I have no idea if you're right, and I don't really care, because the point is that this is a common technique, and the patent is ridiculous. McDonnell-Douglas should disavow this idiotic move by one of their clueless employees, and even more clueless lawyers. Are the lawyers getting a percentage of the take, perhaps? As to the patent office, if this doesn't prove that they're incompetent, then I don't know what would.
So, no, I didn't patent (or publish) the idea because I'm not an idiot, and I don't try to patent things that are obvious. In fact, one of the criteria of a valid patent is that it must be non-obvious.
Here are some other things I have never patented: 1. Measuring 30 feet with a 20 foot tape measure by first marking off 20 feet, then measuring from the mark. 2. Determining the weight of a liquid by first weighing the empty container, and subtracting. 3. Calculating the mileage of my car by keeping track of whether the odometer has rolled over, and adding 100,000 if it has.
You're not really suggesting that this patent requires formal prior art in order to be invalidated, are you?
I suspect this Y2K patent will fail to stand up in court, I'm guessing that some far-sighted engineer did something similar in the 80s and "prior art" will apply.
Dear IBM Canada, Call me. I used this technique in the common date routines in your CPLC software inventory system in 1976. I have the printouts to prove it.
Linux is my O/S, and the O/S I'm pushing to become dominant in homes and businesses.
However, for military hardware (e.g. battleships), I suggest that a microkernel imbedded O/S, such as QNX or VxWorks, would be the proper solution.
The beauty of a microkernel O/S is that it is made up of small modules, each of which can be independently verified to work perfectly.
Now, Linux offers similarly high levels of security, and reliability, through Open Source, and the intense review of thousands of developers. The military, on the other hand, will most certainly want to keep their source to themselves (wisely or not), and will want to do their own reviews.
The other advantage of a microkernel imbedded O/S, for military applications, is that it's better suited to real-time guaranteed-response systems, whereas a more monolithic O/S such as Linux may offer better peak performance, as is generally required by a PC user.
Some years ago, when Microsoft wanted to convince Unix programmers to switch to Windows, they contracted with Bristol to provide a Windows API layer on Unix. In theory, by using this API layer, applications could be developed on less expensive Windows systems, with full confidence that they would run on Unix. Also, use of the API would allow developers to make the choice (*ahem*) to move their apps to Windows later on (when NT inevitably took over). In order to gain developer acceptance, Microsoft made public promises about their ongoing support:
"Microsoft is committed to providing WISE licensees with future versions of Windows family source code, thereby continuing to maximize application compatibility and performance for today's and tomorrow's applications. Customers can be confident that their investments today will continue to realize benefits well into the future."
Of course, it wasn't a real commitment:
"I have heard it said that we want to allow our top 50 ISVs to be able to innovate on Windows as their primary platform, and port back onto their Unix platforms, *** but we don't want it to work too well ***." (emphasis added)
When Microsoft went back on their promise to supply the source code, Bristol took them to court. Unfortunately for Bristol, Microsoft hadn't made the same promises in their contract as they had to the public, so Bristol lost their case.
Although the case was lost, it brings into question Microsoft's real intentions behind their current Unix interoperability efforts. Does anyone want to argue that this time Microsoft really, really means it? Do you think desparation will make honest men of them? I'll believe it when I see it.
You forgot to mention that Microsoft is being taken to court for their Mosaic license. Do you think MS knew that they were going to be giving IE away when they licensed Mosaic for a percentage of revenue, rather than a fixed price? Can you say "negotiating in bad faith"?
Okay, I've re-read the original post again, and maybe Aaron is too disrespectful of Microsoft for his post to be real astroturf (No, wait--thats just what they want us to think! Bwahahaha:^). Aaron's post, however, is not helping the cause.
First, Aaron's description of Netscape's early days is a mischaracterization, at best.
Second, while Netscape did tend to add their own html extensions, as far as I can tell it was always done to add new capabilities. This is a far cry from many of Microsoft's extensions, some of which seem mainly intended to break Netscape.
Third, even if Netscape did extend standards in the past, the current AOL/Netscape/Mozilla efforts seem to be strongly supportive of standards.
And last, I had a serious problem with the phrase "Microsoft dominated web". If ever there was a lie that MS wanted us to believe, this is it. MS is not even close to dominating the web, but if we keep repeating that phrase long enough, it will become a reality.
My warning about MS astroturf is still valid. If you look further up among the posts, and look over the last couple of weeks of Slashdot (since about the time of Ed Muth's anti-Linux comments), I think you will find dozens of candidates. Just--be careful. Question everything you read, and always check it against your knowledge of the facts. Don't let your opinions be formed for you.
For those who don't know the background, astroturfing refers to an attempt to use paid help to fake grassroots support for something (e.g. a bill in congress). Microsoft was caught in the act preparing for an astroturf campaign to help them against the DOJ. The plan was to pay people to write letters-to-the-editor, put posts in newsgroups, and so on. All the letters and posts would be written so as to appear to come from an individual, i.e. a concerned member of the public, but they would always follow a script designed to push the Microsoft agenda. The controversy may have subsided, but there is ample evidence that the astroturf continued.
With JWZ'z resignation, Microsoft can smell blood, and you can bet your last dollar that they have moved quickly to figure out how to turn it to their advantage. After all, the Netscape browser is one of the greatest impediments to MS expanding their OS monopoly to the Internet (others include Linux, Apache, and Java). As long as there is a viable browser on the market besides IE, it prevents MS from playing with the web standards. This is a greater danger than anything they could do to Linux, because an OS runs independently, while the web lives or dies on standards. If MS could manage to become the leader in the browser market, they could do what they have always done--implement complex, hidden, and quickly-changing standards that no one else can follow. It would become almost impossible to avoid using IE, at least part of the time, in order to access the web. And, to keep up with the changes, we would be forced to pay, and pay, and pay for the "upgrades" (whether to IE or Windows).
The biggest victory MS could have would be if they could get people (users and developers) to lose confidence in Netscape, and especially in the open source version, Mozilla. Remember that OS/2 didn't lose to Windows for technically reasons--it lost through lack of support by IBM, which was a direct result of loss of confidence by IBM management. IBM lost because the were foolich enough to listen to the barrage of MS-sponsered articles and editorials, with their constant message: "Of course it's inevitable that Windows will take over".
