Seems to me an easy (well, maybe not always practical, but I can't think of much alternative) answer to this would be, in cases where this kind of thing is expected, to have the appropriate sort at Andover contact the site owner ahead of time with some kind of release for short-term mirroring. (Honestly, if you look at it, most of this type of content will keep for a few days before posting anyway, and it saves all manner of sysadmins the headache of scrambling to keep up with the Slashdot effect.)
The complaint I keep seeing over and over in this is that people can't recover their data. To this, the answer seems simple enough: for static content and scripts, always keep a local copy. For databases, etc., don't rely on the provider's advertised backups - mirror the data, preferably at least nightly, even if it means fighting your provider to get them to allow you access to those files. This is a job that can be easily performed by a script (though admittedly, depending on your data, it could get bandwidth-intensive).
As for other services (mail, DNS, etc.), my best advice is what's been offered before: don't put all your eggs in one basket. Sure, it costs more to have two or three different services for mail, DNS and web hosting, but it allows you a smoother transition should one provider suddenly prove unreliable.
A final word on this: in some cases in just makes better sense to host your own servers. I don't really think this option makes sense for high-volume web operations in small companies (I personally don't have the pocket money to shell out for an OC3), but in cases where you've got a pipe and an IP block anyway, it might just make good sense to set yourself up a mail server and a DNS, even if they're just fallbacks from what your hosting service provides. That way, when whatever MCSE they've hired as admin comes along and tries this new fdisk command he just learned on the primary DNS (apologies to all the MCSEs out there who actually know what they're doing), your operation still has a decent chance of surviving, even if it does take a speed hit for a while.
I moderated him down because I hate Tom. Do I need any other reason?
Well, I was hoping for something more like "I moderated him down because his argument seemed pointless." Would've at least been more like what moderation's supposed to be for - scoring the usefulness of the comment instead of just whether you like the opinion and/or the poster.
In all truth, though, I don't know that there's much to be done to fix the moderation system. Like it or not, it does seem to improve S/N in a mostly fair and reasonable manner. And granted, there are those who'll abuse it (show me a way to create an abuse-proof situation with any sentient interaction and I'll show you something likely not worth using).
Finally, in Tom's defense, I've found his to be some of the best-constructed, most sensible arguments on Slashdot, even when I disagree with his views. And that is something that Slashdot is all too sorely lacking.
A lot of the time, KDE and GNOME seem to follow somewhat-different courses to very similar ends. I understand there are probably some philosophical differences in the code used and so on, but from a functional point of view, would the projects benefit from something a little more cooperative, perhaps creating a modular environment where users can pick and choose their functionality, or where the underlying interface could be tailored to accomodate whatever methods a particular developer preferred?
Honestly, I've got to look at this and say that the class-action suits on the consumer level aren't necessarily going to be very much of a concern for MS. As has been pointed out, the vast majority of Windows installations are bundled packages, not people buying individual CDs. This makes it difficult to come by a preponderance of evidence showing harm to consumers. At the corporate level, I think there's a more real threat, but I don't really personally see much potential in the consumer suits.
A little more difficult, though. X-Win32 (also MI/X, eXceed, and probably a few I've missed) is simply an X server, designed to implement X11 (already closely tied to IP networking) over the Windows GDI. The more difficult proposition is intercepting GDI calls (designed to run with a local display) and actually convert those to X calls. It requires implementation at a deeper level, essentially adding a third interface (network in addition to existing hardware and application interfaces) to an existing abstraction, rather than just implementing the abstraction to overlay another abstraction like XWin32 or eXceed (an X server just has the same old network and hardware interfaces).
Maybe I'm just not visiting the right sites, but I've never been able to make out much that I've missed by using Netscape as my primary browser. Now granted, IE is indisputably the most-used browser out there right now (bundling with the OS and OEM/ISP agreements assured that), at least by those who are new to the Internet (still the most powerful market segment), but so far I haven't seen much that really excludes me as a Netscape user (and all of that in Intranet apps). Now I've seen a few things that render oddly in IE (and the fact that IE -still- can't handle MindTerm), but that's another matter.
