challenge of either the constitutional or statutory basis of the IP legal framework.
Nearly all patents are (and have been for a long time) developed in corporate or .edu research labs and are automatically assigned by
employment contract. The reason: very few engineers
have the funds for the research, let alone the legal
expense of defending IP.
As IP is (for the forseeable future) an asset, how is restricting transfer / assignment any different from restricting transfer of any other asset or property.
All of my (5) (medical device) patents are assigned to a (former) employer. That's ok I was well paid for the time I spent developing these inventions.
Sure Disney and RIAA are pushing the politicians into some stupid extensions, that does not render the entire concept wrong (or as easy to change as the denizens of /. seem to wish.
Seems to me a false economy.
What's news in this? break the law in a foreign state and you're potentially in very big trouble, and under the local rules.
If a US citizen leaves the working compounds in Saudi Arabia and runs afoul of the authorities the penalty for stealing is the loss of a hand.
These boys were stupid enough to come to the US and demonstrate their 'prowess' and got arrested.
Nation states laws only apply within thier jurisdiction, what's more, usually foreign policy is clear on these matters. It's not illegal under US law for companies to practice monopoly methods outside the US, never has been and probably never will be.
What's more, this is recognized. So while I have no idea how Russia feels about this, I'm dubious on any concerns that the FBI was breaking Russian law. FBI may have been somewhat unethical in sniffing the suspect's passwords but those are the rules of law enforcement: if you can trick the perp into doing something stupid you win. Once the passwords were in hand no cracking was needed to gain access.
Finaly this is how it's done in the 'real world'. and just as there are places where the US State Departement can't get a US citizen out of trouble, just so, even if Russia wanted these guys back (which I doubt) I rather expect they understand the score.
This is similar to how Chris Tresco was nabbed, Dod was infiltrated, evidence gathered and a bunch of kids who thought they were all 'leet discovered otherwise. As it happens I've met Chris - he did a linux server config for me several years back, nice enough guy (but not all that sharp with linux)
Ok, 'lilofree' came out of a beef in 2001!?? old news imho
I work on OPN irc daily, lilo's requests for funds (which admittedly fall on deaf ears here) hit my server window, and are pretty unobtrusive.
Contention is surely the nature of IRC, snotty OPs dripping attitude, /banning folks who don't kowtow. The OPN net is well run imo in that it
actively discourages that. Of course as an ircnet
which is nominally intended to aid opensource collaboration, that's as it should be.
I don't really care about politics around Mr. Levin. As long as this ircnet is better run than most (admittedly damnation by faint praise), I'm happy to continue using it. From what I can see the rants are just more IRC noise and certainly if Debian doesn't want advertising, all they need do is move.
And no surprise that the Debianistas would be the ones making noises. Same crowd that's been picking a fight over GPL with Linux -- Pffft!
I say Debian - good riddance, just as they can now use the long-awaited Hurd if Linux isn't ideologically pure enough. Personally I'm happy to live in a real world and deal with real people, who sometimes don't live up to my expectations. I can always choose to move elswhere, no cause I can see for making a big flap.
Hard to beat(sic) the Power Exchange @Otis St. Plan to go with a date, stag males pay $75 entry on weekends for the mixed space.
If the linux kernel had been released using the BSD license it would still be popular
Why do I mention this? The point is that *how* the project is run has a lot more to do with how the development is managed than the details of the license. BSD *could* just as easily be run as a bazaar and I've seen plenty of GPL-licensed projects whither on the vine because for one reason or another the founders don't allow adequate control to the 'troops'. And OSS troops have a way of voting with their feet when projects don't meet their needs.
Many of todays license 'zealots' are seriously misinformed /. that BSD is somehow beholden to GNU because because "most of BSD is GPL code" Of course this is nonsense, yes lots of GPL code runs on BSD and some (e.g. gcc) is indeed part of the *BSD variants. However people who actually use the BSD's have probably observed that most of the OS tools are not the GPL variants; ls, cat, more ... are all quite different on BSD vs Linux. Hell, OpenBSD (the only current BSD I use) provides csh, and it's /bin/sh is a POSIX shell, not bash.
I have read opinions on
For all the "idealism" RMS and FSF are practical in what they do
Myth would have us believe that Linus is the king of practical and RMS will sacrifice no practicality on the alter of the GPL. I'm sorry but facts do not bear this out.
