"Supporting" UNIX clients means testing against them and dealing with issues that come up with technical support. That's expensive, and it is sensible for HP not to do that.
The box may or may not still accept UNIX line printer connections. If I were putting together the product, I would disable that because it raises all sorts of additional security issues.
You can almost certainly print to the box using SMB from a UNIX client. But, of course, if it doesn't work, you have to figure out yourself why.
I think you could make an argument that Verant can't prohibit you from doing this based on their EULA. But if they are sufficient persistent, even if you can make that argument, getting your way in court would be a lengthy and costly process.
What I don't understand is why you bothered doing this in the first place. It was quite predictable that Verant wouldn't like this and might take you to court. And you are cloning a server for a proprietary game with proprietary clients, helping increase the popularity of the very company that's now causing you legal problems.
In some areas, interoperability is important. Open source software really does need to be able to clone things like the MS Office file format and UI. But for something like Everquest, it would have been so much better to start from scratch with building a game that is similar and in the same genre (you could have based the client on any of a number of standalone open source games).
I think this project is not worth spending much time or effort supporting for the open source and hacker community. It may set dangerous legal precedents, consume lots of legal resorces, and cost us good will. In fact, it doesn't even strike me as something that is technically particularly interesting to "hack"--what kind of interesting technical principles are you going to discover?
My recommendation would be this: either stop the development for good and start an open source game effort to produce something in the same genre, or make what you have incompatible with EverQuest and develop your own client instead. I think there are more important and more rational battles to fight.
Why again do you think you have to record all E-mail? Are you supposed to listen in on all telephone conversation and bug people's offices as well? Who is going to pay for the effort that that kind of monitoring requires?
I don't think monitoring is feasible. In fact, it may you expose to even more liability because it puts you in the position of being able to discover problems, and the presumption then may be that you knew about a problem but chose to ignore it.
I'd prohibit any personal use of company E-mail (there is no need for it--web-based mailers provide an excellent alternative), have a clear policy on how employees can get help with problems, and indicate to external recipients of E-mail messages (in a header or signature) who they can contact in case of problems with mail they received. But if it really worries you, why not talk to a lawyer?
I agree: the Cato institute doesn't parrot the agenda of corporate America. But they are clearly not without an agenda: they push a brand of libertarianism, and their political and philosphical beliefs suffuse any argument they are trying to make.
Whether their arguments are well thought out or merely drivel, you have to judge for yourself. In my opinion, most of their arguments are intellectually lightweight, based on polemics and hot button phrases, without deeper reflection, sound economic analysis, or hard data. They appear to be arguing and judging from the position of reasonably well off middle class academics, often missing the point behind the social policies they are pontificating about entirely.
The number of applications doesn't matter much when it comes to barriers to entry. What matters is the nature and use of the applications.
Microsoft has some very "sticky" software, including Microsoft Office, financial software, custom developed software, proprietary codecs, and proprietary device drivers. The cost for any customer to switch away from those to a competing product is prohibitive, both in terms of disruption of business processes, retraining, and replacement. With that kind of cost structure, Microsoft could afford to be much worse or much more expensive than anybody else and people still wouldn't switch.
In fact, that's what you see: StarOffice, Lotus, and other packages are quite usable and essentially free, yet people stick with Microsoft office.
Even if microgravity were essential for some experiment, you don't need a manned space station in order to perform them: the experiments can be automated, or you can carry them out using telepresence.
You also mischaracterize the particular experiment. It isn't about "fuel efficiency", it's about getting more gasoline from a barrel of oil by improving the efficiency of cracking (the "hydrogen storage" angle in the article is even more of a long shot). That doesn't improve energy efficiency or greenhouse gas emissions. There are much simpler ways of achieving better fuel efficiency, foremost by taxing gas guzzlers like SUVs.
As for light pollution, you can escape from city lights by putting your telescopes in remote places, but you can't escape from shiny objects orbiting the earth. At best, you can try to avoid having them pass through your field of view.
That means even more light pollution. And if it has reflective surfaces (solar panels), we can expect a glint from it as well.