Bottom line, my objection still isn't the browser (IE is impressive most of the time), it's the combination of bloated, often-buggy code (NS is equally guilty here) with nasty market practices.
Here we are a month and a half away and all I have to hope for is a Voodoo5! (sigh)
What I see, though, is that the Voodoo5 (especially the 32-processor, 2GB V5 - yipes!) may be one more step in that direction. In order to have video, 3D or otherwise, be anything more than a series of recorded frames, we still need that assembly of polygons (to allow true on-the-fly creation of the image).
The other thing I see is that the Good Enough hardware may be enough for most people, but there are still enough hardcore gamers out there who will pay $200 for the AGP Voodoo3 just to have that extra degree of speed and quality, even if it does seem negligible to most. In my house we have six boxen we use to play EverQuest. The two most comparable machines perform identically in most respects, and most people would say they don't see any difference when running EverQuest at the same resolution (other than the fact that the All-In-Wonder Pro's brightness just bites). The fact is, though, I do perceive a noticeable difference in the frame rate and texture quality of the Voodoo3, and I'm willing to pay the extra $50 or $100 for that, because it's one less thing to distract me, one more thing that allows me to become absorbed in the game (hence less hack-and-slash and more real roleplaying, which is otherwise EQ's greatest weakness).
Bottom line is, whether standard hardware levels out at Good Enough or not, vendors are still going to have enough of a market among the die-hards to keep making better and better products.
Yes. In all honesty I think Netscape is struggling at this point, but still has a lot of life left in it if they can put out a strong product (5.0) quickly, and perhaps if something develops soon in the DOJ vs. MS case that would allow Netscape to start setting up agreements to bundle browsers with new systems (that's still where most impressions are formed). It's a tough case, but it's not hopeless.
Opera 3.60 seemed to handle it fine. I can't wait for the linux release.
Off-topic, I know, but I've heard that for months, and I've yet to see a Linux release. And I'd love to see just how many big ugly bugs Opera has (there have got to be a few - I'm sick of hearing people praise Opera as though it were immune to all the headaches people see with IE/NS).</spleen>
In Microsoft, respect is gained by getting stuff out the door, not by taking the time to do it properly.
"Rant" or no, herein lies the crux of the problem, I think. Netscape knows very well that to keep any viability they must continue to be visible. Hence the reason we see new revs of Communicator with nothing much different every time we turn around. (I use 4.5/4.6 at work and 4.04/4.08/4.6/4.7 on my various boxen at home, and the only real difference I've seen is the multiple IMAP server capability in 4.5+ - nice to have for people like me who actually use IMAP, but better would be something like Outlook Express's multiple POP capability, which would actually appeal to the mass market. "Smart Browsing" is BS, and another Netcenter button does not a minor rev number make.) They're trying to stay visible, to look busy (as defined by Joe Consumer) while the Mozilla coders finish v5. And from a technical standpoint, if v5 in whatever form is half what it claims to be, I'm willing to wait.
We didn't ship because you didn't help us. If you would have helped us, we would have shipped
That was never stated nor implied. The more important statement here is that they were rebuilding the product from the ground up. Yes, it delays releases, but if the net result is a better product, so be it. Microsoft should learn from this approach, maybe start building Windows v6 now - might regain them some credibility.
On the article in general, I'm going to have to agree that it seems to suffer from a disturbing lack of unforced facts. I don't think I saw a single direct quote from a current employee, and as far as the attrition, etc., it looks like that was all from vague guesses by former employees. You want turnover figures, ask the HR department, and don't be surprised when they tell you where to go.
The approach I think I'd take to this one would have to be the swap-out program I've seen a few people mention. Now obviously, this is going to require a replaceable or rechargeable battery (in the latter case, there might also be a way to make the phone a true inexpensive replacement to an analog cell phone), but that, in effect, reduces the cost even more. It's the kind of thing I'd be willing to do, just because I don't have great credit, I don't have -frequent- use for a cell phone, but I've had times when I know it would be nice to be able to reach somebody now instead of whenever I can find a phone.