Consider the above post in this light, and what do we see? It has all the earmarks of MS propaganda. It does not really discuss JWZ's resignation. It doesn't talk about how we can move forward to ensure the success of Mozilla, or some other good standards-supporting browser. The only purpose of this post seems to be to support the MS message: "It is time to give up on this Netscape foolishness. Though you may not like it, IE is inevitable".
All I can say is, don't fall for this. As tragic as JWZ's resignation may be, it is not the end of the world--Internet standards will survive as long as we keep up our confidence. We need to, and will, do whatever it takes, whether that is supporting AOL's Communicator, Mozilla, Forkzilla, Opera, the KDE File Manager (try it), or some completely new alternative.
If the only voices I had heard, after discovering Linux, had been RMS and the anti-IP crowd, I would have turned around and left.
It is only because of voices of reason, such as ESR, Bob Young, and Linus himself, that I am here.
I believe in property rights, including intellectual property. I believe that what a man (or woman, of course) creates belongs to him, and it is his choice whether to trade it for pay, sell it, or (in this case) give it away. I believe property rights are the foundation of a free society.
I also support Linux. I view the Linux effort as "men of good character voluntarily choosing to co-operate for mutual benefit and a better world".
Thus, I find no conflict between support for open standards, support for open source projects such as Linux, and involvement by the corporate world. There are men of good character in the corporate world, who will co-operate honestly, and enhance the Linux effort (there are also, of course, some glaring exceptions).
I can understand why ESR wrote this, because I have considered posing a similar question. For a long time, I have been working on an idea of my own, and if (no--when, damn it!) it gets released, it would be as proprietary, closed source software. This raises the question: Am I welcome here? I would rather release my software under Linux than Windows, despite the fact that, for the moment at least, there is probably more money to be made under Windows. I would consider this to be an act of support for Linux. It is disheartening to know that so many in the Linux community would loathe me for doing so.
I don't know who else can do the job that Eric is doing (I sure know it's not me). All I can do is say to Eric that his efforts are appreciated.
I tried to give it a careful reading to be fair, but I couldn't make it to the end. It was just TOO PAINFUL. It was remeniscent of that episode in the original Star Trek series, the one with the theraputic beam that could be used to torture one's victim by EMPTYING HIS MIND.
In most Windows shops, the users are given full access to their machines. While ultimately this may be unwise, the sysadmin knows that it's extremely unlikely that the users will start replacing parts of Windows, let alone hack the binaries.
I am not saying (and I don't think Mettler is either) that Linux has a fatal security flaw, only that open source presents some new security issues that must be taken into account by taking extra steps. You have described one of those steps (obvious in a Unix shop), that of denying root access to the users.
Looking at it from the other side, as a user in a Windows shop (shudder:^), I can be fairly confident that I am working with the bona-fide original versions of Windows and other programs. While it is possible for an unscrupulous sysadmin to hack the binaries, doing so is generally beyond the ability of most sysadmins (or at least more effort than he is likely to expend). With open source, however, my sysadmin could quite easily introduce changes that would give him access to all my data (he could, for example, modify the shop's encryption program to store passwords in a hidden file). This means, as Mettler has pointed out, that as a user of open source software, I need better tools for ensuring that I am using the (unmodified) version of the code that I think I am. Other open source advocates have suggested the same thing.
We can't assume that normal Unix security is enough to make us safe. We have to consider the new challenges presented by open source, then do everything we can think of to deal with them.
I've read lots of L.A.M.'s posts on ZDNet, mostly on the subject of the DOJ vs Microsoft. His repeated point was that, speaking as a lawyer, Microsoft has acted illegally, and should be found guilty for their crimes. L.A.M. tends to be somewhat pedantic and repetitive, but his posts were also intelligent and entertaining, and I used to watch for them.
Regarding Linux's security, he has a point--opening the source does introduce a new security risk. It is ESR's point in reverse. If opening the source means many eyes to look for bugs, then it also means many eyes to look for weaknesses to exploit. It's not a fatal flaw. As others have said, the large number of reviewers makes open source the antidote for the very problem it introduces.
Regarding physical access versus root-authority-based security, I don't think that is his point. The point is as follows. With proprietary software, the company buys an official copy of the software, the sysadmin loads it onto everyone's PCs, and no one is expected to ever change it. In fact, all the anti-virus software needs to look for is whether anything has changed--if it has, that's a problem. With open source software, on the other hand, everything is expected to change. Individual users are expected to pick up new software or new versions and add it to their systems, even to the point of recompiling the kernal itself. That's a security risk because, it introduces a dozen new avenues for attack. Who's to say that every user has obtained his software from only trusted sources. Of course, the problem is not unique to Linux. Word macros, and IE's VB-Script represent similar open source, "anyone can pick it up and run it" security problems (in many ways, the Word and VB-Scipt problems are worse, because the user can end up running the virus when all he intended to do was browse). This other security concern means that, as others have pointed out, we need better tools for ensuring that we are working with correct versions of the libraries (I forget the details of the exploit that recently reignited this discussion).
Mettler's other point is that when someone tries to discuss these concerns, he runs headlong into the rabid response that "Linux is perfect, therefore there can't possibly be anything wrong with it, and curse you for even suggesting it". Then, when he tries to point that out, he runs into "The Linux community is perfect, therefore we can't possibly be arrogant". I think there are enough posts here to demonstrate that there is some truth to those points as well.
I didn't see any of the earlier discussions, so I don't know if Mettler has become angry and pedantic to the point of leaving the field of reason. What I do know is that we must not ignore the new security issues that are raised by opening the source. Rather than burying our heads in the sand, we must identify the risks, discuss them openly and rationally, then, using the benefits provided by open source, create the means to deal with them.
Rob, I posted this secure in the knowledge that you have nothing better to do than to wait for my suggestions, and cater to my needs:^). Waiting eagerly...
Rob, now that I've gotten started, and I've had time to think, here's what I really want:
Beside each "Read More..." link on the main page, add a second link named "Select...". The "Read More" link would operate as it currently does, based on the user's defaults. The "Select" link would open a selection-criteria page. The story-id would be specified in the URL, but the user would then enter the other filtering criteria. I believe that this would not only give me more flexibility, but would also reduce the load on your system, since users would no longer have to download a page one way (e.g. threaded), then download it again another (flat).