In '93 when I chose to push a little corporate $$ at FSF I purchased tapes for GCC, emacs, X11 and various GNU utilities. FSF was as happy to accept their $150 fee for X11 as the GNU pieces. Later, when Linux had quite enlivened the opensource arena RMS (in '98 I think) decided that the X license was bad (I think this is the same timeframe as he started in on the GNU/Linux rants).
I think it's notable that RMS/fsf have redoubled their efforts on the Linux kernel at the time that Hurd is finaly ready for prime time. I wish them well in yet another attempt to balkanize the landscape, I'm sure the Debian folks at least will be on the bandwagon ...Ho-hum ...
Now it's time for the ad-hominem ??
I'm sure RMS is a fine guy, and there are things about him that I find admirable. However, I truly find the tune he dances to to be all about RMS. Over the years he's used whatever tactic seems best to push his personal agenda. Honestly if I want to play power games I'd far prefer to do it in a context of leathersex or BDSM, which is a helluva lot more fun way to play power games imo.
Iff you're willing to have your email, web pages etc plastered with advertising, then by all means go get it free. I don't use macs so mac.com would not be the first thing I'd go to but as near as I can tell they're offering a decent service. Many moons ago I used the e-world service that was basically a mac-centric clone of AOL (same software, for all I know same network and services). IMX Apple did manage to provide a better (larger) signal:noise environment.
The boom year+1/2 of internet-hype surely led a lot of folks to expect they could get services free on the net, and the fact that most of the businesses offering these services were underwriting operations with checks written by investors (i.e. diluting shareholder's equity) meant that *all* services had to be offered for free in order to get customers.
The flip side of this existential coin of course was that the users data was being collected, on the theory that fine-grained tracking/profiling would create lucrative new abilities to target customers.
I for one quickly tire of emails from yahoo et-al subscribers plastered with spam trailers. Mac.com addresses don't have these, so if they're now having to charge for it, then those users will get to make a choice between a relatively higher quality service and annoying people like me who absolutely abhor commercial adds in private emails.
In motorcycling we say 'if you have a $50 head, by all means use a $50 helmet'. If your web pages / content / email doesn't look worse for a commercial trailer over which you have no editorial control then a free+advertising service is the the thing you want.
Trying not to drop into an ad-hominem cheapshot
The last project that I know of which Gates authored was a ROM Basic. The Basic interpreters which followed for various early microcomputers were written by his assciates at MSFT. Of course whether or not Bill still writes code I have no actual knowlege, but nothing I've read from MSFT suggests that he acts in any capacity but architect / vision-leader.
I think this 'aura' of a brilliant coder plus his wealth is exactly the primary MSFT strategic advantage. I know dozens of lawyers, MBA's, executives who seem to beleive the following:
This guy (company) is fabuloulsy successful so their product must be just wonderful
and:
He's this really brilliant programmer / geek and that's the basis of it all
And because these folks haven't got a tech background they're basically taking it on faith. honestly it's insidious, I've seen an entire company (very big one) in a different business say 'wow that's great, lets emulate it ... ohh and yes lets also go with MS in the Data Center! [doh!]. (They fired an MIS director and then a CIO who couldn't make this fine strategy actually work in practice.)
Now what *is* true about Bill (IMO) is that he's really bright (and that his early commercial coding was largely in either assembler or on DEC PDP / Vaxen used for emulation of 8080/z90/x86 systems). Where to my knowlege he applies this is strategy and architecture, and if I don't like his choices, I'm the first to admit they've been effective (if underhanded and illegal) in the market.
Second, for 2 decades MSFT aggressively hired the very best and brightest CS grads. A freind who teaches in one of the better university CS departments observed this and on that basis only started investing in MSFT. That was a very good investment strategy for him :-).
Today I think even the financial types are beginning to realize that some of this is smoke & mirrors. I think the combination of unreasonable licensing changes and the slap on the wrist they just got from SEC are just the sort of thing that these people pay attention to.
Microsoft has always been brought more or less kicking and screaming into standard technologies. netbui vs tcp/ip; WINS vs DNS; NT Domains vs Kerberos|LDAP. Often they have implemented open technology (DCE) in the internals, just not making these the preferred API's.
Of course the whole time I and other opensource types have been looking on and saying *yikes* you want to put this cruft in an enterprise??! MSFT is highly feature driven and lusers love features. Nowhere near enough coders (or architects) work to the priciple that the least code that will do the job is usually the best solution.