I just can't get terribly excited about the ISS. I think we would get much more bang for the buck with unmanned missions and research on new lift vehicles and propulsion systems.
I recently donated a lot of computer equipment and called around for computer-oriented non-profits in the SF Bay Area. The experience was depressing.
From the donation guidelines and other information, it is pretty clear that these organizations were largely only interested in machines that they could run recent version of Windows and MacOS on. Organizations that provided training for nonprofits were largely only looking for volunteer training for Windows.
In part, this is probably because Microsoft and other companies provide Windows software free or at a relatively low cost to nonprofits and educational institutions. This means that the most easily understood argument in favor of Linux, "it's free", isn't really that convincing.
I think nonprofits and schools should be very interested in Linux. Linux runs on low end hardware, it is stable, it doesn't require a lot of retraining from version to version, and it uses mature, industry standard, open APIs. And in particular at schools, we want students to be exposed to something other than Windows administration and the Windows APIs. The best way for Linux proponents to address that is through targetted donations and volunteering.
Whoever is at fault, it remains an example of silly restrictions on the kind of equipment you can legally buy.
But I don't believe that the cell phone industry is prevented by law to do a better job on security. If cryptography was the issue, they could have worked out some deal, for example, giving keys to law enforcement. And they don't even bother much with security in areas where it clearly matters: authentication and fraud prevention. The export restrictions are also a red herring: most cell phone equipment is imported, not exported, and incorporating encryption can be just a question of reprogramming the software, something that needs to be done anyway for export.
Yes, with current hardware, Sony doesn't stand a chance with their vision. But Sony and their allies are a big companies with lots of money and lots of influence. They can influence future hardware and software standards, as well as future legislation that affects what computing devices will be legal.
A lot of stupid laws restricting technology have been passed, both here and in Europe. For example, in the US, you cannot buy a receiver that covers the analog cell phone bands anymore (although it's easy for a criminal to put one together). Why? Because rather than holding the cellular companies responsible for their appalling lack of security, the cellular companies prevailed on congress to simply outlaw the production of receivers. It didn't make the world much safer, but it removed the responsibility from them and put the burden on the tax payer and consumer.
Or consider that in Europe, people pay a tax on blank tapes, money that is then shipped directly to the music industry. The presumption is that you use tapes for illegal copying, so you might as well pay the "legitimate artists" for that.
And, of course, in the US, the industry already succeeded in getting cumbersome copy protection devices into digital audio systems.
Don't take the current situation for granted, where computer hardware is reasonably open, well documented, and programmable. We may well end up with a situation where most people use proprietary, limited hardware (like a future PlayStation) with tightly controlled interfaces and software, and in which other people have to pay a steep premium for "professional" equipment that is programmable; and even that kind of "professional equipment" may be tightly controlled.
It's pretty clear that Sony and other content providers are going to fight tooth and nail to protect their obsolete business models. We need to be vigilant to keep them from succeeding. Sony needs to figure out how to make money in a world of open hardware and cheap distribution; if they can't do that, they should go out of business. As a society, we have no obligation to protect companies that can't adapt to new technological realities.
Sure it is, it just isn't MacOS
on
Is UNIX An OS?
·
· Score: 2
Every should go back and read up on computing history. Operating systems have come with user interfaces, productivity suites, tools, typesetting, and other components since long before even the original Macintosh was released. Even a MacOS 9 or Windows NT installation comes with a lot less functionality than, say, a BSD UNIX distribution from 1987 or an IBM mainframe from the 70's. One of the reasons many UNIX users found (and continue to find) MacOS and Windows such pathetic environments is because they come with so few tools and because the tools on those platforms are so poorly integrated. For serious applications, integration needs to go beyond being able to drag a file from here to there or to cut and paste a snippet of text.
Even if the yardstick were an MS/Apple-style GUI-oriented desktop, Linux with KDE or Gnome hardly has to fear comparison. Linux distributions are still more complete and have a much larger and much better integrated toolset than MacOS X is likely to have.