The criteria I would like to be able to specify are: - Threaded or flat - Page threshhold (solves AC's problem) - Headers-only or messages - Sort by newest/oldest/score - Newer-than date/time - First 50 msgs etc. (i.e. first/last msg seq#) - Later than cid#
The last two points would make it easy to come back later and continue reading. If you wanted to implement it perfectly, the criteria would be "messages listed after, or >= time of, cid#".
Other suggestions: - Include msg scores on the headers-only display. - List every msg on the headers-only display, with a flag (*) to indicate which ones meet the filter criteria. - On the message display, always include the cid#. On the flat display, also show the parent cid#. This would allow users to piece together the thread without having to opening a second brower page for a threaded display (showing hierarchy level#s would normally be an alternative, but it falls apart when messages can be filtered out).
Rob, how about providing a date/time filter field on the comments page. This would allow us to come back to a page later, enter a date/time, and only view the comments that are new, even if they are buried in the hierarchy.
I agree that the $99 charge could be to cover Dell's Linux support costs (although a hidden M$ tax would'nt surprize me either).
I disagree that their support costs are going to turn out to be low, at least not at first.
It's not that Linux is harder to maintain than Windows. My experience agrees with others--Linux is more reliable, and once you get something working, it stays working.
The problem is the learning curve. A large percentage of Dell's customers will be starting with no Linux experience (or Unix for that matter). The fact that Linux is pre-installed will help of course, but Dell is going to have to deal with questions like "How do I get my ADSL modem working?", or "How do I install application X?".
It also won't help that there is a lot of good support on the Internet. Current Linux users are mostly Hackers, and they are, by definition, self-teaching. Future Linux users (not to mention your typical barely competent sysadmin) will not be Hackers, and they will need their hands held.
In order for Linux to realize the low support costs that should accompany its quality, a few things need to happen:
1. Linux distributions need to come with better documentation (Redhat--your book sucks!).
2. Installation procedures for Linux applications need to be simplified, standardized, and documented.
3. There needs to be enough experienced Linux users around that new users can turn to their co-workers, friends, and sons and daughters for help, the way they do today with Windows.
I predict a rosy future for Linux, but we all know that getting there is not going to be a cakewalk, especially when it comes to getting the general public involved.
You're forgetting something. Wiping Microsoft off the face of the earth, as enjoyable, and as beneficial for mankind as that might be, is not the primary goal of Linux. It is, at best, a side benefit, and it's likelihood is questionable.
Remember--the primary goal is to have a great O/S (plus apps). All that is required for this is a critical mass of users who are dedicated to Linux and open standards.
If MS created their own proprietary version of Linux, then it's true that many, if not most businesses would support it. If MS violated the GPL, the lawsuits would fly (class-action suit--all Linux users please donate $1:^). MS might win or lose in marketshare, and in court.
In the end, though, none of it would matter. The market would be split between MS Linux (Win 2001) supported by a single vendor, and Standard Linux, supported by many vendors, IBM, Oracle, etc. If you don't think that a critical mass of support would survive for Standard Linux, just look at the Debian crowd:^).
1. MS wants to make money on the ported Office, plain and simple.
2. MS can't sabotage Linux unless they first get a foot in the door with a Linux app.
3. A flaky port of Office would hurt the credibility of Linux.
4. A promised port of Office that never emerges would hurt the credibility of Linux.
5. MS Office will require proprietary MS libs, allowing MS to gradually take over or corrupt Linux.
6. Office and IE on Linux will help MS to decommoditize HTML and other protocols.
To these we now add the following:
7. Office on Linux is just a lost leader, with the upgrade to Office 2000 requiring a corresponding "upgrade" to Windows 2000 (your theory).
Finally, I would add the following variation:
8. Office on Linux is a desparation ploy, to keep users from migrating away from MS Office (due to Y2K), while MS finishes Windows 2000.
This last theory is interesting. Office 97, Win95, and NT are not Y2K-ready. Windows 2000 will not be ready in time for Y2K. Office beta-2000 may be Y2K-ready, but there is no MS platform to support it. If customers migrate to Linux to solve their Y2K problems, they will also migrate to other office apps, making it nearly impossible for MS to get them back. The only way out of this is to port MS Office (beta-2000) to Linux in order to keep Office users in a holding pattern until W2K plus O2K are ready. It's their only chance.
2. Never waste money on research. You can save money by letting others innovate for you.
3. Never compete on the basis of quality. FUD and customer lock-in are cheaper and easier.
4. Never allow your products to become compatible. If others manage to adopt your standards, change them. This will ensure customer lock-in.
5. Never wait until your product is stable before releasing it. By releasing early, you save money on development, while ensuring extra cash for the "upgrade" (e.g. Windows 98).
6. Never provide full backward compatibilty with your own products. Provide just enough to claim compatibility, while ensuring that your customers will eventually be forced to pay you for the upgrade.
7. Ignore your competitor's protocols. If a competitor's protocol looks like it will succeed, embrace and extend that protocol until you destroy it or control it.
8. Never lie about your competitor's products--unless you can get away with it.
10. If a competitor's product runs on your platform, never use a false error message to discredit it. Use an incomprehensible message, and allow the user's doubts to do your work for you.
11. Never publish a complete spec for your platform. Always leave enough undocumented to ensure that your competitor's products can be made incompatible on the next release.
12. Never provide your customers with complete documentation for your products. Complete documentation leads to customers being in control of their systems, and the resulting confidence will interfere with your ability to control their decisions.
Once the masses have caught on to something, it can really hurt (my personal gripe in this regard was FM radio). The famed Linux community support will be severely tested when there are enough poorly-informed Linux users to ensure that most of the questions are about obvious/documented things. Remember, though, that the Internet, unlike radio, is not a mass media. With the benefits of unlimited bandwidth (in the logical sense), and narrow-casting, separate communities will form for the casual Linux users, and the techie Linux developers.
I am not too worried about a non-GPL layer of commercial software running on top of Linux. In fact, I think that such a layer would tend to reinforce the standardization of Linux (of course, then we have to worry about stagnation - the world is always tossing in new challenges), and would provide more motivation for getting rid of the quirks in the various GUI's.
