It may take a relatively meager amount of brains to be a lawyer (that is, write your bar exams), but it takes considerable intellectual ingenuity to be a good lawyer - especially a litigator.
Of course, I could say the same thing about programming as opposed to analysis or computer science. Most geeks are intellectual prima donnas - of course, I'm a geek, too, and I can hardly claim innocence.
The problem with law is that it is rarely clean cut. Both sides bring convincing (or at least not ludicrous) arguments to the table, and then to trial. Both are consistent, even in matters of new law, with existing statute and case law. Geeks, who tend to be more mathematically and technically oriented, rarely find themselves in such conundrums - where both sides are "right" (or at least as right as possible) but one must win despite that.
Very few laws are made with the cynical knowledge that they will be immediately overturned. The CDA, of course, was one of those - a law written into the books entirely for political brownie points. However, the Supreme Court of any nation may have a different idea of the relative importance of different priorities, or different applications of law. It's inevitable that many laws be overturned on constitutional grounds (both in the US and in Canada) in such a situation. This is a sign that the system works, not that it's broken.
Hard to deal with, of course, if you're a geek and given to common sense. Common sense, however, has nothing to do with the law; what the law seeks is internal consistency. The eventual implications of any individual case rarely matter.
This isn't to say I disagree with you that geeks and lawyers have different mind-sets. I agree wholeheartedly - look at any psychometric study of professions, for example. But to demean the character of lawyers is unnecessary. Shakespeare might have said "First, kill all the lawyers", but he meant that as being the first step towards anarchy. Lawyers may be slimy characters in general (and I know quite a few) but the wheels of any state need ample lubrication with such slime.:)
A salesperson isn't in control of their environment; the people whom you visit are. I'm sure OSHA visits their offices to repond to complaints, just like they visit your own.
The parts of your travelling menagerie that are under your employer's control (such as your laptop, sales props, and the likes) are presumably subject to OSHA control. If you get carpal tunnel from your laptop, your employer is going to have to pay for rehabilitation.
Working at home isn't a perk; it's work. A perk is secondary to your job function; working is your job function. Your boss still retains the responsibility for what happens there. Your employer couldn't send you into a coal mine and then disavow responsibility when you come down with black lung. Even if you asked to do sales calls in a coal mine, they'd still be responsible.
I have no idea what motivates your employer, but consider this: the cost of a good desk, chair, computer and network connection is a small fraction of the salary of a good technical employee. The main expenditure of an IT organization is for skilled personnel, not hardware or software.
If you produce significantly more by being able to work at home, it's easy to make a strong business case for that. On the other hand, if it's merely a matter of personal convenience, then don't expect an enthusiastic response; they aren't in business for your convenience, after all.
I suspect, although I have no personal knowledge in this instance, that they're simply leery about having people work at home, away from teams and/or management. Perhaps they have been burned in the past in this respect; perhaps they're merely conservative and slow to change. Or they're authoritarian. There are any number of reasons.
My own firm brings consulting work from across North America to our home base in Toronto for that reason, and very few of us work onsite (and even then only for as long as necessary); we can work from home at our own discretion, but that defeats the purpose of having every member of a team in one geographical location. As for me, I only work from home when I have to; working with a team is far more productive.
If they own the machine that you use at home, then they're entitled to recover it and examine it at any time. My own company has this regulation spelled out in their policy - and it's just common sense, really. I use my own computer, for obvious reasons.
They, theoretically, could get a court order to examine my computers at home - but they could only do that if they suspected I'd stolen company property and it was located there. Keep a log of what you're working on for every minute of your work at home, and make it clear to your manager what you will do and what you did do for offsite work. For that matter, keep a log of what you're working on for your work at work, too.
If they're paying for your line, fast or slow, it stands to reason that at one point they could examine what you're downloading using it. So don't let them pay for your connection; if you really want to skirt the edge of it, charge them for the use of it, pro rated on an hourly basis. But, really, don't do that if you don't want to give them an expense to snoop. I have my own cable modem connection, and use an encrypted connection to work when I want to work. If I want to download pornography, I'm obviously not going to route it through there.
If they buy you furniture, it's their property. The cushy leather ergonomic executive chair goes back to the company when you leave them. Likewise for anything that you've expensed. I know that geeks love to accumulate geek books, and if it's at their employer's expense so much the better (and I've done that too) but right now if I want a book I pay for it myself; if I think that the company needs a book for my work, on the other hand, I buy it, expense it, and turn it over to the company's library when I'm done with it.
This is called simple professionalism and common sense. That it would have to be codified is guaranteed employment for lawyers; it also boggles the imagination that an explanation is even needed.
I think that this was inevitable; it revisits the distinction between a contractor/consultant and an employee.
An employee works as an extension of the employer. They are required to perform at an adequate, previously discussed level but are paid on a time (even salaried workers - they merely have a larger time increment than hourly) rather than work performed basis.
A consultant performs work for or on the behalf of a company, but they are professionals who work for themselves. The work produced is the basis for compensation, not the length of time which it takes to produce said work. (I know about time and materials contracts, btw, but the point is that you're compensated for the work you produce).