What Every is really saying is that he likes MacOS. To each their own. But to hide Macintosh advocacy behind "what is an operating system?" is silly. And given how stripped down MacOS is and continues to be, he's on the losing end of that argument anyway.
Fortran 77 is a lousy language. Fortran 9x, on the other hand, is quite decent for numerical computations.
C doesn't have efficient array bounds checking, multidimensional arrays, parallel array operations, or a host of other features useful for numerical computations. The workarounds people use in C are inefficient and/or cumbersome.
I think most people agree that, while USB 1 was excellent as a low cost device interconnect, USB 2.0 is a stupid "me-too" kind of product--technically inferior to FireWire and likely to be not significantly cheaper. So what are we going to do about it?
Buy FireWire for your external interface needs and contribute to the Linux FireWire effort. FireWire devices are here. You can pick up FireWire PCI cards for under $100, and external FireWire disk enclosures taking an IDE drive for $150. There are some scanners, CD-RWs, and other peripherals, in addition to the A/V FireWire devices. FireWire is quickly becoming a feasible substitute for SCSI.
You mean that CD that will be obsolete in a few years because the industry wants to see you the same music all over again in a new format? And the CD whose metallic layer will oxidize away? The CDs that only fit into large, cumbersome portable CD players?
I paid for the music once, and I don't intend to buy it again. I have converted my CDs into MP3 so that I can keep them around easily and use them in a small portable player. I have also converted vinyl to MP3 because some classical recordings aren't even available anymore in modern formats.
All of that is fair use of content that I paid for, not piracy. If the music industry wants to go to a complete rental model for music, they better start by lowering their prices drastically. Otherwise, their are engaging in fraud: claiming to sell you content while knowing full well that its lifetime is limited to a few years.
Will KDE ever have
a corporate-backed "foundation" deciding it's future?
I think it always has: in a sense, KDE is "backed by" Troll Tech, and it seems to me that KDE's future has already been influenced strongly through that, starting with the discontinuation of Harmony.
Your response is generally "well, AFS doesn't do this, but you shouldn't be doing that anyway".
The fact is that people do deal with gigabyte files over networked file systems (video editing, scientific datasets, server logs, etc.), they do run UNIX installations that don't use AFS Kerberos as their authentication method, they do need to create named pipes and UNIX domain sockets on networked partitions, and they do expect that UNIX access semantics are preserved when using remote files. AFS fails to deliver on all of those. The AFS designers simply thumbed at UNIX semantics and didn't give a damn.
NFS does deliver on all those points. For small and mid-size installations, NFS management is pretty simple, and NFS security is getting better and less of a problem with switched Ethernets anyway.
The suggestion that AFS is good for content distribution to web server farms also strikes me as silly. Installing AFS to achieve synchronization between web servers is like driving a truck to pick up a carton of milk. rsync and similar tools are much simpler to deploy and much more flexible.
People can get excited about whatever they want, and if AFS makes you happy, great! I have used AFS for many years, and my recommendation is that people should look at its incompatiblities and quirks very carefully. I think for most UNIX environments, it is not a very good choice.
But what this announcement may mean is that, after years of neglect, maybe people will roll up their sleeves and fix those rather serious problems that AFS has. Then, it could potentially become a good distributed file system. Until then, it is a solution that, in its own way, is as flawed as NFS, and quite a bit more work to manage.
If you want to keep a bunch of web sites synchronized over a WAN, rsync is much simpler to deploy (it's just a user program) and considerably more efficient than AFS. You also get much more control over what gets updated where and when.
For serious wide area, distributed authoring, no distributed file system, no matter how good, is going to be adequate by itself. Distributed authoring requires workflow support, version and revision control, support for disconnected operation, and other features. For that, something like WebDAV or CVS are more appropriate choices.
Don't get me wrong: AFS isn't all bad. Some of its core ideas are really great. But some of its practical aspects (e.g., differences in semantics from UNIX, simplistic caching strategy) make it a pain and rather inefficient in many real world settings. There are some areas where people can live with those limitations (e.g., university computer labs), but I think for most environments, NFS and SMB, despite their many warts, are still more practical systems.