As for Microsoft, if they just became one application supplier among many, then that alone would be a great benefit. I also don't think the fact that MS was making money would kill the Linux spirit. I am a recent convert to Linux, and I made the move as much to escape MS as anything. Now that I am here, however, I have rediscovered the joy of computing. I have come to agree with those who say that Linux is a positive force, and does not need MS as an enemy in order to survive.
Of course, we both know that MS would never be satisfied with competing on even ground. The have the mentality of a thief, and they will always believe that they can tilt the system in their favor, and get away with it. They would follow their usual strategy. The MS Linux apps and distribution would initially start out compatible, but gradually, over time, we would find that MS apps would only run reliably (as much as MS apps ever do) on the MS distribution, or would require proprietary MS libraries. What would happen then? We've already seen what would happen with regards to HTML and the Internet. A grass roots Standard Linux group would form, similar to W3C, and would get the backing of much of the industry. It would once again be Robin Hood versus the bad guys, and the Linux community would be re-energized.
The Internet has changed everything for Microsoft. In the old days, MS could buy half the media, and silence the other half with legal threats. Like all good con men, they were masters at the image game. Today, with the Internet, there are countless individuals who will not be silenced, and the word gets out. Also, unlike the old days when computing was an end in itself, more and more, the network and PCs are becoming a means to an end, i.e just tools. Commercial interests on the Internet want to reach a large audience of consumers. Thus, when MS tries to put up roadblocks, it creates a huge financial incentive to find a way around them, and, unlike the pioneering days, lots of trails have been blazed - too many for MS to control.
It may not be world domination, but I think Linux has a bright future.
Say Microsoft did create their own distribution of Linux? What would happen? Businesses would be faced with a choice between a 100% pure, open-standard, stable Linux, available from multiple trusted vendors, and a more expensive, proprietary, less-stable Linux, supplied by a single, sneaky vendor. Given that Microsoft would not have their current IBM-deeded head start, I'll bet that the open standard would win, just as is now happening with Corba, Java, and the Internet in general.
Let's say I'm wrong. Say Microsoft once again sucked the PHBs into using their adulterated Linux. What will have changed? What's the difference between a proprietary MS Windows, and a proprietary MS Linux? The answer is: no difference! The Linux community will carry on, continuing to support open-standard, open-source Linux. And, to those who argue that an MS bizarro-Linux will hurt the credibility of Linux, I would answer that, in the eyes of the PHBs, MS will have legitimized it, thus making the leap to real Linux even easier.
I wasn't betting that Linux would kill Microsoft - I was betting that Linux would fulfill my computing needs, now and in the future. Even if none of the promised software support comes from the major companies, such as IBM, I will still have won.
As to Microsoft, they will ultimately die of their own poison and lies.
The FSF is against information ownership,..., they don't seem to oppose capitalism in other areas...
You are right, of course, about the FSF. My comments were aimed more toward those Slashdot posters who take the OSS phenomenon as a validation of Socialism in general (and who really do mean free as in free lunch).
As to the FSF, I can't agree with their goal of ending information ownership. Like you, I think they are trying to live a contradiction. In fact, I consider that part of their philosophy to be dangerous. Note that, while they want to to remove my IP rights, the reason that I choose not to violate the GPL, whether it's legally enforced or not, is out of respect for their IP rights.
Despite my differences with the FSF, I continue to support Linux and other OSS projects. It makes sense for the operating system to be open, and it's an excellent free-market solution to the threat of a Microsoft monopoly (and it's fun, and you meet nice people). As another poster pointed out (smithdog, below), cooperation has always gone hand-in-hand with Capitalism. There is no reason why free individuals, working toward a common goal, can't share the fruits of their labor, while still respecting each other's, and everyone else's, property rights.
The popular understanding of Capitalism has become so corrupted, it's hard to discuss it intelligently anymore. This was largely done on purpose (which is not to say there is a conscious plan involved) by the major segments of society whose power and money would be threatened by true Capitalism, including Politicians, University Professors, Union Leaders, Bankers, and Big (so-called) Businessmen. Despite what you have heard, Capitalism is not about international money-changing, nor can it be characterized by pointing to the problems created by the huge government-regulated, government-controlled stock market and public corporations.
Capitalism is about private ownership of Capital. The central point is the private ownership, which is closely linked to freedom. This means that each person is entitled to (i.e. owns) the results of his own work (unless he sells it or trades it for pay), and the definition of ownership is that you get to say what can or cannot be done with your property.
The GPL is a license setting down the wishes of the creators of GNU and other OSS software. We expect those wishes to be honored. We even expect the courts to enforce the license. This implies that the creators have a legal right to set the rules for their creation. That, in turn, implies ownership (even is the ownership is subsequently transferred to the public), which is a Capitalist concept, whether some OSS-ers think so or not.
Under Socialism, there is no ownership. Everything is shared by society, for the good of the people (or the state). Of course, there still needs to be rules to manage the sharing, otherwise, what's to stop me and my guests from using up the entire world's supply of Truffles and Dom Perignon. Thus, whenever Socialism is implemented, it breaks down into Ownership of Everything by the Government.
Under Socialism, the GPL would be ignored, since it only represents the wishes of a few private individuals. Under Socialism, the Government could decide that it was in the best interest of society that some companies (say those located in Redmond) be allowed to provide custom versions, and binaries only, and there would be nothing you could say about it. Socialism would destroy the GPL and the OSS community.
If you really want a better understanding of Capitalism, read Milton Friedman's Free To Choose. It's insightful, entertaining, and written for the general public. For the more philosophically inclined, I would also suggest Ayn Rand's Capitalism: The Unknown Ideal (though I risk re-igniting a recent flame war - some here think Rand is full of crap, but I'm definitely not one of them).
To summarize: If you support the GPL, then you must support Capitalism, because that's the only system under which the GPL could exist.
Slashdot Mirror
I have no idea if you're right, and I don't really care, because the point is that this is a common technique, and the patent is ridiculous. McDonnell-Douglas should disavow this idiotic move by one of their clueless employees, and even more clueless lawyers. Are the lawyers getting a percentage of the take, perhaps? As to the patent office, if this doesn't prove that they're incompetent, then I don't know what would.