An employee, who is paid for time, therefore, is entitled to a proper working environment during that time. That is, wherever the work is performed. I haven't heard that occupational health and safety regulations were automatically suspended when an employee is out making a client call, for example. Why should they be suspended merely because the employee is working at a different but still acceptable location (in this case, his or her home)? The essence of this is that the home is an acceptable location to the employer; when they sanction work at home they acquire the responsibility to provide a proper work environment there, just like they couldn't send you down a mineshaft without a helmet.
A consultant produces work, and the contractual nature of the work means that they must provide for themselves or their subcontractors. They aren't subject to most occupational health and safety rules because of the nature of that employment - or, at least, the company tendering the contract is not responsible for conformation to those rules. The company employing the individual (in many cases themself or a small consultancy) is.
To sum this up: employees don't stop being employees just because they're working at a different site. The act of approving the home as a suitable worksite by a company makes them responsible for occupational injury there, just like it was any other jobsite - because it is just that, any other jobsite. They don't turn into ersatz consultants when they work from home, and employers which treat them like that will no doubt discover the error of their ways.
It does give them a reference distribution, which they control, on which to base their porting efforts. There's nothing wrong with that; in fact, it's quite shrewd of them. They'll probably lose buckets of money on the distribution proper, but they're probably planning to make it back and more on applications software.
As for basing it on Debian, that the core of Debian is 100% free software means that they won't have to worry about licensing issues looming overhead in the future. Dpkg/apt is arguably the best package management tool presently avaliable and Debian has an excellent QA process to boot. They could put together their own distribution and their own package management format, but there's no advantage in doing that.
Consider it as a business case and it becomes quite clear. There's no advantage in reinventing the wheel merely to get into the linux market, but you still have to have some control over where the wheel goes.
The problem really isn't the greed of corporations; it's their litigiousness. Network Solutions Inc. has been the proverbial jellyfish in response to domain name challenges in the past, but that's mostly because it has no motivation to challenge court orders sought by corporations - really, it has no desire to even stand up to the threat of a court order. If you want to change the system, make it a lot harder for a corporation to get a cease and desist order on the use of a domain name. The courts should consider an active domain name to be a de facto business address, which it is, rather than a promotional widget, which it isn't.
(If it sounds like I have no patience for gold-digging cybersquatters, you'd be entirely correct).
I wrote a slashdot post recently on the foolishness of letting NSI arbitrate valuable domain names with no legislative muscle behind it. Click on my user info and look at the article entitled "Capital Wins, News at 11" if you'd like to read it.
Not to be too harsh, but almost everyone in this discussion is confusing application development with equity markets.
Lots of great applications - technically proficient, even brilliant - were dogs when it came to sales.
Companies which marketed them lost money, went belly-up. IBM lost five billion dollars in a single fiscal year - more money than most third world countries take in - while holding 10% of the US's patents and an immense share of R&D expenditures.
Linux will continue to be a fine operating system. I'm quite sure of it (now, criticise *my* almost wholly uninformed guess:) ). Stock prices will take a header, and this is an slightly more informed guess. But Linux doesn't depend on high stock prices to continue the pace development; individual companies do. If they go bankrupt, c'est la vie.
How many software companies from ten years ago are still in business, anyway? It's the nature of the high-tech industry: live fast, die young, and leave a pretty corpse.
Linux has been over-hyped somewhat in the past year. By over-hyped I mean that valuations of Linux-related stock has far exceeded revenue. Fair enough - it has, by huge margins. Most likely this overvaluation is exacerbated by the paucity of Linux stock out there; most of it remains in the hands of their directors. This is endemic to the high-tech industry, of course, but it does mean that when they do cash in (hello, ESR) stock values will plunge.
I work with a number of business analysts (I'm not one, but they make for good lunch-time conversation), and they've come to the same conclusion: Linux doesn't offer strong enough added value to induce a CIO to switch corporate desktops outright. On the server side, it may well, but the majority of OS licenses are sold on desktop computers rather than servers (good thing that Linux isn't in the licensing game, no?).
In any case. While companies may come and go, and I fully expect at least a couple of Linux pioneers to fold in the new year, it's important to remind ourselves that no company has a monopoly on Linux. If Red Hat should fold, it would be a tragedy to lose so many talented developers who would have to work elsewhere for their suppers, but it would not be the end of Linux as an operating system. So long as people contribute to it, Linux as a phenomenon remains vital.
We all may be a bit sadder for a crash in stock values, and some of us much poorer, but it's nothing unexpected and nothing to worry about.
A U*x admin that has managed to avoid reboot on a production system for so long most likely has not left any known or even supsected h0lez for R00ting...
I have no idea how any admin would do that when it comes to kernel vulnerabilities. Crystal ball, perhaps? Most exploit principles are not new, but the techniques certainly are.