Finally, has anyone ever considered that the best favor the DoJ could do for MS is to break it up?
Yep. It actually seems to be the closest thing I've seen to an "everybody wins" solution. Figure a three-prong split, like you've potentially outlined. Each of the resulting companies is going to be focusing on its own application and how to make it less dependent on the others. This is going to mean that a lot of subsystems are going to require ground-up rebuilds. Several benefits there:
Cleaner software as a whole. I rather like the idea of a leaner, meaner IE, f'rinstance, or a much more compact Word. Either of these applications has a long list of merits, it's mostly poor coding and equally poor internal integration that make them undesirable technically, and the manufacturer's business practices that make it undesirable philosophically.
It'll create jobs, at least if well-planned. Ideally, the OS goes back to providing only the low-level services (hardware abstraction, mostly), and leaves the higher-level services to the middleware, the libraries people create on top of that. Now the OS company can still provide a set of those (your basic libc and whatnot), but other companies can also provide their own extensions or replacements to these. Someone's gotta do that.
Portability. Again, with the ground-up rewrite, there's no good reason to write it to depend on Windows. Another browser for Linux would be a good thing, as would another entry into the office apps market, the development tools market, etc., making real competition bigger and software consequently better. Same idea goes for BSD, MacOS, Be, AmigaOS, etc.
Standards. Forced to compete again, the companies would hopefully realize it's easier to create a product based on an existing standard, selling on merits like performance and usability, than to reinvent the wheel and try to sell on "features" no one else supports.
I could go on, but the point is that it seems to be a fairly good idea. Now I'm certain I've missed some of the negatives - anyone who wants to point those out, I'd enjoy seeing both sides of this argument.
But remember also that law enforcement isn't the only application of this. Imagine your boss sitting two rows over from you in your nice little cubicle hell and being able to tell exactly what porn site you're visiting today, or what manner of disparaging comments you're making about your employer on Slashdot. Even if you're using a flat-panel, you're still within 10m. I know in most tech-support it's already considered kosher to watch an agent's screen for QA purposes (though usually this is done with VNC or the like), what would keep a company from going one step further and making it unnecessary to even have the necessary software on the agent's box? (Yes, BTW, I'm aware that this would require a highly-directional antenna and perhaps even a very good idea of azimuth/elevation for each desk, but the former is available and the latter takes nothing more than first-year trig.)
Reality, though, is that the technology is there, and whether it's legal or not, it's wise to guard against it if you've got something to hide (and yes, there are plenty of legitimate reasons for having something to hide). -Drayke
I'm going to dare to go against the flow here and say no, for me it doesn't work. Mine might be an exceptional case, though,for several reasons:
I'm a firm believer in some form of telepathy, mostly out of firsthand experience. I don't hear throughts per se, but I can definitely read emotions and reactions at close range (within 10-100m). It's something I've come to depend on, and it just doesn't work over an electronic connection for obvious reasons.
People with faces are important to me. I'm still at entry-level geekdom - I work in tech support. This means I talk to 30-50 people every day for whom I never match a face. Now I grew up learning the concept that a couple must be friends before they're anything else, that they must know something about each other. My concept of "knowing someone" includes seeing their face, and I don't just mean in photos, etc.
When you get right down to it, there is still no online substitute for a hug, and I don't see that changing anytime soon.
Now, this all falls under the usual category of personal opinion and experience, so feel free to disagree, ridicule, point and laugh, whatever. Just where I stand.
Yes, we all know it's FUD. But isn't it interesting that all FUD seems to look about the same after a while? Observe:
Linux will mainly replace Windows desktop systems to run browser and X based applications, and maybe even some office applications. Therefore SCO feels no threat [etc.]
Now correct me if I'm wrong, but this sounds a lot like Microsoft's line (that Linux is not a threat to the Windows market but rather to Unix vendors - like SCO).
Linux at this moment can be considered more a play thing... [b]ecause Linux is basically a free-for-all it means that no individual... is accountable should anything go wrong[.]