So, no, I didn't patent (or publish) the idea because I'm not an idiot, and I don't try to patent things that are obvious. In fact, one of the criteria of a valid patent is that it must be non-obvious.
Here are some other things I have never patented:
1. Measuring 30 feet with a 20 foot tape measure by first marking off 20 feet, then measuring from the mark.
2. Determining the weight of a liquid by first weighing the empty container, and subtracting.
3. Calculating the mileage of my car by keeping track of whether the odometer has rolled over, and adding 100,000 if it has.
You're not really suggesting that this patent requires formal prior art in order to be invalidated, are you?
I suspect this Y2K patent will fail to stand up in court, I'm guessing that some far-sighted engineer did something similar in the 80s and "prior art" will apply.
Dear IBM Canada,
Call me. I used this technique in the common date routines in your CPLC software inventory system in 1976. I have the printouts to prove it.
A Technical Comparison of CORBA vs DCOM:
s /HTML/DCOMnCORBA/S.html
o rbacom.html
http://www.research.microsoft.com/~ymwang/paper
The paper concludes that they are roughly equivalent, but the diagrams indicate to me that the design of CORBA is much cleaner.
A survey of CORBA usage vs DCOM, etc. (from 1998):
http://www.cutter.com/consortium/architecture/c
Linux is my O/S, and the O/S I'm pushing to become dominant in homes and businesses.
However, for military hardware (e.g. battleships), I suggest that a microkernel imbedded O/S, such as QNX or VxWorks, would be the proper solution.
The beauty of a microkernel O/S is that it is made up of small modules, each of which can be independently verified to work perfectly.
Now, Linux offers similarly high levels of security, and reliability, through Open Source, and the intense review of thousands of developers. The military, on the other hand, will most certainly want to keep their source to themselves (wisely or not), and will want to do their own reviews.
The other advantage of a microkernel imbedded O/S, for military applications, is that it's better suited to real-time guaranteed-response systems, whereas a more monolithic O/S such as Linux may offer better peak performance, as is generally required by a PC user.
Don't forget Bristol's recent Lawsuit against MS:
e vised.html
http://www.bristol.com/legal/summary_of_proof_r
Some years ago, when Microsoft wanted to convince Unix programmers to switch to Windows, they contracted with Bristol to provide a Windows API layer on Unix. In theory, by using this API layer, applications could be developed on less expensive Windows systems, with full confidence that they would run on Unix. Also, use of the API would allow developers to make the choice (*ahem*) to move their apps to Windows later on (when NT inevitably took over). In order to gain developer acceptance, Microsoft made public promises about their ongoing support:
"Microsoft is committed to providing WISE licensees with future versions of Windows family source code, thereby continuing to maximize application compatibility and performance for today's and tomorrow's applications. Customers can be confident that their investments today will continue to realize benefits well into the future."
Of course, it wasn't a real commitment:
"I have heard it said that we want to allow our top 50 ISVs to be able to innovate on Windows as their primary platform, and port back onto their Unix platforms, *** but we don't want it to work too well ***." (emphasis added)
When Microsoft went back on their promise to supply the source code, Bristol took them to court. Unfortunately for Bristol, Microsoft hadn't made the same promises in their contract as they had to the public, so Bristol lost their case.
Although the case was lost, it brings into question Microsoft's real intentions behind their current Unix interoperability efforts. Does anyone want to argue that this time Microsoft really, really means it? Do you think desparation will make honest men of them? I'll believe it when I see it.
You forgot to mention that Microsoft is being taken to court for their Mosaic license. Do you think MS knew that they were going to be giving IE away when they licensed Mosaic for a percentage of revenue, rather than a fixed price? Can you say "negotiating in bad faith"?
Okay, I've re-read the original post again, and maybe Aaron is too disrespectful of Microsoft for his post to be real astroturf (No, wait--thats just what they want us to think! Bwahahaha :^). Aaron's post, however, is not helping the cause.
First, Aaron's description of Netscape's early days is a mischaracterization, at best.
Second, while Netscape did tend to add their own html extensions, as far as I can tell it was always done to add new capabilities. This is a far cry from many of Microsoft's extensions, some of which seem mainly intended to break Netscape.
Third, even if Netscape did extend standards in the past, the current AOL/Netscape/Mozilla efforts seem to be strongly supportive of standards.
And last, I had a serious problem with the phrase "Microsoft dominated web". If ever there was a lie that MS wanted us to believe, this is it. MS is not even close to dominating the web, but if we keep repeating that phrase long enough, it will become a reality.
My warning about MS astroturf is still valid. If you look further up among the posts, and look over the last couple of weeks of Slashdot (since about the time of Ed Muth's anti-Linux comments), I think you will find dozens of candidates. Just--be careful. Question everything you read, and always check it against your knowledge of the facts. Don't let your opinions be formed for you.
For those who don't know the background, astroturfing refers to an attempt to use paid help to fake grassroots support for something (e.g. a bill in congress). Microsoft was caught in the act preparing for an astroturf campaign to help them against the DOJ. The plan was to pay people to write letters-to-the-editor, put posts in newsgroups, and so on. All the letters and posts would be written so as to appear to come from an individual, i.e. a concerned member of the public, but they would always follow a script designed to push the Microsoft agenda. The controversy may have subsided, but there is ample evidence that the astroturf continued.
With JWZ'z resignation, Microsoft can smell blood, and you can bet your last dollar that they have moved quickly to figure out how to turn it to their advantage. After all, the Netscape browser is one of the greatest impediments to MS expanding their OS monopoly to the Internet (others include Linux, Apache, and Java). As long as there is a viable browser on the market besides IE, it prevents MS from playing with the web standards. This is a greater danger than anything they could do to Linux, because an OS runs independently, while the web lives or dies on standards. If MS could manage to become the leader in the browser market, they could do what they have always done--implement complex, hidden, and quickly-changing standards that no one else can follow. It would become almost impossible to avoid using IE, at least part of the time, in order to access the web. And, to keep up with the changes, we would be forced to pay, and pay, and pay for the "upgrades" (whether to IE or Windows).
The biggest victory MS could have would be if they could get people (users and developers) to lose confidence in Netscape, and especially in the open source version, Mozilla. Remember that OS/2 didn't lose to Windows for technically reasons--it lost through lack of support by IBM, which was a direct result of loss of confidence by IBM management. IBM lost because the were foolich enough to listen to the barrage of MS-sponsered articles and editorials, with their constant message: "Of course it's inevitable that Windows will take over".