Most holes are at the application level, which is all well and good, and as it should be. They can be easily fixed without a reboot. Exploits of kernel services (tcp stack, for example), require patching and rebooting, and they're not unknown - not to mention that fixing them is generally outside the job description and specialized knowledge of a sysadmin.
Now, it's possible to secure your boxes from outside attack through firewalls and the like. Perhaps even to the point that the weakness is unexploitable from the outside. The feeling following such work is known as hubris, the pride that goeth before the fall. Are you willing to make the claim of total security for every machine within your network, or every employee you have or have had? All these things are potential vulnerabilities; a machine secured to the outside, as you can see, isn't really secure at all.
I would like to see more advanced statistical data with respect to this. I would suspect that the average uptime of a linux box is a bimodal distribution, with hobbyists representing the first, larger node with shorter uptime, and professional administrators representing most of the second, smaller node with longer uptime. The first node would dominate the other, and the arithmetical average and median would be pulled towards it through no fault in the operating system.
It is interesting, however. I can see the smaller node being constrained by the release of security fixes (and I wonder about the very long life FreeBSD boxes as well). I do think, however, that the best purpose of an uptimes study would be to find artificial constraints on uptimes, as respresented by abnormal distributions.
This could have pointed out the 49-day limit in MS operating systems well in advance of it being reported by MS proper, for example. If a kernel bug in Linux (or any other operating system) caused frequent crashes over time, it would reveal itself in the distribution of uptimes. Like I said above, the point should be to improve our (and by "our" I mean everyone's, from BSD to Linux to Windows) OS reliability, not merely to dick-size about ludicrously long uptimes.
Perhaps this calls for a more advanced massaging of data from the uptimes people:). I wouldn't mind helping. It's a project well worth the effort.
--
Capital wins, news at 11
on
Etoy Update
·
· Score: 3
When domain names started having intrinsic value, rather than just being pointing to locations, then this kind of kerfuffle became inevitable. NSI, having little power of its own and no stomach (much less cash reserve) for a legal fight, will inevitably cave to the stronger party, and that's not what we need in a name service administrator.
If you'd like an analogue, imagine the US Patent and Trademark Office awarding ownership of any given patent or trademark to the strongest comer. Anyone can see how this would become disastrous.
The above analogue wouldn't happen, of course, because the Patent and Trademark Office is a government agency with the rest of the federal government behind it (note that this age may have changed, now that large corporations are lobbying to have funding slashed to agencies which prosecute them). You just don't fight city hall without a good reason and deep pockets. Fighting an independent NSI, on the other hand, is incredibly trivial.
I suggest that legal ownership of NSI's name service databases be recovered by the NSF or another agency of the US government (or better yet the UN), while the business of selling registrations and administering the databases remains in private hands. The point is that ICANN authorized name service vendors would become service providers rather than posessing goods (domain names) which are valuable in their own right and which they have next to no interest in protecting or ability to protect.
This isn't to say that the government would do a better job, or would be less of a jellyfish than NSI was. However, I'd lay good odds that it would at the very least appear much less powerless. This would be enough, I think, to discourage much petty banditry surrounding the issue of domain names.
I might add that a code base, even if it is not used in a profitable product, still has a certain book value. This keeps most companies from releasing their code to the public; they don't want to take the book loss (on the other hand, profitable companies might want to do this with their unprofitable product lines; it would cause a book loss which would reduce their profit and hence their corporate taxes). The code existing in the old product can be reworked into a new one - even if that possibility is very slim.
This may seem contrarian, but the GPL has some significant benefits for a corporation releasing its code to the public. The corporation retains control of the original codebase, and can spin it into a new product without adverse repercussions. The corporation infects the released code with a viral license, and prevents it from being used in a competing product. These are not minor advantages; they may make the difference between releasing and not.
We should promote more of this practical and pragmatic thinking among software corporations.:) What better way to keep one's product lines alive? It was a revolutionary act when Netscape did it to Navigator (which it never made a profit on; NN was a dog in every sense of the word); now, it's no longer farsighted. It just makes sense.
Is that necessarily true? Remember that three of the four main BSD's remain freely available and code is freely shared between them. A number of OpenBSD architectures are based on ports to those architectures made by the NetBSD team. It would be easy to create a "main" BSD distribution by folding changes made in any of these back in.
I think that you can attribute the forking not to commercial gain or a need to keep changes secret, but to different directions in development philosophy. FreeBSD concentrates on x86 development, NetBSD on maintaining the maximum number of available architectures, and OpenBSD (my personal favourite) on security.
It's not altogether inconceivable that Linux might fork the same way in the future. Remember that Linux is just a kid compared to *BSD, even though it's outpaced *BSD in recent years.
On the contrary. The most important part is to determine why he was right, why he thought that, what did he mean when he said it. What comes last and least important is what it might mean to us.
To do otherwise is to reduce us to a million monkeys on a million typewriters. If I say something profound by accident, it isn't nearly as profound - in fact, it's absolutely vacant.