This again sounds like the Microsoft/Jesse Berst line of "there's no one to sue if it goes wrong." Well, frankly, to me that's fine because most of the time the problems I've seen with an OS on the server level can be attributed to operator error or "act of God" type situations anyway.
d) It is often perceived to be dangerous when a company uses an operating system where the source code is available on the web to the whole world[.]
Again, it seems to me I've seen MS use this argument before, and it's just as fallacious. There's an easy solution: it's very easy to provide "look-but-don't-touch" or less access to the kernel sources or anything else that is considered potentially hazardous. Or if kernel hacking is necessary, do the work on another box and don't introduce it to the mission-critical systems until it's been reviewed and tested.
The last bit, even if it doesn't fall into the the category of repetitive FUD, seems to destroy the whole argument:
For 3 years now, SCO has supplied free of charge UnixWare and OpenServer for educational and non commercial use. SCO does not supply source code with its products. (emphasis mine)
Well, it's nice to get the OS free of charge, but the license is still restrictive, and you're still stuck with the kernel SCO gives you. The flexibility of the code is the #1 reason I've found for using Linux. -Drayke
This one had me rolling, really. Customers are asking for bloat? I've yet to meet one. Frankly, if it's really expected of us all to have a PIII/500 and 256MB of RAM just to run a word processor, a spreadsheet, a half-baked database and a broken contact manager/mail client, then I for one am content to fall behind - I do as well just installing something like StarOffice that does 95% of the same in useful functions and has a quarter of the disk footprint.
More frightening yet, though, is that best as I can tell, the author believes everything he's said. He doesn't realize that he's going to have to put on an asbestos suit to open his mail for the next several days (I for one won't be a part of this - he's below my response threshold, too brainwashed to bother). -Drayke
Trouble is that when the driver is buggy, version-specific (herein lies its biggest problem and the biggest issue with being binary-only), -and- they're holding their cards close to their chest, well, they're not doing themselves any favours either 'cos it means they're doing Linux development half in the dark (99% of their development team is still Windows-oriented and will remain that way). By making the thing binary-only, they're screwing themselves out of the best development team they could have, namely the people who, whether for money or whatever, deal with and develop for Linux on a daily basis, and therefore know the APIs they're developing for and so on.
Personally, I'm inclined to stay with the consensus on this one, that it's not enough. Granted, Creative's worried about their proprietary information and so on, but why have they been dominant up to this point? Not because they alone implement the standard, but because they provide the best, most compatible implementation. Opening it up gets the developer base, gives Linux users a good reason to buy an SBLive, and while it opens up the interface to competitors, it also means that people are going to take the time to make that driver something worth using.
Slashdot Mirror
Seems to me an easy (well, maybe not always practical, but I can't think of much alternative) answer to this would be, in cases where this kind of thing is expected, to have the appropriate sort at Andover contact the site owner ahead of time with some kind of release for short-term mirroring. (Honestly, if you look at it, most of this type of content will keep for a few days before posting anyway, and it saves all manner of sysadmins the headache of scrambling to keep up with the Slashdot effect.)
-Drayke
The complaint I keep seeing over and over in this is that people can't recover their data. To this, the answer seems simple enough: for static content and scripts, always keep a local copy. For databases, etc., don't rely on the provider's advertised backups - mirror the data, preferably at least nightly, even if it means fighting your provider to get them to allow you access to those files. This is a job that can be easily performed by a script (though admittedly, depending on your data, it could get bandwidth-intensive).
As for other services (mail, DNS, etc.), my best advice is what's been offered before: don't put all your eggs in one basket. Sure, it costs more to have two or three different services for mail, DNS and web hosting, but it allows you a smoother transition should one provider suddenly prove unreliable.
A final word on this: in some cases in just makes better sense to host your own servers. I don't really think this option makes sense for high-volume web operations in small companies (I personally don't have the pocket money to shell out for an OC3), but in cases where you've got a pipe and an IP block anyway, it might just make good sense to set yourself up a mail server and a DNS, even if they're just fallbacks from what your hosting service provides. That way, when whatever MCSE they've hired as admin comes along and tries this new fdisk command he just learned on the primary DNS (apologies to all the MCSEs out there who actually know what they're doing), your operation still has a decent chance of surviving, even if it does take a speed hit for a while.
