I thought SuSE used to be download available, but only piecewise. In other words, you could do a live internet install or download packages individually, but from SuSE you would not have gotten a simple set of ISOs.
Anyway, now there is SLES and OpenSuSE, with the same relationship as RHEL and Fedora.
But you are right, both RH and SuSE can be downloaded for free with registration, with cutoff on updates. I wasn't even aware of this. It does ease some of the problems, but the simpler Ubuntu approach still seems more straightforward. A large chunk of commercial software pain is tracking entitlement, and coping with limited-period 'evaluations' don't exactly make that easier.
Well, SuSE's distribution was always aggravating to those who wanted it for free (no free ISO downloads back in the day meant it was hard to install for free). RH was amenable to first-party free distribution until RH9, after which they decided this was their way.
Namely, both SuSE and RHEL have a 'commercial-only' distribution with those enterprise sensibilities and a free 'first-party' offering that is ostensibly an enthusiast endeavor which really translates to recruiting enthusiasts as testers. They bank on trademark/copyright of text and images to keep clones from looking *too* much like their first-party offerings. CentOS is from a technical standpoint, a clone (plus some other stuff, but the clone-only behavior is default), but distinguishable enough to preclude Vendor and ISV support (both don't want to go the linux support path alone generally).
Meanwhile, here comes Canonical. They truly keep the distribution and support model independent. They have rapid release cycles, but denote a more 'enterprise-friendly' LTS cycle underscoring things. Regardless, the distribution is free to download and distribute. So clients can prototype and train and even do production as they feel comfortable with doing so without support, and then when they do need support, the contract is available without reinstall or other drastic measures. Suddenly, the mark of whether another party will support it or not is not keyed on the distribution, instead requiring a Canonical support contract to be in place.
I think SuSE/RH's approach is botching the market. I know of a *lot* of CentOS installs going in to places that might feel more comfortable with the option of purchasing a support contract. Knowing the strict distinction between RH and CentOS, Ubuntu will be very appealing to those places. The absolute identical nature of free training/development/prototyping systems with low support requirements and production use is also appealing.
If you know your material well and can find prior art, politely point that out for the short term. If you can't, you simply must accept in this instance, you likely explicitly signed over rights to this.
Long term, if you are so fundamentally against a core procedure like this, and will likely antagonize and be at odds with your boss over it, you should look elsewhere for a situation that you agree with. It sounds to an extent callous to say "it's the company's", but non-decision-making people have this sort of action as the counter-leverage. If the top talent tends to hate software patents, patent loving companies would have hard time attracting what they need.
Android is in trouble wrt to fulfilling the hope of an 'open' platform. So far:
-We still know next to nothing about the current state of their development situation. The M5 SDK released all the way back in March is the latest hard technical resource people have without an NDA. Meanwhile, Google is refreshing the SDK for the cherry-picked few (http://groups.google.com/group/android-developers/browse_thread/thread/f031c33fe9e5b992). For a platform trying to leverage a large development community, they sure are making it hard for those people. -The bits we do have come without platform source. There seems to be a good chance Google might keep their middleware closed-source. Otherwise, why be so secretive about it even today?
So far we've seen promise of being open falling through to date, we've seen the supposed source of strength of android (the community), hamstringed by Google's own actions. I've seen promises of 'once the phone is on the market, we'll make good!', but I fail to see why they can't allow the SDK to be in public hands because of that excuse.
Then we have LiMo, which so far looks not to be user-centered, and more cell-phone manufacturer centered, so I'd not expect that to change the world significantly.
The *only* platform that so far in spirit lives up to those promises is OpenMoko. Unfortunately, the 'best' platform for it (FreeRunner) is a tad underpowered technically.
I blame to an extent the love affair with abstraction layers. Applied correctly, the concept is good, but try to frontend too much and you end up with a layer that doesn't make things easier and adds complexity. Too many times I see an abstraction layer constructed that only frontends one technology, in the name of presenting only the 'necessary' bits. Over time, they realize why the underlying technology had those bits and why they need them. At the end, they end up with a layer that has to represent 99% of the technology underneath and didn't let them move from one underlying technology to another.
Generalize that to dozens of layers in some cases. That is a common cause that I see where software projects increase their complexity to unreasonable degrees without payoff.
I worked on one project with many many lines of code, many abstraction layers, so that ultimately the underlying interface would be obfuscated into XML-RPC calls. Meanwhile, I also worked on something at the same time that didn't bother trying to follow that and just coded. This was the 'legacy' application that would be sunset. After a couple of years, the legacy was still going strong and in fact was able to fix problems that were a challenge for the 'new' project. Functionality of the new never caught up, and couldn't be as agile. The new project had a headcount of 40 and a large budget. The old project had 2 people.
The new project has been canceled, and the 'legacy' strategy continued, as ultimately, KISS wins. Complexity for the sake of 'simplicity' is a dumb idea.
For people to use it to develop an understanding that makes you comfortable about recommending purchase of QS blade servers, which have none of those limitations of the PS3.
Howerver, I still want a good PS3 Myth frontend. I already have one that is on an AMD PC, but I would love to have one fewer device in the mix.
If they didn't know how to develop for it, they wouldn't have gotten to #1 supercomputing slot. IBM knows how.