Througout, Gates has pushed Basic as the language of choice [shrug]. Gates I don't really want to meet, his original partner Paul Allen, also a billionaire who has said "Blame me for having to type the backslash" ... he doesn't want to meet me in an alley :-)
I will say that I'm glad Gates is focussing on technology again. .NET has promise, and the mono initiative will make it open. His foundation is also giving big money in important areas of medical research and he cares about the right stuff, (e.g. HIV/AIDS).
As the article I read states "software, cash and consulting services", I imagine MSFT will be providing support.
It doesn't always work out that way, I read years ago that in the '70s W.R. Hearst 'gave' a dismantled monastery to the city of San Francisco (probably reaping a nice tax benefit). However as he neglected to provide any funds to assemble the stones, they sat in Golden Gate Park, rain washed off the identifying marks (which goes where) and thus today there's just a pile of stones.
Anyhow, in the automotive industry, my uderstanding is that most of the *profit* is in 'support', auto manufacturerers typically sell the vehicle at or near their manufacturing cost and make their money on parts and service.
In most states of the US (and most developed nations) you are not allowed to operate an automobile without maintaining basic safety (and emissions) equipment. I expect sometime in the near future similar requirements may be made of systems connected to the internet.
Today the conversations may look like: .. I pay for this service and I'm not responsible / can't afford to fix it ...
ISP: Your system is being used for attack by an intruder, if you don't take it offline and get it fixed we will enforce our AUP and take you offline.
customer1: Ooops, sorry ok we'll spend the $$ / time to fix it
customer2:YOU CAN'T DO THAT
ISP: CLICK
Today, while it's feasible to keep systems patched / audited for a reasonable level of safety, many (most?) orgainizations don't have the skillset / funds allocated to keep their systems secure against even the 'kiddies, let alone a determined attacker. That's gonna have to change IMO either thru systems that are harder to break into in the first place or better practices.
Some of the provisions of this bill are also simple clarifications of existing statutes. For instance see the provision: Specify that an existing ban on the "advertisement" of any device that is used primarily for surreptitious electronic surveillance applies to online ads. -- apparently while it's illegal to advertise wiretapping equipment in print, this will extend the restriction to online ads also.
This explains why I've been seeing the adds and spame for keyboard keystroke recorders (shame on you thinkgeek!) and packet sniffers to protect (spy on) your kids or spouse.
More recently I've run short 1.8x on a dual P3 w/ ECC Ram & hardware RAID for a year now, first on kernel 2.2.19, and more recently 2.4.17.
Not a single failure in 12+ months of either snort or the kernel. I suspect that the combination of a marginal backplane and IDE disk, coupled with the older kerenel probably caused the first machine's problems.
The cost analysis was done based on linux, however most of the code analysed in fact is for things that run on other platforms, and much of which was in development for years before linux 0.9 hit the 'Net.
So the measure of value based on who uses Linux includes everyone who uses linux-hosted apache servers. The more general case includes everyone who accesses servers that depend on (Perl, BIND, sendmail, mysql .... etc) or were/are developed using (X11, CVS, bitkeeper, emacs, gcc .... etc)
The economic value isn't small. That much I'm pretty certain of, just how big, well it works for me, I'll leave the analysis to the economists.
So much for other companies trying to raise funding to develop products in this market.
Oft-observed MS behavior:
No offense intended but this looks like an attempt to take rubes (the poster's 26+ karma notwithstanding.
security model is too coarse grained: ... move towards ACLs, for example in NSA's SE Linux, as well as LIDS
Actually SELinux does not implement ACL's, but rather Type Enforcement. It also has potential (and experimental impementation) to implement MLS or other security policies / methods.
What type enforcement gets you is the ability to create highly fine-grained security controls, so that the program and user-security-context have privilege to execute critical functions and that privilege can be removed from the root user.
One of the debian SELinux implementers placed an SELinux system on the 'net with root-password / ssh access advertised. This is not a proof of safety, but in fact noone succeeded in escalating privilege.
As it looks like LSM is on track to be in kernel 2.6, at least the way is presently paved.
As it happens I have a couple of posts today about the what goes on in the OS market, so I'll just link Exchange on nt 3.1? & lessons for OSS and (long) Is Linux Dead. And yes, the market will hand irrelevancy to systems which don't adapt. If GPL does not adapt it will follow the same path, and by my read it has:
Beyond all that, free is not just GPL. FSF used to distribute X11 from the X-Consortium at ~$150 / tape Sometime later RMS decided the X11 license was 'bad'. Today, (perhaps with Debian/Hurd as his ace in the hole?) RMS is trying to push Linux to a strict (activist?) interpretation of GPL.