Consider the above post in this light, and what do we see? It has all the earmarks of MS propaganda. It does not really discuss JWZ's resignation. It doesn't talk about how we can move forward to ensure the success of Mozilla, or some other good standards-supporting browser. The only purpose of this post seems to be to support the MS message: "It is time to give up on this Netscape foolishness. Though you may not like it, IE is inevitable".
All I can say is, don't fall for this. As tragic as JWZ's resignation may be, it is not the end of the world--Internet standards will survive as long as we keep up our confidence. We need to, and will, do whatever it takes, whether that is supporting AOL's Communicator, Mozilla, Forkzilla, Opera, the KDE File Manager (try it), or some completely new alternative.
If the only voices I had heard, after discovering Linux, had been RMS and the anti-IP crowd, I would have turned around and left.
It is only because of voices of reason, such as ESR, Bob Young, and Linus himself, that I am here.
I believe in property rights, including intellectual property. I believe that what a man (or woman, of course) creates belongs to him, and it is his choice whether to trade it for pay, sell it, or (in this case) give it away. I believe property rights are the foundation of a free society.
I also support Linux. I view the Linux effort as "men of good character voluntarily choosing to co-operate for mutual benefit and a better world".
Thus, I find no conflict between support for open standards, support for open source projects such as Linux, and involvement by the corporate world. There are men of good character in the corporate world, who will co-operate honestly, and enhance the Linux effort (there are also, of course, some glaring exceptions).
I can understand why ESR wrote this, because I have considered posing a similar question. For a long time, I have been working on an idea of my own, and if (no--when, damn it!) it gets released, it would be as proprietary, closed source software. This raises the question: Am I welcome here? I would rather release my software under Linux than Windows, despite the fact that, for the moment at least, there is probably more money to be made under Windows. I would consider this to be an act of support for Linux. It is disheartening to know that so many in the Linux community would loathe me for doing so.
I don't know who else can do the job that Eric is doing (I sure know it's not me). All I can do is say to Eric that his efforts are appreciated.
Whew. Thank you for the Shakespere quote. My mind is full again.
I tried to give it a careful reading to be fair, but I couldn't make it to the end. It was just TOO PAINFUL. It was remeniscent of that episode in the original Star Trek series, the one with the theraputic beam that could be used to torture one's victim by EMPTYING HIS MIND.
In most Windows shops, the users are given full access to their machines. While ultimately this may be unwise, the sysadmin knows that it's extremely unlikely that the users will start replacing parts of Windows, let alone hack the binaries.
:^), I can be fairly confident that I am working with the bona-fide original versions of Windows and other programs. While it is possible for an unscrupulous sysadmin to hack the binaries, doing so is generally beyond the ability of most sysadmins (or at least more effort than he is likely to expend). With open source, however, my sysadmin could quite easily introduce changes that would give him access to all my data (he could, for example, modify the shop's encryption program to store passwords in a hidden file). This means, as Mettler has pointed out, that as a user of open source software, I need better tools for ensuring that I am using the (unmodified) version of the code that I think I am. Other open source advocates have suggested the same thing.
I am not saying (and I don't think Mettler is either) that Linux has a fatal security flaw, only that open source presents some new security issues that must be taken into account by taking extra steps. You have described one of those steps (obvious in a Unix shop), that of denying root access to the users.
Looking at it from the other side, as a user in a Windows shop (shudder
We can't assume that normal Unix security is enough to make us safe. We have to consider the new challenges presented by open source, then do everything we can think of to deal with them.
I've read lots of L.A.M.'s posts on ZDNet, mostly on the subject of the DOJ vs Microsoft. His repeated point was that, speaking as a lawyer, Microsoft has acted illegally, and should be found guilty for their crimes. L.A.M. tends to be somewhat pedantic and repetitive, but his posts were also intelligent and entertaining, and I used to watch for them.
Regarding Linux's security, he has a point--opening the source does introduce a new security risk. It is ESR's point in reverse. If opening the source means many eyes to look for bugs, then it also means many eyes to look for weaknesses to exploit. It's not a fatal flaw. As others have said, the large number of reviewers makes open source the antidote for the very problem it introduces.
Regarding physical access versus root-authority-based security, I don't think that is his point. The point is as follows. With proprietary software, the company buys an official copy of the software, the sysadmin loads it onto everyone's PCs, and no one is expected to ever change it. In fact, all the anti-virus software needs to look for is whether anything has changed--if it has, that's a problem. With open source software, on the other hand, everything is expected to change. Individual users are expected to pick up new software or new versions and add it to their systems, even to the point of recompiling the kernal itself. That's a security risk because, it introduces a dozen new avenues for attack. Who's to say that every user has obtained his software from only trusted sources. Of course, the problem is not unique to Linux. Word macros, and IE's VB-Script represent similar open source, "anyone can pick it up and run it" security problems (in many ways, the Word and VB-Scipt problems are worse, because the user can end up running the virus when all he intended to do was browse). This other security concern means that, as others have pointed out, we need better tools for ensuring that we are working with correct versions of the libraries (I forget the details of the exploit that recently reignited this discussion).
Mettler's other point is that when someone tries to discuss these concerns, he runs headlong into the rabid response that "Linux is perfect, therefore there can't possibly be anything wrong with it, and curse you for even suggesting it". Then, when he tries to point that out, he runs into "The Linux community is perfect, therefore we can't possibly be arrogant". I think there are enough posts here to demonstrate that there is some truth to those points as well.
I didn't see any of the earlier discussions, so I don't know if Mettler has become angry and pedantic to the point of leaving the field of reason. What I do know is that we must not ignore the new security issues that are raised by opening the source. Rather than burying our heads in the sand, we must identify the risks, discuss them openly and rationally, then, using the benefits provided by open source, create the means to deal with them.
Rob, I posted this secure in the knowledge that you have nothing better to do than to wait for my suggestions, and cater to my needs :^). Waiting eagerly...