Of course, IBM uses AFS extensively internally, as they have been for many years, and there are environments where people like and use AFS.
My point was not that IBM has somehow discontinued AFS (clearly they haven't), my point was that IBM themselves has recognized the technical problems that AFS has and tried to address many of them with DFS.
Porting IE, Office, and other Microsoft products to Linux is likely pocket change for Microsoft, in particular when they can outsource it and use a compatibility library.
The question is whether and in what form they will release those products, or how else they will use them. They might, for example, ship substandard versions of IE simply to have an argument for management of the form "yes, you can standardize on IE, it runs on Windows, Mac, Linux, and Solaris". Or they might want to harm third party office suite efforts (open source or proprietary), for example by giving pre-release demos, maybe releasing a so-so version of Office, etc.
You can bet that whatever Microsoft will be doing with these ports, it will be done to increase their business and harm competitors, including Linux and Mozilla, and push their proprietary protocols and formats. Altogether, I think it's best not to get distracted by Microsoft and ignore what they are doing. If writing Gnome Office or KOffice was a good idea a week ago, it still ought to be a good idea today.
The Olympics seems no different to me than Coca Cola or Tommy Hilfiger: it's a brand that sells youth and a lifestyle to consumers. They are a hugely profitable international commercial enterprise with questionable accounting and financial practices.
Are you going to get all pushed out of shape about Coca Cola commercials not being broadcast on the web? So why worry about what happens with the Olympics? Do you really care whether one doped athlete beats another one by 1/100's of a second?
You don't need an expensive, canned media extravaganza to aspire to the "olympic ideals". Participate in sports yourself, travel, learn languages, and get to know people. You'll be healthier and do more for the world than the "official" Olympics ever did.
You are putting up a strawman. Reality is that copyright legislations has strongly restricted access and use of information since the 19th century, and that at the end of the 20th century, technology is being used to make a mockery of even the limited fair use provisions copyright law still had left.
People in a democratic society need to be able to exchange information. A vibrant culture needs artists that can build on each other's work and that can reuse cultural icons (as we know from several hundred years of experience with the arts and culture in Europe). A society and culture in which every bit of information and every cultural icon is owned in perpetuity by corporations with specific financial interests and concerns about their "brand image" is a dead end.
It is this tradeoff that is at the core: should we put corporate profits over the interests of our society? Or should we be conservative and return to the roots of copyright law: to fair use provisions that allows individuals to use content, build on each other's ideas, and reproduce content for scientific and analytical purposes, and to limited (20-30 years) protection?
To me, the answer is pretty clear. Kaplan is right that this needs to be done by the legislative branch, not by the judicial branch of government, but without a serious political finance refort, that is unlikely.
AFS semantics are very different from UNIX file system semantics: permissions are associated with directories only, access is determined only by the containing directory, if multiple clients modify the same file, updates are lost, you can't have any special files in an AFS file system, etc. AFS uses its own authentication, it doesn't work well for big files, it always requires extra work to get it to work with daemons, and it has severe problems for scientific compute clusters. IBM has long ago moved onto DFS (unrelated to Microsoft DFS), which fixes many of the problems of AFS (but is itself big, even more complex than AFS, and hard to administer). Many places are trying to get rid of AFS because it's just too much of a hassle to run it (and converting back to a UNIX file system isn't easy because AFS encourages permissions and ACLs to mushroom unnecessarily).
AFS may be acceptable for specific applications (in fact, what it was designed for originally): a large untrusted user population, dedicated system management staff, and smallish files and problems (text file editing, small programming jobs). But for many environments where Linux is used--big software development projects, web servers, scientific computing, home networking--it just doesn't seem like a good fit.
If it's the security you care about, NFSv4 might be for you, although it clearly also has some problems. If you want something AFS-like, Coda might be an option (but I don't know how mature it is yet). MFS and GFS are options for compute clusters. Maybe we can get 9P or Styx up on Linux.