When I look a the history I think Stallman for all his principles exhibits pragmatism in his actions which is so often attributed to Linus. Linus made it clear long ago that he was not going to give FSF/Deb the blank cheque that many GPL developers do licensing under "gpl-current or whatever later version"
I daresay Linus drew a line in the sand saying "2.0 and no later version of GPL", and I bet if he hadn't we'd be looking at GPL-V3 today.
To be clear I'm not trying to knock either approach. I happen to have a bit more sympathy for Linus's views but that doesn't invalidate the strengths and accrued benefits of other approaches.
Here's my letter to the MSNBC author suggesting different interpretation / conclusions from his analysis.
Dear Mr Schoen,
A few notes on your column regarding the absence of Linux presence (at a recent tech expo in nyc? the context / event was not clearly named).
First I'd like to note different versions of a few of the statistics you present.
The statement that Windows (NT/2k/XP) servers are the largest server installed base: That is probably true if you look at number of licenses or CPUs. However if you look at a more meaningful number like total computing or transaction capacity running under Unix / Linux I think the number rather changes.
Win32 has it's best penetration in small organization and departmental applications. Windows does not scale well past 16 cpu SMP or clusters beyond a few dozen nodes. Therefor virtually all very large servers run Unix, which runs in SMP beyond 64 processors and can aggregate a thousand or more processors into supercomputers or large clusters.
Linux still lags behind the vendor-based unix platforms (IBM, HP, Sun), but is already considerably more capable of high-end scaling than MS's offerings, and is penetrating that market at a pace which reflects it's strengths (and weaknesses).
It is not news that Microsoft has aggressivly targetted both Unix and Linux in the data center. And for all I can see they are not winning anywhere near as often as they would like. IBM has won a number of very large accounts with AIX / Linux in direct competition with MS. This is pointed to in your own statement "Linux server sales jumped by more than 50 percent to $400 million, with IBM leading"
Part of the reason for that is that many Unix shops are able to deploy less-expensive Linux servers where priorly they were locked into proprietary RISC hardware running vendor-based Unix. However they *had* invested in platforms with fundamentally open programming interfaces.
The point here is that these savings are exactly realizable by organizations which had initially invested in open-platform investments. Merril Lynch recently announced a conversion of internal applications from Unix to Linux, citing the lower cost of maintaining applications within the opensource framework of Linux.
The key point here is that while porting applications between proprietary Unix versions (e.g. HP, IBM, Sun) is relatively expensive compared to standardizing on Linux, Windows simply doesn't play in that equation in a meaninful way. Microsoft is working very hard to counter this equation, but the economics don't work in their favor in the data center.
It is simply not meaningful today to compare market penetration from Desktop to Enterprise in either tech or economic terms. The market environments of servers at application, database, web and department are all markedly different. MS has excellentpenetration at the departmental level and of course the Desktop, but they have yet to translate this into serious penetration in any of the other three.
Those of us who do adopt linux on the desktop (I have been using Unix for professional / technical / server computation for 10 years) continue to marvel at the low reliability expectations of Win32 desktop users. Yes things are a lot better than win 3.1 / 95 / 98 days when a networked workstation could not be counted on to stay up for a working day.
NT and Win2k have improved on that record, however they are still a far sight behind Linux in reliability. I run Linux on a late-model laptop (which is probably Linux's toughest challenge, given that virtually no vendors support Linux on these systems, due to the low market share / returns to investment). My experience remains that I have better stability than I get with Win2k Pro
Sure I'm not the average user. My work is either technical or programming and the tools that work best for me are from the same domain as the servers which I run, deploy and support. And I'm not religious about this. As long as end-users have an easier time with proprietary systems they will (and should) stick with them.
However I also know that when the New YorkTimes runs anarticle on the relative compatibility, ease of use and low cost ($50 supported) of Open Office (On win32 or Linux) compared with $600+ for MS Office. I think it's fair to say that Opensource is beginning to competing with MS on the desktop. http://www.nytimes.com/2002/06/20/technology/circu its/20STAT.html
Many observers equate Open Source software with Linux, or the Free Software Foundation, or GPL. In fact the concept of OSS is probably older than Unix but Unix and the original Arpanet / Internet were the substantive proving grounds for these concepts. Today this has paid off in an infrastructure potent enough that it can develop cheap, reliable open systems that also run in the closed platforms.