What Jefferson said at any one point, good quote or not, is made important by his involvement in world affairs, in the intellectual life of colonial Virginia and the early United States, his time in France. Not to mention his Presidency and founding of the University of Virginia. Jefferson was a very ambitious and astute politician who made good on his potential and that's why he's important; he didn't just make good quote.
No, the same programs don't work (or don't even compile) on all of them. There are library incompatibilities, for one. BSD has gone farthest by releasing a user-space linux emulator.
I haven't had trouble compiling. However, I was referring to binary compatibility, not source compatibility. (Of course, there's someone else in this thread who insists that emulation equates compatibility, so perhaps it's possible to go too far the other way).
For that matter, I really haven't had that much difficulty porting stuff from BSD to Linux, either. Maybe I *am* lucky, or I just haven't tried anything really really difficult yet.
Larry's keeping tight hold of perl (as perl; you're free to reuse anything to do with perl so long as you don't call it perl) for the sake of perl. He trusts his own judgement, in other words, but he's always amenable to new arguments.
Sun, on the other hand, is keeping tight hold of Java as a revenue centre. The good of the application as a functional and usable piece of work comes second, or sometimes not at all, if I can borrow from Austin Powers;-). I don't think that they'd incorporate architectural changes which come from outside their core group (it seems that they've rejected all proposals so far) and the SCSL makes a mockery of open source.
I would feel hurt, both emotionally and professionally, if I'd taken Sun's goodwill at face value, like the Blackdown team did. Despite their occasional gesture towards open source, they're Just Another Corporation at heart.
They run them through binary emulation. By the same token, you could say that Linux on x86 is compatible with old atari systems, merely because an emulator exists. It's clearly not so.
BSD advocates on a public discussion forum are all well and good, but look at the documentation, even the most basic documentation, before you make an assertion. From the OpenBSD FAQ:
1.1 - What is OpenBSD?
The [7]OpenBSD project produces a freely available, multi-platform 4.4BSD-based UNIX-like operating system. Our efforts place emphasis on portability, standardization, correctness, and security. [8]OpenBSD supports binary emulation of most binaries from SVR4 (Solaris),FreeBSD, Linux, BSDI, SunOS, and HPUX.
I think that the point is that most users don't demand compatibility - systems analysts and other such pedantic characters may desire it and even write it in, but when you're dealing with joe developer, much less joe user, thoughts of compatibility are relatively irrelevant. It's almost always a matter of price-performance and relevance to the business case.
Don't make Jefferson out as a proto-libertarian, which you're tending towards. He most certainly wasn't. He was a revolutionary and a political liberal, but intellectually he remained a child of the scottish enlightenment. His rejection of british conceptions of liberalism was a political, rather than a philosophical, move.
RMS does much better when he describes the philosophical and technical issues behind his positions. As it is, this article is full of unsupported assertions, questionable facts, and general gobbledygook.
For example, the assertion that "putting users in control" (that is, opening the source, preferably under the GPL) is the best way to assure continued compatibility. I am no philosopher of science, but it seems to me that putting users in control and allowing code forking is to encourage incompatibility. Witness the fate of the BSD's - originally compatible due to their shared code base in BSD4.4, presently incompatible due to different directions in development. And that over a relatively short time span.
Another example, his trust in the market to favour a Java-compatible implementation. As such things go, I doubt that the market would favour any such thing. The most popular pieces of software, word processors and web browsers, are perpetually crippled with respect to backwards and forwards compatibility, much less interoperability. What the market favours is price, availability, and support. Compatibility, it seems, has been left by the wayside. This may be short-sighted, and I believe it is, but RMS may be giving us, as an industry, much too much credit here.
I can only hope that he didn't intend this piece of correspondence for publication. As is the case with many visionaries (and I have no doubt that he is a visionary), he needs to tone down his message to sell it to the masses. Right now, I doubt that anyone is any more sold on open source, and a golden opportunity wasted.
Don't forget that many anti-communist agitators in eastern europe were priests. Several were killed for such activity in Poland alone; the pope himself only barely escaped that sad fate.
That said, anti-communism strikes me as a one-note philosophy, lacking a connection to a larger whole or principle. Which isn't to say that anticommunists are unprincipled, merely that the anti-communist movement as a whole is such an incredibly mixed bag that we can use it as proof of almost nothing - except for anti-communism itself, of course. It spans the breadth of political thought, from die-hard leftist intellectuals (Orwell) to totalitarian dictators of the worst order (Pinochet).
When selecting a man of the century, one should select a man who exemplifies the thought of the century. Should we select a man who exemplifies only a single thought - perhaps a man who only had a single thought? If we are to select a laughingstock, then we should abandon all pretense and seek out the greatest laughingstock available.
If we are to seek out a great man, on the other hand, then someone who thought great thoughts and performed great deeds (that is, someone who wasn't along for the ride of history) is as good a choice as you could possibly make. Marshall is an entirely apt choice in this respect.
Slashdot Mirror
It may take a relatively meager amount of brains to be a lawyer (that is, write your bar exams), but it takes considerable intellectual ingenuity to be a good lawyer - especially a litigator.