Isn't that what Microsoft wants? Kill competing efforts? Microsoft has been operating this way for as long as they have been around. If they ever deliver a product, it will likely be years down the road.
But, yes, Microsoft is probably dropping a little bit of money (percentage-wise, less than you or I spend on a Latte) on getting Office ported to Linux. Why not? They can do it just to cover their bases and "learn about the enemy"; they don't ever have to release it. And if they do decide to release it, they can release it in a form that leaves their Linux customers begging for the Windows version.
Altogether, I think people in the Linux community shouldn't get confused by these kinds of rumors. Pick a goal and follow it. I don't think that reimplementing a Windows environment (KDE or Gnome) on top of Linux is a particularly good goal, but if that's what tickles your fancy, at least don't get distracted by Microsoft's usual business and marketing practices; you should know about them by now.
It's great that Debian 2.2 is out, and I appreciate all the testing that has gone into it. I also liked the package system and the installation on Debian 2.1. The floppy-based install is a real boon on oddball hardwawre.
But with the infrequent releases, Debian just became too outdated for me. Because I have some machines that don't have an Internet connection or only have a slow Internet connection, the latest official release on CD is sometimes the only thing I can install. I haven't found any "Debian packages up to last month" kind of CD distributions (if you know of any, please let me know).
Also, lack of interim version numbers means that Debian can be hard to use in an environment with many users who handle their own installations, yet want to share software: if you want to tell other people to bring up their machines to a known level to run some software, what are you going to tell them? If operating system releases are infrequent, the last numbered release is often too out of date.
I hope Debian will be able to stamp version numbers on interim releases, maybe every other month. This shouldn't involve any significant extra work, just a declaration "these packages are Debian 2.1.7". CD-ROM vendors can then burn those versions into CDs and ship them out, and everybody has a reference point. From my point of view, those interim releases don't have to be extensively tested: if there is some real problem with some release (which is probably rare), that will become widely known and people will avoid it.
Various universities and scientific publishers "require" copyright transfers. You may be able to negotiate these and at least retain the right to distribute the stuff yourself.
In fact, I have no problem granting UMI, my university, or a journal a transferable right to publish my thesis or papers. For me, the problem occurs when those institutions want to prohibit me from distributing or reusing my own work.
Slashdot Mirror
The box may or may not still accept UNIX line printer connections. If I were putting together the product, I would disable that because it raises all sorts of additional security issues.
You can almost certainly print to the box using SMB from a UNIX client. But, of course, if it doesn't work, you have to figure out yourself why.
What I don't understand is why you bothered doing this in the first place. It was quite predictable that Verant wouldn't like this and might take you to court. And you are cloning a server for a proprietary game with proprietary clients, helping increase the popularity of the very company that's now causing you legal problems.
In some areas, interoperability is important. Open source software really does need to be able to clone things like the MS Office file format and UI. But for something like Everquest, it would have been so much better to start from scratch with building a game that is similar and in the same genre (you could have based the client on any of a number of standalone open source games).
I think this project is not worth spending much time or effort supporting for the open source and hacker community. It may set dangerous legal precedents, consume lots of legal resorces, and cost us good will. In fact, it doesn't even strike me as something that is technically particularly interesting to "hack"--what kind of interesting technical principles are you going to discover?
My recommendation would be this: either stop the development for good and start an open source game effort to produce something in the same genre, or make what you have incompatible with EverQuest and develop your own client instead. I think there are more important and more rational battles to fight.
I don't think monitoring is feasible. In fact, it may you expose to even more liability because it puts you in the position of being able to discover problems, and the presumption then may be that you knew about a problem but chose to ignore it.
I'd prohibit any personal use of company E-mail (there is no need for it--web-based mailers provide an excellent alternative), have a clear policy on how employees can get help with problems, and indicate to external recipients of E-mail messages (in a header or signature) who they can contact in case of problems with mail they received. But if it really worries you, why not talk to a lawyer?