-Drayke
I moderated him down because I hate Tom. Do I need any other reason?
Well, I was hoping for something more like "I moderated him down because his argument seemed pointless." Would've at least been more like what moderation's supposed to be for - scoring the usefulness of the comment instead of just whether you like the opinion and/or the poster.
In all truth, though, I don't know that there's much to be done to fix the moderation system. Like it or not, it does seem to improve S/N in a mostly fair and reasonable manner. And granted, there are those who'll abuse it (show me a way to create an abuse-proof situation with any sentient interaction and I'll show you something likely not worth using).
Finally, in Tom's defense, I've found his to be some of the best-constructed, most sensible arguments on Slashdot, even when I disagree with his views. And that is something that Slashdot is all too sorely lacking.
</grouse>
-Drayke
A lot of the time, KDE and GNOME seem to follow somewhat-different courses to very similar ends. I understand there are probably some philosophical differences in the code used and so on, but from a functional point of view, would the projects benefit from something a little more cooperative, perhaps creating a modular environment where users can pick and choose their functionality, or where the underlying interface could be tailored to accomodate whatever methods a particular developer preferred?
-Drayke
Honestly, I've got to look at this and say that the class-action suits on the consumer level aren't necessarily going to be very much of a concern for MS. As has been pointed out, the vast majority of Windows installations are bundled packages, not people buying individual CDs. This makes it difficult to come by a preponderance of evidence showing harm to consumers. At the corporate level, I think there's a more real threat, but I don't really personally see much potential in the consumer suits.
-Drayke
A little more difficult, though. X-Win32 (also MI/X, eXceed, and probably a few I've missed) is simply an X server, designed to implement X11 (already closely tied to IP networking) over the Windows GDI. The more difficult proposition is intercepting GDI calls (designed to run with a local display) and actually convert those to X calls. It requires implementation at a deeper level, essentially adding a third interface (network in addition to existing hardware and application interfaces) to an existing abstraction, rather than just implementing the abstraction to overlay another abstraction like XWin32 or eXceed (an X server just has the same old network and hardware interfaces).
-Drayke
Maybe I'm just not visiting the right sites, but I've never been able to make out much that I've missed by using Netscape as my primary browser. Now granted, IE is indisputably the most-used browser out there right now (bundling with the OS and OEM/ISP agreements assured that), at least by those who are new to the Internet (still the most powerful market segment), but so far I haven't seen much that really excludes me as a Netscape user (and all of that in Intranet apps). Now I've seen a few things that render oddly in IE (and the fact that IE -still- can't handle MindTerm), but that's another matter.
Bottom line, my objection still isn't the browser (IE is impressive most of the time), it's the combination of bloated, often-buggy code (NS is equally guilty here) with nasty market practices.
-Drayke
Here we are a month and a half away and all I have to hope for is a Voodoo5! (sigh)
What I see, though, is that the Voodoo5 (especially the 32-processor, 2GB V5 - yipes!) may be one more step in that direction. In order to have video, 3D or otherwise, be anything more than a series of recorded frames, we still need that assembly of polygons (to allow true on-the-fly creation of the image).
The other thing I see is that the Good Enough hardware may be enough for most people, but there are still enough hardcore gamers out there who will pay $200 for the AGP Voodoo3 just to have that extra degree of speed and quality, even if it does seem negligible to most. In my house we have six boxen we use to play EverQuest. The two most comparable machines perform identically in most respects, and most people would say they don't see any difference when running EverQuest at the same resolution (other than the fact that the All-In-Wonder Pro's brightness just bites). The fact is, though, I do perceive a noticeable difference in the frame rate and texture quality of the Voodoo3, and I'm willing to pay the extra $50 or $100 for that, because it's one less thing to distract me, one more thing that allows me to become absorbed in the game (hence less hack-and-slash and more real roleplaying, which is otherwise EQ's greatest weakness).
Bottom line is, whether standard hardware levels out at Good Enough or not, vendors are still going to have enough of a market among the die-hards to keep making better and better products.