:)
Of course, I could say the same thing about programming as opposed to analysis or computer science. Most geeks are intellectual prima donnas - of course, I'm a geek, too, and I can hardly claim innocence.
The problem with law is that it is rarely clean cut. Both sides bring convincing (or at least not ludicrous) arguments to the table, and then to trial. Both are consistent, even in matters of new law, with existing statute and case law. Geeks, who tend to be more mathematically and technically oriented, rarely find themselves in such conundrums - where both sides are "right" (or at least as right as possible) but one must win despite that.
Very few laws are made with the cynical knowledge that they will be immediately overturned. The CDA, of course, was one of those - a law written into the books entirely for political brownie points. However, the Supreme Court of any nation may have a different idea of the relative importance of different priorities, or different applications of law. It's inevitable that many laws be overturned on constitutional grounds (both in the US and in Canada) in such a situation. This is a sign that the system works, not that it's broken.
Hard to deal with, of course, if you're a geek and given to common sense. Common sense, however, has nothing to do with the law; what the law seeks is internal consistency. The eventual implications of any individual case rarely matter.
This isn't to say I disagree with you that geeks and lawyers have different mind-sets. I agree wholeheartedly - look at any psychometric study of professions, for example. But to demean the character of lawyers is unnecessary. Shakespeare might have said "First, kill all the lawyers", but he meant that as being the first step towards anarchy. Lawyers may be slimy characters in general (and I know quite a few) but the wheels of any state need ample lubrication with such slime.
--
A salesperson isn't in control of their environment; the people whom you visit are. I'm sure OSHA visits their offices to repond to complaints, just like they visit your own.
The parts of your travelling menagerie that are under your employer's control (such as your laptop, sales props, and the likes) are presumably subject to OSHA control. If you get carpal tunnel from your laptop, your employer is going to have to pay for rehabilitation.
Working at home isn't a perk; it's work. A perk is secondary to your job function; working is your job function. Your boss still retains the responsibility for what happens there. Your employer couldn't send you into a coal mine and then disavow responsibility when you come down with black lung. Even if you asked to do sales calls in a coal mine, they'd still be responsible.
--
I have no idea what motivates your employer, but consider this: the cost of a good desk, chair, computer and network connection is a small fraction of the salary of a good technical employee. The main expenditure of an IT organization is for skilled personnel, not hardware or software.
If you produce significantly more by being able to work at home, it's easy to make a strong business case for that. On the other hand, if it's merely a matter of personal convenience, then don't expect an enthusiastic response; they aren't in business for your convenience, after all.
I suspect, although I have no personal knowledge in this instance, that they're simply leery about having people work at home, away from teams and/or management. Perhaps they have been burned in the past in this respect; perhaps they're merely conservative and slow to change. Or they're authoritarian. There are any number of reasons.
My own firm brings consulting work from across North America to our home base in Toronto for that reason, and very few of us work onsite (and even then only for as long as necessary); we can work from home at our own discretion, but that defeats the purpose of having every member of a team in one geographical location. As for me, I only work from home when I have to; working with a team is far more productive.
--
Doesn't your dorm room contract prevent you from running a business from it? ;-). Mine did, at least, although that was nearly ten years ago.
--
If they own the machine that you use at home, then they're entitled to recover it and examine it at any time. My own company has this regulation spelled out in their policy - and it's just common sense, really. I use my own computer, for obvious reasons.
They, theoretically, could get a court order to examine my computers at home - but they could only do that if they suspected I'd stolen company property and it was located there. Keep a log of what you're working on for every minute of your work at home, and make it clear to your manager what you will do and what you did do for offsite work. For that matter, keep a log of what you're working on for your work at work, too.
If they're paying for your line, fast or slow, it stands to reason that at one point they could examine what you're downloading using it. So don't let them pay for your connection; if you really want to skirt the edge of it, charge them for the use of it, pro rated on an hourly basis. But, really, don't do that if you don't want to give them an expense to snoop. I have my own cable modem connection, and use an encrypted connection to work when I want to work. If I want to download pornography, I'm obviously not going to route it through there.
If they buy you furniture, it's their property. The cushy leather ergonomic executive chair goes back to the company when you leave them. Likewise for anything that you've expensed. I know that geeks love to accumulate geek books, and if it's at their employer's expense so much the better (and I've done that too) but right now if I want a book I pay for it myself; if I think that the company needs a book for my work, on the other hand, I buy it, expense it, and turn it over to the company's library when I'm done with it.
This is called simple professionalism and common sense. That it would have to be codified is guaranteed employment for lawyers; it also boggles the imagination that an explanation is even needed.
--
I think that this was inevitable; it revisits the distinction between a contractor/consultant and an employee.
An employee works as an extension of the employer. They are required to perform at an adequate, previously discussed level but are paid on a time (even salaried workers - they merely have a larger time increment than hourly) rather than work performed basis.
A consultant performs work for or on the behalf of a company, but they are professionals who work for themselves. The work produced is the basis for compensation, not the length of time which it takes to produce said work. (I know about time and materials contracts, btw, but the point is that you're compensated for the work you produce).