Rob, now that I've gotten started, and I've had time to think, here's what I really want:
Beside each "Read More..." link on the main page, add a second link named "Select...". The "Read More" link would operate as it currently does, based on the user's defaults. The "Select" link would open a selection-criteria page. The story-id would be specified in the URL, but the user would then enter the other filtering criteria. I believe that this would not only give me more flexibility, but would also reduce the load on your system, since users would no longer have to download a page one way (e.g. threaded), then download it again another (flat).
The criteria I would like to be able to specify are:
- Threaded or flat
- Page threshhold (solves AC's problem)
- Headers-only or messages
- Sort by newest/oldest/score
- Newer-than date/time
- First 50 msgs etc. (i.e. first/last msg seq#)
- Later than cid#
The last two points would make it easy to come back later and continue reading. If you wanted to implement it perfectly, the criteria would be "messages listed after, or >= time of, cid#".
Other suggestions:
- Include msg scores on the headers-only display.
- List every msg on the headers-only display, with a flag (*) to indicate which ones meet the filter criteria.
- On the message display, always include the cid#. On the flat display, also show the parent cid#. This would allow users to piece together the thread without having to opening a second brower page for a threaded display (showing hierarchy level#s would normally be an alternative, but it falls apart when messages can be filtered out).
Rob, how about providing a date/time filter field on the comments page. This would allow us to come back to a page later, enter a date/time, and only view the comments that are new, even if they are buried in the hierarchy.
I agree that the $99 charge could be to cover Dell's Linux support costs (although a hidden M$ tax would'nt surprize me either).
I disagree that their support costs are going to turn out to be low, at least not at first.
It's not that Linux is harder to maintain than Windows. My experience agrees with others--Linux is more reliable, and once you get something working, it stays working.
The problem is the learning curve. A large percentage of Dell's customers will be starting with no Linux experience (or Unix for that matter). The fact that Linux is pre-installed will help of course, but Dell is going to have to deal with questions like "How do I get my ADSL modem working?", or "How do I install application X?".
It also won't help that there is a lot of good support on the Internet. Current Linux users are mostly Hackers, and they are, by definition, self-teaching. Future Linux users (not to mention your typical barely competent sysadmin) will not be Hackers, and they will need their hands held.
In order for Linux to realize the low support costs that should accompany its quality, a few things need to happen:
1. Linux distributions need to come with better documentation (Redhat--your book sucks!).
2. Installation procedures for Linux applications need to be simplified, standardized, and documented.
3. There needs to be enough experienced Linux users around that new users can turn to their co-workers, friends, and sons and daughters for help, the way they do today with Windows.
I predict a rosy future for Linux, but we all know that getting there is not going to be a cakewalk, especially when it comes to getting the general public involved.
You're forgetting something. Wiping Microsoft off the face of the earth, as enjoyable, and as beneficial for mankind as that might be, is not the primary goal of Linux. It is, at best, a side benefit, and it's likelihood is questionable.
:^). MS might win or lose in marketshare, and in court.
:^).
Remember--the primary goal is to have a great O/S (plus apps). All that is required for this is a critical mass of users who are dedicated to Linux and open standards.
If MS created their own proprietary version of Linux, then it's true that many, if not most businesses would support it. If MS violated the GPL, the lawsuits would fly (class-action suit--all Linux users please donate $1
In the end, though, none of it would matter. The market would be split between MS Linux (Win 2001) supported by a single vendor, and Standard Linux, supported by many vendors, IBM, Oracle, etc. If you don't think that a critical mass of support would survive for Standard Linux, just look at the Debian crowd
Theories so far:
1. MS wants to make money on the ported Office, plain and simple.
2. MS can't sabotage Linux unless they first get a foot in the door with a Linux app.
3. A flaky port of Office would hurt the credibility of Linux.
4. A promised port of Office that never emerges would hurt the credibility of Linux.
5. MS Office will require proprietary MS libs, allowing MS to gradually take over or corrupt Linux.
6. Office and IE on Linux will help MS to decommoditize HTML and other protocols.
To these we now add the following:
7. Office on Linux is just a lost leader, with the upgrade to Office 2000 requiring a corresponding "upgrade" to Windows 2000 (your theory).
Finally, I would add the following variation:
8. Office on Linux is a desparation ploy, to keep users from migrating away from MS Office (due to Y2K), while MS finishes Windows 2000.
This last theory is interesting. Office 97, Win95, and NT are not Y2K-ready. Windows 2000 will not be ready in time for Y2K. Office beta-2000 may be Y2K-ready, but there is no MS platform to support it. If customers migrate to Linux to solve their Y2K problems, they will also migrate to other office apps, making it nearly impossible for MS to get them back. The only way out of this is to port MS Office (beta-2000) to Linux in order to keep Office users in a holding pattern until W2K plus O2K are ready. It's their only chance.
Of course, there's always all-of-the-above.
1. Never buy what can be safely copied or stolen.
2. Never waste money on research. You can save money by letting others innovate for you.
3. Never compete on the basis of quality. FUD and customer lock-in are cheaper and easier.
4. Never allow your products to become compatible. If others manage to adopt your standards, change them. This will ensure customer lock-in.
5. Never wait until your product is stable before releasing it. By releasing early, you save money on development, while ensuring extra cash for the "upgrade" (e.g. Windows 98).
6. Never provide full backward compatibilty with your own products. Provide just enough to claim compatibility, while ensuring that your customers will eventually be forced to pay you for the upgrade.
7. Ignore your competitor's protocols. If a competitor's protocol looks like it will succeed, embrace and extend that protocol until you destroy it or control it.
8. Never lie about your competitor's products--unless you can get away with it.
10. If a competitor's product runs on your platform, never use a false error message to discredit it. Use an incomprehensible message, and allow the user's doubts to do your work for you.
11. Never publish a complete spec for your platform. Always leave enough undocumented to ensure that your competitor's products can be made incompatible on the next release.
12. Never provide your customers with complete documentation for your products. Complete documentation leads to customers being in control of their systems, and the resulting confidence will interfere with your ability to control their decisions.
You make some excellent points.
Once the masses have caught on to something, it can really hurt (my personal gripe in this regard was FM radio). The famed Linux community support will be severely tested when there are enough poorly-informed Linux users to ensure that most of the questions are about obvious/documented things. Remember, though, that the Internet, unlike radio, is not a mass media. With the benefits of unlimited bandwidth (in the logical sense), and narrow-casting, separate communities will form for the casual Linux users, and the techie Linux developers.
