Evoken

OpenSource Software

61 posts in this topic

All these systems were very successful in the evolution of computing and are worthy of praise for their accomplishments at the time.

I agree, as far as that goes - my criticism is that, for what was intended to be a de novo operating system, it was probably too heavily influenced by past architecture, and one that I don't personally see as superior to Unix's (and derivatives) relative, but capable, simplicity. I would bet that much more scientific and engineering work got done (later) on various Unix workstations such as Sun (Solarix Unix), SGI (Irix), HP, etc. workstations than was ever done on DEC machines. If there are any modern supercomputers today that don't consist of many CPUs running a Unix variant, I haven't heard of it (other than a handful of extremely specialized systems for e.g. playing chess, doing quantum chromodynamic calculations in particle physics, etc.)

Share this post


Link to post
Share on other sites
I would bet that much more scientific and engineering work got done (later) on various Unix workstations such as Sun (Solarix Unix), SGI (Irix), HP, etc. workstations than was ever done on DEC machines. If there are any modern supercomputers today that don't consist of many CPUs running a Unix variant, I haven't heard of it (other than a handful of extremely specialized systems for e.g. playing chess, doing quantum chromodynamic calculations in particle physics, etc.)

DEC went out of business primarily because it was too slow to pick up on the trend in workstations and PCs, even though it had them, and was emphasizing them, too late, at the end. Just as VAXes were taking off a new company, Apollo, came out with its own proprietary network OS graphics workstation that showed what was coming, but which were expensive. Ironically, Apollo was started primarily by refugees from Prime Computer, which was also too slow to see the coming workstations as it tried to retain its own minicomputer business attempting to compete with DEC! After Apollo workstations hit the market, Sun then started with its unix based workstations, which were signficantly inferior, but which eventually put Apollo out of business because they were so much cheaper.

Even PCs have long been much faster than the old VAX minicomputers, so it's not surprising that more scientific computing has been done on workstations than was done on VAXs during its relatively shorter lifespan. But at the time, it was a major breakthrough, as was Apollo afterwards even though Apollo did not develop as large a market because of the expensive technology. Unix workstations are still used for serious work, but now PC's can do more than they could. The next development -- a co-processor and flash drive that plug into your head to run unix processes?

Share this post


Link to post
Share on other sites
Unix workstations are still used for serious work, but now PC's can do more than they could. The next development -- a co-processor and flash drive that plug into your head to run unix processes?

I got my start circa. 1978 on the TRS-80, so I was never personally involved with DECs. Certainly, my Intel quad-core CPU with a modern NVidia 3D graphics card can computationally run circles around practically any workstation of just a few years ago, from anybody. For practical reasons it runs Windows XP Pro, though I rather wish it were Mac OS X - but the history of commercial success is certainly not always due to technical superiority (e.g. the hideous 20 bit-with-segmentation architecture of the Intel 8088/86 with asymmetrical, small, register set vs. the power and beauty of the symmetrical MC68000 32 bit architecture, way back when.)

Share this post


Link to post
Share on other sites
I got my start circa. 1978 on the TRS-80, so I was never personally involved with DECs.

VAX/VMS was just becoming commercially available about that time [1] [2], though it was not widespread until a few years later and you would have had to be in the right kind of environment to encounter it directly. (Anyone could have a trash-80 :lol:.) VAX/MS was such a technical and commercial success that it is easy to see why usoft would have hired one of its top architects to work on NT as its new OS using that experience, but that was almost 20 years ago and much has changed since then.

Share this post


Link to post
Share on other sites

I just got finished reading this article on Microsoft's plans to attack linux through patents:

"Microsoft takes on the free world. Microsoft claims that free software like Linux, which runs a big chunk of corporate America, violates 235 of its patents. It wants royalties from distributors and users. Users like you, maybe. Fortune's Roger Parloff reports."

Obviously I am a supporter of intellectual property rights but I am also concerned about Microsoft (or any major public corporation in this Politically Correct world we live in) becoming the sole supplier of something as critical as operating systems.

I know patents are a booming field of litigation (as with the Palm and Blackberry lawsuits) and I have a gut impression that the aggressors in this field are overreaching.

Can anyone comment on whether Microsoft's game plan should be supported or opposed under the premises of Objectivism?

Share this post


Link to post
Share on other sites
Can anyone comment on whether Microsoft's game plan should be supported or opposed under the premises of Objectivism?

Despite my comments on the history of free/open source software (FOSS), I think Microsoft's actions are an abomination. I wrote two posts to HBL about yesterday, which may or may not be posted, but in a nutshell:

1) Software patents should not exist. They're an aberration of American patent law, and not allowed anywhere else in the world for good reason (in this case). They should not exist because they are non-objective and amount to patenting abstract ideas, not concrete inventions. Copyrights and individually issued licenses are an excellent, objective way to protect software IP.

2) Linux and other targeted FOSS's have been in development and in use for many years at this point, and represent a vast and now indispensible resource. If they all had to go away, the internet would literally collapse, and hundreds of millions of embedded devices that use Linux (e.g. TiVo) would be illegal. It would be an economic cataclysm. It is outrageous, even if one grants the validity of software patents, that Microsoft now comes to claim that the backbone of the entire internet and the companies that rely on it, are massively infringing on its purported rights. It isn't as though this progress happened overnight.

Some have postulated that Microsoft's real goal is to get all software patents thrown out. I only wish that were true, because I don't believe it.