An employee, who is paid for time, therefore, is entitled to a proper working environment during that time. That is, wherever the work is performed. I haven't heard that occupational health and safety regulations were automatically suspended when an employee is out making a client call, for example. Why should they be suspended merely because the employee is working at a different but still acceptable location (in this case, his or her home)? The essence of this is that the home is an acceptable location to the employer; when they sanction work at home they acquire the responsibility to provide a proper work environment there, just like they couldn't send you down a mineshaft without a helmet.
A consultant produces work, and the contractual nature of the work means that they must provide for themselves or their subcontractors. They aren't subject to most occupational health and safety rules because of the nature of that employment - or, at least, the company tendering the contract is not responsible for conformation to those rules. The company employing the individual (in many cases themself or a small consultancy) is.
To sum this up: employees don't stop being employees just because they're working at a different site. The act of approving the home as a suitable worksite by a company makes them responsible for occupational injury there, just like it was any other jobsite - because it is just that, any other jobsite. They don't turn into ersatz consultants when they work from home, and employers which treat them like that will no doubt discover the error of their ways.
--
It does give them a reference distribution, which they control, on which to base their porting efforts. There's nothing wrong with that; in fact, it's quite shrewd of them. They'll probably lose buckets of money on the distribution proper, but they're probably planning to make it back and more on applications software.
As for basing it on Debian, that the core of Debian is 100% free software means that they won't have to worry about licensing issues looming overhead in the future. Dpkg/apt is arguably the best package management tool presently avaliable and Debian has an excellent QA process to boot. They could put together their own distribution and their own package management format, but there's no advantage in doing that.
Consider it as a business case and it becomes quite clear. There's no advantage in reinventing the wheel merely to get into the linux market, but you still have to have some control over where the wheel goes.
--
The problem really isn't the greed of corporations; it's their litigiousness. Network Solutions Inc. has been the proverbial jellyfish in response to domain name challenges in the past, but that's mostly because it has no motivation to challenge court orders sought by corporations - really, it has no desire to even stand up to the threat of a court order. If you want to change the system, make it a lot harder for a corporation to get a cease and desist order on the use of a domain name. The courts should consider an active domain name to be a de facto business address, which it is, rather than a promotional widget, which it isn't.
(If it sounds like I have no patience for gold-digging cybersquatters, you'd be entirely correct).
I wrote a slashdot post recently on the foolishness of letting NSI arbitrate valuable domain names with no legislative muscle behind it. Click on my user info and look at the article entitled "Capital Wins, News at 11" if you'd like to read it.
--
(see above)
--
Not to be too harsh, but almost everyone in this discussion is confusing application development with equity markets.
:) ). Stock prices will take a header, and this is an slightly more informed guess. But Linux doesn't depend on high stock prices to continue the pace development; individual companies do. If they go bankrupt, c'est la vie.
Lots of great applications - technically proficient, even brilliant - were dogs when it came to sales.
Companies which marketed them lost money, went belly-up. IBM lost five billion dollars in a single fiscal year - more money than most third world countries take in - while holding 10% of the US's patents and an immense share of R&D expenditures.
Linux will continue to be a fine operating system. I'm quite sure of it (now, criticise *my* almost wholly uninformed guess
How many software companies from ten years ago are still in business, anyway? It's the nature of the high-tech industry: live fast, die young, and leave a pretty corpse.
--
Strangely enough, I agree. Sort of.
Linux has been over-hyped somewhat in the past year. By over-hyped I mean that valuations of Linux-related stock has far exceeded revenue. Fair enough - it has, by huge margins. Most likely this overvaluation is exacerbated by the paucity of Linux stock out there; most of it remains in the hands of their directors. This is endemic to the high-tech industry, of course, but it does mean that when they do cash in (hello, ESR) stock values will plunge.
I work with a number of business analysts (I'm not one, but they make for good lunch-time conversation), and they've come to the same conclusion: Linux doesn't offer strong enough added value to induce a CIO to switch corporate desktops outright. On the server side, it may well, but the majority of OS licenses are sold on desktop computers rather than servers (good thing that Linux isn't in the licensing game, no?).
In any case. While companies may come and go, and I fully expect at least a couple of Linux pioneers to fold in the new year, it's important to remind ourselves that no company has a monopoly on Linux. If Red Hat should fold, it would be a tragedy to lose so many talented developers who would have to work elsewhere for their suppers, but it would not be the end of Linux as an operating system. So long as people contribute to it, Linux as a phenomenon remains vital.
We all may be a bit sadder for a crash in stock values, and some of us much poorer, but it's nothing unexpected and nothing to worry about.
Hysteria indeed.
--
A U*x admin that has managed to avoid reboot on a production system for so long most likely has not left any known or even supsected h0lez for R00ting...
I have no idea how any admin would do that when it comes to kernel vulnerabilities. Crystal ball, perhaps? Most exploit principles are not new, but the techniques certainly are.