Whether their arguments are well thought out or merely drivel, you have to judge for yourself. In my opinion, most of their arguments are intellectually lightweight, based on polemics and hot button phrases, without deeper reflection, sound economic analysis, or hard data. They appear to be arguing and judging from the position of reasonably well off middle class academics, often missing the point behind the social policies they are pontificating about entirely.
Microsoft has some very "sticky" software, including Microsoft Office, financial software, custom developed software, proprietary codecs, and proprietary device drivers. The cost for any customer to switch away from those to a competing product is prohibitive, both in terms of disruption of business processes, retraining, and replacement. With that kind of cost structure, Microsoft could afford to be much worse or much more expensive than anybody else and people still wouldn't switch.
In fact, that's what you see: StarOffice, Lotus, and other packages are quite usable and essentially free, yet people stick with Microsoft office.
You also mischaracterize the particular experiment. It isn't about "fuel efficiency", it's about getting more gasoline from a barrel of oil by improving the efficiency of cracking (the "hydrogen storage" angle in the article is even more of a long shot). That doesn't improve energy efficiency or greenhouse gas emissions. There are much simpler ways of achieving better fuel efficiency, foremost by taxing gas guzzlers like SUVs.
As for light pollution, you can escape from city lights by putting your telescopes in remote places, but you can't escape from shiny objects orbiting the earth. At best, you can try to avoid having them pass through your field of view.
I just can't get terribly excited about the ISS. I think we would get much more bang for the buck with unmanned missions and research on new lift vehicles and propulsion systems.
From the donation guidelines and other information, it is pretty clear that these organizations were largely only interested in machines that they could run recent version of Windows and MacOS on. Organizations that provided training for nonprofits were largely only looking for volunteer training for Windows.
In part, this is probably because Microsoft and other companies provide Windows software free or at a relatively low cost to nonprofits and educational institutions. This means that the most easily understood argument in favor of Linux, "it's free", isn't really that convincing.
I think nonprofits and schools should be very interested in Linux. Linux runs on low end hardware, it is stable, it doesn't require a lot of retraining from version to version, and it uses mature, industry standard, open APIs. And in particular at schools, we want students to be exposed to something other than Windows administration and the Windows APIs. The best way for Linux proponents to address that is through targetted donations and volunteering.
But I don't believe that the cell phone industry is prevented by law to do a better job on security. If cryptography was the issue, they could have worked out some deal, for example, giving keys to law enforcement. And they don't even bother much with security in areas where it clearly matters: authentication and fraud prevention. The export restrictions are also a red herring: most cell phone equipment is imported, not exported, and incorporating encryption can be just a question of reprogramming the software, something that needs to be done anyway for export.
A lot of stupid laws restricting technology have been passed, both here and in Europe. For example, in the US, you cannot buy a receiver that covers the analog cell phone bands anymore (although it's easy for a criminal to put one together). Why? Because rather than holding the cellular companies responsible for their appalling lack of security, the cellular companies prevailed on congress to simply outlaw the production of receivers. It didn't make the world much safer, but it removed the responsibility from them and put the burden on the tax payer and consumer.
Or consider that in Europe, people pay a tax on blank tapes, money that is then shipped directly to the music industry. The presumption is that you use tapes for illegal copying, so you might as well pay the "legitimate artists" for that.
And, of course, in the US, the industry already succeeded in getting cumbersome copy protection devices into digital audio systems.
Don't take the current situation for granted, where computer hardware is reasonably open, well documented, and programmable. We may well end up with a situation where most people use proprietary, limited hardware (like a future PlayStation) with tightly controlled interfaces and software, and in which other people have to pay a steep premium for "professional" equipment that is programmable; and even that kind of "professional equipment" may be tightly controlled.
It's pretty clear that Sony and other content providers are going to fight tooth and nail to protect their obsolete business models. We need to be vigilant to keep them from succeeding. Sony needs to figure out how to make money in a world of open hardware and cheap distribution; if they can't do that, they should go out of business. As a society, we have no obligation to protect companies that can't adapt to new technological realities.