I am not too worried about a non-GPL layer of commercial software running on top of Linux. In fact, I think that such a layer would tend to reinforce the standardization of Linux (of course, then we have to worry about stagnation - the world is always tossing in new challenges), and would provide more motivation for getting rid of the quirks in the various GUI's.
As for Microsoft, if they just became one application supplier among many, then that alone would be a great benefit. I also don't think the fact that MS was making money would kill the Linux spirit. I am a recent convert to Linux, and I made the move as much to escape MS as anything. Now that I am here, however, I have rediscovered the joy of computing. I have come to agree with those who say that Linux is a positive force, and does not need MS as an enemy in order to survive.
Of course, we both know that MS would never be satisfied with competing on even ground. The have the mentality of a thief, and they will always believe that they can tilt the system in their favor, and get away with it. They would follow their usual strategy. The MS Linux apps and distribution would initially start out compatible, but gradually, over time, we would find that MS apps would only run reliably (as much as MS apps ever do) on the MS distribution, or would require proprietary MS libraries. What would happen then? We've already seen what would happen with regards to HTML and the Internet. A grass roots Standard Linux group would form, similar to W3C, and would get the backing of much of the industry. It would once again be Robin Hood versus the bad guys, and the Linux community would be re-energized.
The Internet has changed everything for Microsoft. In the old days, MS could buy half the media, and silence the other half with legal threats. Like all good con men, they were masters at the image game. Today, with the Internet, there are countless individuals who will not be silenced, and the word gets out. Also, unlike the old days when computing was an end in itself, more and more, the network and PCs are becoming a means to an end, i.e just tools. Commercial interests on the Internet want to reach a large audience of consumers. Thus, when MS tries to put up roadblocks, it creates a huge financial incentive to find a way around them, and, unlike the pioneering days, lots of trails have been blazed - too many for MS to control.
It may not be world domination, but I think Linux has a bright future.
Say Microsoft did create their own distribution of Linux? What would happen? Businesses would be faced with a choice between a 100% pure, open-standard, stable Linux, available from multiple trusted vendors, and a more expensive, proprietary, less-stable Linux, supplied by a single, sneaky vendor. Given that Microsoft would not have their current IBM-deeded head start, I'll bet that the open standard would win, just as is now happening with Corba, Java, and the Internet in general.
Let's say I'm wrong. Say Microsoft once again sucked the PHBs into using their adulterated Linux. What will have changed? What's the difference between a proprietary MS Windows, and a proprietary MS Linux? The answer is: no difference! The Linux community will carry on, continuing to support open-standard, open-source Linux. And, to those who argue that an MS bizarro-Linux will hurt the credibility of Linux, I would answer that, in the eyes of the PHBs, MS will have legitimized it, thus making the leap to real Linux even easier.
I'm not worried.
I wasn't betting that Linux would kill Microsoft - I was betting that Linux would fulfill my computing needs, now and in the future. Even if none of the promised software support comes from the major companies, such as IBM, I will still have won.
As to Microsoft, they will ultimately die of their own poison and lies.
The FSF is against information ownership,..., they don't seem to oppose capitalism in other areas...
You are right, of course, about the FSF. My comments were aimed more toward those Slashdot posters who take the OSS phenomenon as a validation of Socialism in general (and who really do mean free as in free lunch).
As to the FSF, I can't agree with their goal of ending information ownership. Like you, I think they are trying to live a contradiction. In fact, I consider that part of their philosophy to be dangerous. Note that, while they want to to remove my IP rights, the reason that I choose not to violate the GPL, whether it's legally enforced or not, is out of respect for their IP rights.
Despite my differences with the FSF, I continue to support Linux and other OSS projects. It makes sense for the operating system to be open, and it's an excellent free-market solution to the threat of a Microsoft monopoly (and it's fun, and you meet nice people). As another poster pointed out (smithdog, below), cooperation has always gone hand-in-hand with Capitalism. There is no reason why free individuals, working toward a common goal, can't share the fruits of their labor, while still respecting each other's, and everyone else's, property rights.
The popular understanding of Capitalism has become so corrupted, it's hard to discuss it intelligently anymore. This was largely done on purpose (which is not to say there is a conscious plan involved) by the major segments of society whose power and money would be threatened by true Capitalism, including Politicians, University Professors, Union Leaders, Bankers, and Big (so-called) Businessmen. Despite what you have heard, Capitalism is not about international money-changing, nor can it be characterized by pointing to the problems created by the huge government-regulated, government-controlled stock market and public corporations.
Capitalism is about private ownership of Capital. The central point is the private ownership, which is closely linked to freedom. This means that each person is entitled to (i.e. owns) the results of his own work (unless he sells it or trades it for pay), and the definition of ownership is that you get to say what can or cannot be done with your property.
The GPL is a license setting down the wishes of the creators of GNU and other OSS software. We expect those wishes to be honored. We even expect the courts to enforce the license. This implies that the creators have a legal right to set the rules for their creation. That, in turn, implies ownership (even is the ownership is subsequently transferred to the public), which is a Capitalist concept, whether some OSS-ers think so or not.
Under Socialism, there is no ownership. Everything is shared by society, for the good of the people (or the state). Of course, there still needs to be rules to manage the sharing, otherwise, what's to stop me and my guests from using up the entire world's supply of Truffles and Dom Perignon. Thus, whenever Socialism is implemented, it breaks down into Ownership of Everything by the Government.
Under Socialism, the GPL would be ignored, since it only represents the wishes of a few private individuals. Under Socialism, the Government could decide that it was in the best interest of society that some companies (say those located in Redmond) be allowed to provide custom versions, and binaries only, and there would be nothing you could say about it. Socialism would destroy the GPL and the OSS community.
If you really want a better understanding of Capitalism, read Milton Friedman's Free To Choose. It's insightful, entertaining, and written for the general public. For the more philosophically inclined, I would also suggest Ayn Rand's Capitalism: The Unknown Ideal (though I risk re-igniting a recent flame war - some here think Rand is full of crap, but I'm definitely not one of them).
To summarize: If you support the GPL, then you must support Capitalism, because that's the only system under which the GPL could exist.