Most holes are at the application level, which is all well and good, and as it should be. They can be easily fixed without a reboot. Exploits of kernel services (tcp stack, for example), require patching and rebooting, and they're not unknown - not to mention that fixing them is generally outside the job description and specialized knowledge of a sysadmin.
Now, it's possible to secure your boxes from outside attack through firewalls and the like. Perhaps even to the point that the weakness is unexploitable from the outside. The feeling following such work is known as hubris, the pride that goeth before the fall. Are you willing to make the claim of total security for every machine within your network, or every employee you have or have had? All these things are potential vulnerabilities; a machine secured to the outside, as you can see, isn't really secure at all.
--
Very good point.
:). I wouldn't mind helping. It's a project well worth the effort.
I would like to see more advanced statistical data with respect to this. I would suspect that the average uptime of a linux box is a bimodal distribution, with hobbyists representing the first, larger node with shorter uptime, and professional administrators representing most of the second, smaller node with longer uptime. The first node would dominate the other, and the arithmetical average and median would be pulled towards it through no fault in the operating system.
It is interesting, however. I can see the smaller node being constrained by the release of security fixes (and I wonder about the very long life FreeBSD boxes as well). I do think, however, that the best purpose of an uptimes study would be to find artificial constraints on uptimes, as respresented by abnormal distributions.
This could have pointed out the 49-day limit in MS operating systems well in advance of it being reported by MS proper, for example. If a kernel bug in Linux (or any other operating system) caused frequent crashes over time, it would reveal itself in the distribution of uptimes. Like I said above, the point should be to improve our (and by "our" I mean everyone's, from BSD to Linux to Windows) OS reliability, not merely to dick-size about ludicrously long uptimes.
Perhaps this calls for a more advanced massaging of data from the uptimes people
--
When domain names started having intrinsic value, rather than just being pointing to locations, then this kind of kerfuffle became inevitable. NSI, having little power of its own and no stomach (much less cash reserve) for a legal fight, will inevitably cave to the stronger party, and that's not what we need in a name service administrator.
If you'd like an analogue, imagine the US Patent and Trademark Office awarding ownership of any given patent or trademark to the strongest comer. Anyone can see how this would become disastrous.
The above analogue wouldn't happen, of course, because the Patent and Trademark Office is a government agency with the rest of the federal government behind it (note that this age may have changed, now that large corporations are lobbying to have funding slashed to agencies which prosecute them). You just don't fight city hall without a good reason and deep pockets. Fighting an independent NSI, on the other hand, is incredibly trivial.
I suggest that legal ownership of NSI's name service databases be recovered by the NSF or another agency of the US government (or better yet the UN), while the business of selling registrations and administering the databases remains in private hands. The point is that ICANN authorized name service vendors would become service providers rather than posessing goods (domain names) which are valuable in their own right and which they have next to no interest in protecting or ability to protect.
This isn't to say that the government would do a better job, or would be less of a jellyfish than NSI was. However, I'd lay good odds that it would at the very least appear much less powerless. This would be enough, I think, to discourage much petty banditry surrounding the issue of domain names.
--
Excellent point.
:) What better way to keep one's product lines alive? It was a revolutionary act when Netscape did it to Navigator (which it never made a profit on; NN was a dog in every sense of the word); now, it's no longer farsighted. It just makes sense.
I might add that a code base, even if it is not used in a profitable product, still has a certain book value. This keeps most companies from releasing their code to the public; they don't want to take the book loss (on the other hand, profitable companies might want to do this with their unprofitable product lines; it would cause a book loss which would reduce their profit and hence their corporate taxes). The code existing in the old product can be reworked into a new one - even if that possibility is very slim.
This may seem contrarian, but the GPL has some significant benefits for a corporation releasing its code to the public. The corporation retains control of the original codebase, and can spin it into a new product without adverse repercussions. The corporation infects the released code with a viral license, and prevents it from being used in a competing product. These are not minor advantages; they may make the difference between releasing and not.
We should promote more of this practical and pragmatic thinking among software corporations.
--
Is that necessarily true? Remember that three of the four main BSD's remain freely available and code is freely shared between them. A number of OpenBSD architectures are based on ports to those architectures made by the NetBSD team. It would be easy to create a "main" BSD distribution by folding changes made in any of these back in.
I think that you can attribute the forking not to commercial gain or a need to keep changes secret, but to different directions in development philosophy. FreeBSD concentrates on x86 development, NetBSD on maintaining the maximum number of available architectures, and OpenBSD (my personal favourite) on security.
It's not altogether inconceivable that Linux might fork the same way in the future. Remember that Linux is just a kid compared to *BSD, even though it's outpaced *BSD in recent years.
--
On the contrary. The most important part is to determine why he was right, why he thought that, what did he mean when he said it. What comes last and least important is what it might mean to us.
To do otherwise is to reduce us to a million monkeys on a million typewriters. If I say something profound by accident, it isn't nearly as profound - in fact, it's absolutely vacant.