Even if the yardstick were an MS/Apple-style GUI-oriented desktop, Linux with KDE or Gnome hardly has to fear comparison. Linux distributions are still more complete and have a much larger and much better integrated toolset than MacOS X is likely to have.
What Every is really saying is that he likes MacOS. To each their own. But to hide Macintosh advocacy behind "what is an operating system?" is silly. And given how stripped down MacOS is and continues to be, he's on the losing end of that argument anyway.
C doesn't have efficient array bounds checking, multidimensional arrays, parallel array operations, or a host of other features useful for numerical computations. The workarounds people use in C are inefficient and/or cumbersome.
Buy FireWire for your external interface needs and contribute to the Linux FireWire effort. FireWire devices are here. You can pick up FireWire PCI cards for under $100, and external FireWire disk enclosures taking an IDE drive for $150. There are some scanners, CD-RWs, and other peripherals, in addition to the A/V FireWire devices. FireWire is quickly becoming a feasible substitute for SCSI.
You mean that CD that will be obsolete in a few years because the industry wants to see you the same music all over again in a new format? And the CD whose metallic layer will oxidize away? The CDs that only fit into large, cumbersome portable CD players?
I paid for the music once, and I don't intend to buy it again. I have converted my CDs into MP3 so that I can keep them around easily and use them in a small portable player. I have also converted vinyl to MP3 because some classical recordings aren't even available anymore in modern formats.
All of that is fair use of content that I paid for, not piracy. If the music industry wants to go to a complete rental model for music, they better start by lowering their prices drastically. Otherwise, their are engaging in fraud: claiming to sell you content while knowing full well that its lifetime is limited to a few years.
I think it always has: in a sense, KDE is "backed by" Troll Tech, and it seems to me that KDE's future has already been influenced strongly through that, starting with the discontinuation of Harmony.
The fact is that people do deal with gigabyte files over networked file systems (video editing, scientific datasets, server logs, etc.), they do run UNIX installations that don't use AFS Kerberos as their authentication method, they do need to create named pipes and UNIX domain sockets on networked partitions, and they do expect that UNIX access semantics are preserved when using remote files. AFS fails to deliver on all of those. The AFS designers simply thumbed at UNIX semantics and didn't give a damn.
NFS does deliver on all those points. For small and mid-size installations, NFS management is pretty simple, and NFS security is getting better and less of a problem with switched Ethernets anyway.
The suggestion that AFS is good for content distribution to web server farms also strikes me as silly. Installing AFS to achieve synchronization between web servers is like driving a truck to pick up a carton of milk. rsync and similar tools are much simpler to deploy and much more flexible.
People can get excited about whatever they want, and if AFS makes you happy, great! I have used AFS for many years, and my recommendation is that people should look at its incompatiblities and quirks very carefully. I think for most UNIX environments, it is not a very good choice.
But what this announcement may mean is that, after years of neglect, maybe people will roll up their sleeves and fix those rather serious problems that AFS has. Then, it could potentially become a good distributed file system. Until then, it is a solution that, in its own way, is as flawed as NFS, and quite a bit more work to manage.
For serious wide area, distributed authoring, no distributed file system, no matter how good, is going to be adequate by itself. Distributed authoring requires workflow support, version and revision control, support for disconnected operation, and other features. For that, something like WebDAV or CVS are more appropriate choices.
Don't get me wrong: AFS isn't all bad. Some of its core ideas are really great. But some of its practical aspects (e.g., differences in semantics from UNIX, simplistic caching strategy) make it a pain and rather inefficient in many real world settings. There are some areas where people can live with those limitations (e.g., university computer labs), but I think for most environments, NFS and SMB, despite their many warts, are still more practical systems.
My point was not that IBM has somehow discontinued AFS (clearly they haven't), my point was that IBM themselves has recognized the technical problems that AFS has and tried to address many of them with DFS.