-Drayke
Do you honestly think ZDNet was off base?
Yes. In all honesty I think Netscape is struggling at this point, but still has a lot of life left in it if they can put out a strong product (5.0) quickly, and perhaps if something develops soon in the DOJ vs. MS case that would allow Netscape to start setting up agreements to bundle browsers with new systems (that's still where most impressions are formed). It's a tough case, but it's not hopeless.
-Drayke
Opera 3.60 seemed to handle it fine. I can't wait for the linux release.
Off-topic, I know, but I've heard that for months, and I've yet to see a Linux release. And I'd love to see just how many big ugly bugs Opera has (there have got to be a few - I'm sick of hearing people praise Opera as though it were immune to all the headaches people see with IE/NS).</spleen>
-Drayke
In Microsoft, respect is gained by getting stuff out the door, not by taking the time to do it properly.
"Rant" or no, herein lies the crux of the problem, I think. Netscape knows very well that to keep any viability they must continue to be visible. Hence the reason we see new revs of Communicator with nothing much different every time we turn around. (I use 4.5/4.6 at work and 4.04/4.08/4.6/4.7 on my various boxen at home, and the only real difference I've seen is the multiple IMAP server capability in 4.5+ - nice to have for people like me who actually use IMAP, but better would be something like Outlook Express's multiple POP capability, which would actually appeal to the mass market. "Smart Browsing" is BS, and another Netcenter button does not a minor rev number make.) They're trying to stay visible, to look busy (as defined by Joe Consumer) while the Mozilla coders finish v5. And from a technical standpoint, if v5 in whatever form is half what it claims to be, I'm willing to wait.
-Drayke
We didn't ship because you didn't help us. If you would have helped us, we would have shipped
That was never stated nor implied. The more important statement here is that they were rebuilding the product from the ground up. Yes, it delays releases, but if the net result is a better product, so be it. Microsoft should learn from this approach, maybe start building Windows v6 now - might regain them some credibility.
On the article in general, I'm going to have to agree that it seems to suffer from a disturbing lack of unforced facts. I don't think I saw a single direct quote from a current employee, and as far as the attrition, etc., it looks like that was all from vague guesses by former employees. You want turnover figures, ask the HR department, and don't be surprised when they tell you where to go.
-Drayke
The approach I think I'd take to this one would have to be the swap-out program I've seen a few people mention. Now obviously, this is going to require a replaceable or rechargeable battery (in the latter case, there might also be a way to make the phone a true inexpensive replacement to an analog cell phone), but that, in effect, reduces the cost even more. It's the kind of thing I'd be willing to do, just because I don't have great credit, I don't have -frequent- use for a cell phone, but I've had times when I know it would be nice to be able to reach somebody now instead of whenever I can find a phone.
-Drayke
Yep. It actually seems to be the closest thing I've seen to an "everybody wins" solution. Figure a three-prong split, like you've potentially outlined. Each of the resulting companies is going to be focusing on its own application and how to make it less dependent on the others. This is going to mean that a lot of subsystems are going to require ground-up rebuilds. Several benefits there:
I could go on, but the point is that it seems to be a fairly good idea. Now I'm certain I've missed some of the negatives - anyone who wants to point those out, I'd enjoy seeing both sides of this argument.
-Drayke
But remember also that law enforcement isn't the only application of this. Imagine your boss sitting two rows over from you in your nice little cubicle hell and being able to tell exactly what porn site you're visiting today, or what manner of disparaging comments you're making about your employer on Slashdot. Even if you're using a flat-panel, you're still within 10m. I know in most tech-support it's already considered kosher to watch an agent's screen for QA purposes (though usually this is done with VNC or the like), what would keep a company from going one step further and making it unnecessary to even have the necessary software on the agent's box? (Yes, BTW, I'm aware that this would require a highly-directional antenna and perhaps even a very good idea of azimuth/elevation for each desk, but the former is available and the latter takes nothing more than first-year trig.)