What Jefferson said at any one point, good quote or not, is made important by his involvement in world affairs, in the intellectual life of colonial Virginia and the early United States, his time in France. Not to mention his Presidency and founding of the University of Virginia. Jefferson was a very ambitious and astute politician who made good on his potential and that's why he's important; he didn't just make good quote.
--
No, the same programs don't work (or don't even compile) on all of them. There are library incompatibilities, for one. BSD has gone farthest by releasing a user-space linux emulator.
--
I haven't had trouble compiling. However, I was referring to binary compatibility, not source compatibility. (Of course, there's someone else in this thread who insists that emulation equates compatibility, so perhaps it's possible to go too far the other way).
For that matter, I really haven't had that much difficulty porting stuff from BSD to Linux, either. Maybe I *am* lucky, or I just haven't tried anything really really difficult yet.
--
I disagree.
;-). I don't think that they'd incorporate architectural changes which come from outside their core group (it seems that they've rejected all proposals so far) and the SCSL makes a mockery of open source.
Larry's keeping tight hold of perl (as perl; you're free to reuse anything to do with perl so long as you don't call it perl) for the sake of perl. He trusts his own judgement, in other words, but he's always amenable to new arguments.
Sun, on the other hand, is keeping tight hold of Java as a revenue centre. The good of the application as a functional and usable piece of work comes second, or sometimes not at all, if I can borrow from Austin Powers
I would feel hurt, both emotionally and professionally, if I'd taken Sun's goodwill at face value, like the Blackdown team did. Despite their occasional gesture towards open source, they're Just Another Corporation at heart.
--
They run them through binary emulation. By the same token, you could say that Linux on x86 is compatible with old atari systems, merely because an emulator exists. It's clearly not so.
BSD advocates on a public discussion forum are all well and good, but look at the documentation, even the most basic documentation, before you make an assertion. From the OpenBSD FAQ:
1.1 - What is OpenBSD?
The [7]OpenBSD project produces a freely available, multi-platform 4.4BSD-based UNIX-like operating system. Our efforts place emphasis on portability, standardization, correctness, and security. [8]OpenBSD supports binary emulation of most binaries from SVR4 (Solaris),FreeBSD, Linux, BSDI, SunOS, and HPUX.
Oh, and I use OpenBSD as well as Linux. Nice try.
--
I think that the point is that most users don't demand compatibility - systems analysts and other such pedantic characters may desire it and even write it in, but when you're dealing with joe developer, much less joe user, thoughts of compatibility are relatively irrelevant. It's almost always a matter of price-performance and relevance to the business case.
--
Don't make Jefferson out as a proto-libertarian, which you're tending towards. He most certainly wasn't. He was a revolutionary and a political liberal, but intellectually he remained a child of the scottish enlightenment. His rejection of british conceptions of liberalism was a political, rather than a philosophical, move.
--
RMS does much better when he describes the philosophical and technical issues behind his positions. As it is, this article is full of unsupported assertions, questionable facts, and general gobbledygook.
For example, the assertion that "putting users in control" (that is, opening the source, preferably under the GPL) is the best way to assure continued compatibility. I am no philosopher of science, but it seems to me that putting users in control and allowing code forking is to encourage incompatibility. Witness the fate of the BSD's - originally compatible due to their shared code base in BSD4.4, presently incompatible due to different directions in development. And that over a relatively short time span.
Another example, his trust in the market to favour a Java-compatible implementation. As such things go, I doubt that the market would favour any such thing. The most popular pieces of software, word processors and web browsers, are perpetually crippled with respect to backwards and forwards compatibility, much less interoperability. What the market favours is price, availability, and support. Compatibility, it seems, has been left by the wayside. This may be short-sighted, and I believe it is, but RMS may be giving us, as an industry, much too much credit here.
I can only hope that he didn't intend this piece of correspondence for publication. As is the case with many visionaries (and I have no doubt that he is a visionary), he needs to tone down his message to sell it to the masses. Right now, I doubt that anyone is any more sold on open source, and a golden opportunity wasted.
--
Don't forget that many anti-communist agitators in eastern europe were priests. Several were killed for such activity in Poland alone; the pope himself only barely escaped that sad fate.
That said, anti-communism strikes me as a one-note philosophy, lacking a connection to a larger whole or principle. Which isn't to say that anticommunists are unprincipled, merely that the anti-communist movement as a whole is such an incredibly mixed bag that we can use it as proof of almost nothing - except for anti-communism itself, of course. It spans the breadth of political thought, from die-hard leftist intellectuals (Orwell) to totalitarian dictators of the worst order (Pinochet).
When selecting a man of the century, one should select a man who exemplifies the thought of the century. Should we select a man who exemplifies only a single thought - perhaps a man who only had a single thought? If we are to select a laughingstock, then we should abandon all pretense and seek out the greatest laughingstock available.
If we are to seek out a great man, on the other hand, then someone who thought great thoughts and performed great deeds (that is, someone who wasn't along for the ride of history) is as good a choice as you could possibly make. Marshall is an entirely apt choice in this respect.
--