The question is whether and in what form they will release those products, or how else they will use them. They might, for example, ship substandard versions of IE simply to have an argument for management of the form "yes, you can standardize on IE, it runs on Windows, Mac, Linux, and Solaris". Or they might want to harm third party office suite efforts (open source or proprietary), for example by giving pre-release demos, maybe releasing a so-so version of Office, etc.
You can bet that whatever Microsoft will be doing with these ports, it will be done to increase their business and harm competitors, including Linux and Mozilla, and push their proprietary protocols and formats. Altogether, I think it's best not to get distracted by Microsoft and ignore what they are doing. If writing Gnome Office or KOffice was a good idea a week ago, it still ought to be a good idea today.
Are you going to get all pushed out of shape about Coca Cola commercials not being broadcast on the web? So why worry about what happens with the Olympics? Do you really care whether one doped athlete beats another one by 1/100's of a second?
You don't need an expensive, canned media extravaganza to aspire to the "olympic ideals". Participate in sports yourself, travel, learn languages, and get to know people. You'll be healthier and do more for the world than the "official" Olympics ever did.
People in a democratic society need to be able to exchange information. A vibrant culture needs artists that can build on each other's work and that can reuse cultural icons (as we know from several hundred years of experience with the arts and culture in Europe). A society and culture in which every bit of information and every cultural icon is owned in perpetuity by corporations with specific financial interests and concerns about their "brand image" is a dead end.
It is this tradeoff that is at the core: should we put corporate profits over the interests of our society? Or should we be conservative and return to the roots of copyright law: to fair use provisions that allows individuals to use content, build on each other's ideas, and reproduce content for scientific and analytical purposes, and to limited (20-30 years) protection?
To me, the answer is pretty clear. Kaplan is right that this needs to be done by the legislative branch, not by the judicial branch of government, but without a serious political finance refort, that is unlikely.
AFS may be acceptable for specific applications (in fact, what it was designed for originally): a large untrusted user population, dedicated system management staff, and smallish files and problems (text file editing, small programming jobs). But for many environments where Linux is used--big software development projects, web servers, scientific computing, home networking--it just doesn't seem like a good fit.
If it's the security you care about, NFSv4 might be for you, although it clearly also has some problems. If you want something AFS-like, Coda might be an option (but I don't know how mature it is yet). MFS and GFS are options for compute clusters. Maybe we can get 9P or Styx up on Linux.
But, yes, Microsoft is probably dropping a little bit of money (percentage-wise, less than you or I spend on a Latte) on getting Office ported to Linux. Why not? They can do it just to cover their bases and "learn about the enemy"; they don't ever have to release it. And if they do decide to release it, they can release it in a form that leaves their Linux customers begging for the Windows version.
Altogether, I think people in the Linux community shouldn't get confused by these kinds of rumors. Pick a goal and follow it. I don't think that reimplementing a Windows environment (KDE or Gnome) on top of Linux is a particularly good goal, but if that's what tickles your fancy, at least don't get distracted by Microsoft's usual business and marketing practices; you should know about them by now.
But with the infrequent releases, Debian just became too outdated for me. Because I have some machines that don't have an Internet connection or only have a slow Internet connection, the latest official release on CD is sometimes the only thing I can install. I haven't found any "Debian packages up to last month" kind of CD distributions (if you know of any, please let me know).
Also, lack of interim version numbers means that Debian can be hard to use in an environment with many users who handle their own installations, yet want to share software: if you want to tell other people to bring up their machines to a known level to run some software, what are you going to tell them? If operating system releases are infrequent, the last numbered release is often too out of date.
I hope Debian will be able to stamp version numbers on interim releases, maybe every other month. This shouldn't involve any significant extra work, just a declaration "these packages are Debian 2.1.7". CD-ROM vendors can then burn those versions into CDs and ship them out, and everybody has a reference point. From my point of view, those interim releases don't have to be extensively tested: if there is some real problem with some release (which is probably rare), that will become widely known and people will avoid it.
In fact, I have no problem granting UMI, my university, or a journal a transferable right to publish my thesis or papers. For me, the problem occurs when those institutions want to prohibit me from distributing or reusing my own work.