Reality, though, is that the technology is there, and whether it's legal or not, it's wise to guard against it if you've got something to hide (and yes, there are plenty of legitimate reasons for having something to hide). -Drayke
-Drayke
- I'm a firm believer in some form of telepathy, mostly out of firsthand experience. I don't hear throughts per se, but I can definitely read emotions and reactions at close range (within 10-100m). It's something I've come to depend on, and it just doesn't work over an electronic connection for obvious reasons.
- People with faces are important to me. I'm still at entry-level geekdom - I work in tech support. This means I talk to 30-50 people every day for whom I never match a face. Now I grew up learning the concept that a couple must be friends before they're anything else, that they must know something about each other. My concept of "knowing someone" includes seeing their face, and I don't just mean in photos, etc.
- When you get right down to it, there is still no online substitute for a hug, and I don't see that changing anytime soon.
Now, this all falls under the usual category of personal opinion and experience, so feel free to disagree, ridicule, point and laugh, whatever. Just where I stand.-Drayke
-Drayke
Yes, we all know it's FUD. But isn't it interesting that all FUD seems to look about the same after a while? Observe:
... [b]ecause Linux is basically a free-for-all it means that no individual ... is accountable should anything go wrong[.]
Linux will mainly replace Windows desktop systems to run browser and X based applications, and maybe even some office applications. Therefore SCO feels no threat [etc.]
Now correct me if I'm wrong, but this sounds a lot like Microsoft's line (that Linux is not a threat to the Windows market but rather to Unix vendors - like SCO).
Linux at this moment can be considered more a play thing
This again sounds like the Microsoft/Jesse Berst line of "there's no one to sue if it goes wrong." Well, frankly, to me that's fine because most of the time the problems I've seen with an OS on the server level can be attributed to operator error or "act of God" type situations anyway.
d) It is often perceived to be dangerous when a company uses an operating system where the source code is available on the web to the whole world[.]
Again, it seems to me I've seen MS use this argument before, and it's just as fallacious. There's an easy solution: it's very easy to provide "look-but-don't-touch" or less access to the kernel sources or anything else that is considered potentially hazardous. Or if kernel hacking is necessary, do the work on another box and don't introduce it to the mission-critical systems until it's been reviewed and tested.
The last bit, even if it doesn't fall into the the category of repetitive FUD, seems to destroy the whole argument:
For 3 years now, SCO has supplied free of charge UnixWare and OpenServer for educational and non commercial use. SCO does not supply source code with its products. (emphasis mine)
Well, it's nice to get the OS free of charge, but the license is still restrictive, and you're still stuck with the kernel SCO gives you. The flexibility of the code is the #1 reason I've found for using Linux. -Drayke
This one had me rolling, really. Customers are asking for bloat? I've yet to meet one. Frankly, if it's really expected of us all to have a PIII/500 and 256MB of RAM just to run a word processor, a spreadsheet, a half-baked database and a broken contact manager/mail client, then I for one am content to fall behind - I do as well just installing something like StarOffice that does 95% of the same in useful functions and has a quarter of the disk footprint.
More frightening yet, though, is that best as I can tell, the author believes everything he's said. He doesn't realize that he's going to have to put on an asbestos suit to open his mail for the next several days (I for one won't be a part of this - he's below my response threshold, too brainwashed to bother). -Drayke
Trouble is that when the driver is buggy, version-specific (herein lies its biggest problem and the biggest issue with being binary-only), -and- they're holding their cards close to their chest, well, they're not doing themselves any favours either 'cos it means they're doing Linux development half in the dark (99% of their development team is still Windows-oriented and will remain that way). By making the thing binary-only, they're screwing themselves out of the best development team they could have, namely the people who, whether for money or whatever, deal with and develop for Linux on a daily basis, and therefore know the APIs they're developing for and so on.
Personally, I'm inclined to stay with the consensus on this one, that it's not enough. Granted, Creative's worried about their proprietary information and so on, but why have they been dominant up to this point? Not because they alone implement the standard, but because they provide the best, most compatible implementation. Opening it up gets the developer base, gives Linux users a good reason to buy an SBLive, and while it opens up the interface to competitors, it also means that people are going to take the time to make that driver something worth using.