How the race will play out remains to be seen, however if you are going to write about the subject I think it's best to look at all the elements of the market not just the ones which bolster your preferred outcome.
Home
I couldn't agree more, and I was unaware of the steel and agriculture issues, but trade negotiations in those areas are also pretty standard stuff, you'll not I also indicated my concern:
>> They still have opportunities to snatch defeat from the waiting hands of victory,
haven't even touched upon what this can do to China
Asia today trades more with itself than the west. That is a sign of a long-term shift to a more healthy economy. The only thing I think we can count on is the outcomes aren't easy to guage / guess. The world economy is a complex beast and will pretty well do what it wants. Most of the world understands this today (which is why OPEC hasn't tried to push oil prices to where they hurt the overall economy as happened in the '70s)
The scariest news is that we'll probably come out of this even bigger
I wouldn't bet on that (either way), Asia has good opportunities to realize real growth, but anyhow I'm not an economist / don't play one on tv :-). Dunno what's coming, but I expect it to be interesting
(Linus speaking): moving this (binary drivers with which Stallman / deb take issue) into user space is a sign of mental disorder .... we are clear from a copyright standpoint ... linux has intentionally taken a non-rabid standpoint ... as I've shown with my use of bitkeeper I don't care about black and white people.
[issues about firmware && binary modules]
(Alan Cox?) The kernel developers do not have energy to sit down and determine a clear set of rules ... Debian has an endless supply of people who have nothing better to do than study legal issues....
[Linus points out that actual GPL violating files get addressed in ca 24 hr timeframe]
The conclusion was to send a message back to the Debian users to "put up or shut up"
I'm sure RMS will have a press release out later this week.
I suggest you look up 'Kondratiev cycle technology'. Jay Forrester (inventor of magnetic-core memory), studied this at MIT/Sloan school and determined that the a long-term economic cycle develops due to the 'self-ordering' nature of capital equipment.
Basically Forrester's group found evidence for the following feedback loop: Early in the deployment of any technology there is a scarcity of capital. Capital equipment is expensive, and early investments involve high degrees of risk accompanied by high profits in a given technology sector. That in turn brings investment in the businesses developing this capital. However, a large part of this capital is used in the development of the capital itself (i.e. IT tends to need advanced hardware and software to develop the bleeding-edge new hardware and software for actual end-use).
Thus the 'buildup phase' of new technology creates a high demand (for both the acutal equipment and the stock of the companies that make it). At some point, however the generation of this new (and expensive) equipment (or software) exceeds the actual (end-user) demand. When this happens the high profit margins that were being realized during the build-up phase disappear quite rapidly, the investment-value follows (crashing stock prices) and the investment money looks for other places. See this article
Sound familiar? Whether or not you buy into the economic details, this is one of the behaviors seen in economics. The inflated acquisition prices mentioned are the direct result of this effect.
Because sure people make stupid mistakes even (especially?) with billion dollar transactions. But the funny thing about the stock market is that the money doesn't ever go away. Whenever someone loses in the market, someone else has made a profit.
And yes it sucks when the players break the rules but especially on the financial rules the market punishes you very hard. I worked for a biotech firm that was growing well, showing solid net earnings ca $300m on $2b sales, a 30:1 p/e ratio. Our japan division was found to have been moving inventories to the tune of changing the sales #'s by $50m. This lie, accounting for only 2.5% caused a nearly 50% drop in stock price and a (justified) shareholders lawsuit.
Whenever someone fsck's with the rules of the game (fixing the books, insider trading, breaking anti-trust rules, whatever), real people get hurt and we have SEC, IRS etc to try and keep up with the process. MS imo is an excellent example of how a determined and unscrupulous competitor can harm while evading the systems controls(sic).
I'm just thanking my stars that (so far) the politicians havent fscked up like they did after the '29 stock market crash. The US enacted protectionist trade tarrifs which effectively were the first blow in killing off the *world* economy.
Post-sept-11'th fears and RIAA / DMCA idiocies aside, at least across a several bumpy decades our boneheads in Washington, the EU, etc at least so far have managed not to fsck up. They still have opportunities to snatch defeat from the waiting hands of victory, but so far it could be a whole lot worse I think.