Share this post


Link to post
Share on other sites
2) Linux and other targeted FOSS's have been in development and in use for many years at this point, and represent a vast and now indispensible resource. If they all had to go away, the internet would literally collapse, and hundreds of millions of embedded devices that use Linux (e.g. TiVo) would be illegal. It would be an economic cataclysm.

For that reason, that people have relied on FOSS's being free and not subject to patent and Microsoft has waited all this time to press their case, I think the defendants would plead estoppel and win. Any lawyers care to comment?

Share this post


Link to post
Share on other sites
Can anyone comment on whether Microsoft's game plan should be supported or opposed under the premises of Objectivism?

Despite my comments on the history of free/open source software (FOSS), I think Microsoft's actions are an abomination. I wrote two posts to HBL about yesterday, which may or may not be posted, but in a nutshell:

1) Software patents should not exist. They're an aberration of American patent law, and not allowed anywhere else in the world for good reason (in this case). They should not exist because they are non-objective and amount to patenting abstract ideas, not concrete inventions. Copyrights and individually issued licenses are an excellent, objective way to protect software IP.

There are many aspects of software that amount to abstract algorithms similar to theoretical mathematical knowledge. The implementation of specific processess directing the operation of a machine as a specific product is another matter, but what is patentable under that principle must be objectively defined. I haven't seen a detailed description of what it is that usoft is claiming to patent.

2) Linux and other targeted FOSS's have been in development and in use for many years at this point, and represent a vast and now indispensible resource. If they all had to go away, the internet would literally collapse, and hundreds of millions of embedded devices that use Linux (e.g. TiVo) would be illegal.

usoft doesn't want them to go away, it wants to own them. If the royalties usoft demands could not be paid, that is what would happen.

It would be an economic cataclysm. It is outrageous, even if one grants the validity of software patents, that Microsoft now comes to claim that the backbone of the entire internet and the companies that rely on it, are massively infringing on its purported rights. It isn't as though this progress happened overnight.

It would be disruptive and could destroy companies, but not cataclysmic to the economy. It would be similar to bankruptcies such as World Com, in which the infrastructure would continue reorganized under new ownership. But that doesn't mean it is justified. As Phil says, there is a big question over the validity of a company suddenly claiming ownership of key components of an entire industry which have developed over decades -- in many cases several times independently using ideas contextually obvious or well known to those working in the field.

Some have postulated that Microsoft's real goal is to get all software patents thrown out. I only wish that were true, because I don't believe it.

I don't believe that either. But neither does this appear to be a principled defense of property rights by usoft, which has never taken such a stand, even when its existence is threatened by anti-trust. Sadly, it appears that usoft is engaged in a 'pragmatic' legal strategy of 'playing the game' of grabbing as much as it can through legal manipulations in the courts without regard to principled reasons of claiming ownership. Such use of government and litigation in general is something Gates has openly regarded as the way 'business is done' and is to be 'expected from competitors' -- including when it is done to usoft by its competitors in vicious anti-trust cases.

usoft might, however not expect to win, only seeking to force the issue to be resolved one way or the other or somewhere in the 'middle' because it is also endlessly being held up for 'royalties' in shakedown rackets by others seeking to be bought off in 'settlements' to get them to go away. But whatever happens in the courts, usoft is apparently trying to 'play the game' of getting as much out of it as it can, pragmatically without regard to principle of what it is legitimately entitled to. I'm sure there are usoft employees who take a more principled approach, but that appears to be the corporate strategy. The legal system may be in such a mess that they have no choice but to pursue that in the world as it is, but we see no principled objections to it from them either.

Share this post


Link to post
Share on other sites
There are many aspects of software that amount to abstract algorithms similar to theoretical mathematical knowledge.

Yes, but those abstract algorithms must be concretely implemented in particular ways in order to form a real working computer program (as I'm sure you know.) I argue that it is only those implementations which should be protectable. In the same way, it would be improper to permit patenting mathematical principles or laws of physics or engineering principles. I'm not sure whether you agree with that or not; if you believe that software patents are legitimate I would be interested in hearing your reasons.

Share this post


Link to post
Share on other sites
1) Software patents should not exist. They're an aberration of American patent law, and not allowed anywhere else in the world for good reason (in this case). They should not exist because they are non-objective and amount to patenting abstract ideas, not concrete inventions. Copyrights and individually issued licenses are an excellent, objective way to protect software IP.

I'll have to think more about this, but offhand, I tend to agree with you - or at least I'd say that software patents have been issued far too readily. The software patents that I have seen always seem to be so general that if taken literally, they would outlaw lots of software. (E.g., I once saw one that sounded like it would cover recursive-descent parsing of programming languages.)

I'm not sure, but I think that companies sometimes just try to get as many software patents, as broadly construed, as they can, perhaps to use as "bargaining chips," in case they ever had to counterattack against another company that accuses them of infringement. (For instance, company A is accused of infringing company B's software patent. In response, company A drags out a software patent that they happen to have, and accuses company B of infringing its patent, probably hoping that company B will agree to just dismiss the reciprocal suits and call it a day.)

As for what Microsoft is up to, it's probably just their usual pragmatism. Or perhaps they think that everything of significance in the world of software was invented by Microsoft (I've known more than one Microsoft employee who has expressed that sentiment), so they of course ought to be able to collect royalties on it. (As a Microsoft employee said to me over a decade ago "We don't really want to control the world; we just want people to put quarters in.")

Share this post


Link to post
Share on other sites