Now, with this, it's a move to drum up interest in the architecture. IBM has server-class systems with enterprise support for CBE systems (QS22 being the current generation). However, drumming up general community interest in Cell benefits them in many ways.
Businesses contemplating Cell can have developers evaluate the architecture on the cheap. They won't be able to have a high speed interconnect or decent RAM, or even respectable double-precision performance. They'll have just enough to get a feel for how coding for the platform would be. It's IBM's dream scenario, a crippled product that people buy that can serve as an evaluation platform for customers.
IBM has come out with an amazing performance chip and managed to get it into a readily available package people can play with. Whether IBM knew it or not, the PS3 capability to run Linux is a boon to their QS server stragety.
A two processor Xeon 3.0 ghz quad core system would have an xhpl-relevant Rpeak of about 96 Gflops. Two cell processors (of the Poer8XCelli varaint, not the ones in PS3) has an Rpeak of 200 Gigaflops. The power consumption of those are about even. So the processor achieves over twice the aggregate performance within a comparable power envelope. *THAT* is why it's interesting.
See the #1 Top500 system. Not significantly more power usage than Intel systems, but blows the Intel ones out of the water.
The thing about code is, a Cell system can run the same C code other platforms can. Not blazingly fast, mind you, but it isn't like you *must* retool everything. In order to get the 2x boosts in certain code loops, yes, you would have to, but a lot of that is done already.
Though typically pointless to reply, I'll clarify what I really thought.
I'm not a fan of Apple, I'd just rather run Linux. I think the would-be customers of Apple clones explicitly disapproved of by Apple should forget OSX and run something else. Buying an Apple is a valid choice for those who accept Apple's model, but if you really dislike the whole strategy of Apple's platform design, don't send them your money to use in proliferating such a platform. OSX loses a lot of the 'it just works' charms when exposed to a more open ecosystem anyway.
I would like to see development attention on fighting OSX to instead think of improving a GNUstep based platform to compete.
If you really want the OS, then it just makes sense to acquire it in a fashion Apple wouldn't mind.
Do I believe in draconian EULA limitations? No. Is it perfectly legitimate to install OSX on a system not blessed by apple? I think it should be.
However, practically speaking, it's a bad path. Just as you have the right to use the software as you see fit, Apple has the right (even if a bad idea) to withhold updates/support from such people. If an OSX update fixes important security issues *and* kills the loophole you use to run the OS, Apple is not required to split it out. Best case, you always wait for a community that is potentially untrustworthy to bless patches and rerelease. However, Apple can sue those people for illegal redistribution.
If you don't want to play by their rules, why give them money to explicitly fight with them over how to use the software?
I'm not personally particularly interested in this instance, but I do think this long after a creation, it's really not helping the general advancement of society to have a company milking it long after the creator is dead. Everyone reasonably responsible for the idea is gone and their descendants have earned probably as much as could be reasonable without revamping it with new and novel twists. Scrabble becoming a generic would save us from a confusing namespace of tons of different names for the same thing. Imagine if the names chess and checkers were still protected.
Now, I'll preface this with a disclaimer that I avoid Fedora generally. I got reminded of why during a recent attempt to use it and follow it, it really punishes the users with inconsistant updates even after release.
That said, RPM dependencies are no more convoluted than deb dependencies. The difference is that originally, RH distros had only the rpm command and debian out of the gate recognized the need for both dpkg *and* apt. RPM distributions each have at least one repository management strategy now (YaST, yum, etc etc). So dependency hell is not one of their worst problems (though I do prefer some apt defaults more than yum, I won't say yum isn't up to the task).
But judging soley from the price of a single share is incomplete data. If I had a company at $10,000 dollars a share, but only 10 shares, it'd be a very low valued company (whether deservedly so or not). Many companies that are popularly tracked would have split on an ascent like Google's, but they chose not to. A better, but imperfect measure would be market cap.
That said, Google's market cap is 150 billion. CBS is at 12 billion or so. One analogy I've been fond of thinking of is comparing television advertising to internet advertising. It's imperfect, but still. The business models are different, as CBS has a higher demanding expense I would guess (TV show production/content creation/distribution I would guess more expensive than even Google's indexing of content from other people), but still over a magnitude more valuable does strike me as steep...
Ugh. MS's commercial success had little to do with technical merit and everything to do with sane and strategically superior business positioning.
People can bash MS all they want, but still, the bottom line is, if their product didn't work (and in the early days, better than anything else), it wouldn't be as popular.
In the early days, the key is at what price a total platform could be acquired for, and that it worked at least most of the time, even if not the best.
"you can get whatever flavor desktop you want, don't like gnome, get kde"..... Yeah, I want to support THAT mess with > 1000 desktops under my umbrella
Just because the world has choice doesn't mean you have to relay that to your organization. Do you pick a single source hardware vendor? How can you possibly do that when people in the world can buy something different than what you choose? The answer is you don't give a rat's ass about the general choice, because your organization picked one. People gripe and override your choice, well, tough for them if they call for help. MS itself has choices inherent. If you chose IE, and users went with Firefox, even in windows, how would this be different?
Macintosh had a GUI second (GEOS on the COMMODeORE was first), and it cost so much to make the hardware and software, nobody bought that crap (LISA). Then Mac himself came out, and it looked like shit. Windows solved both, in that you could have a clone, or a real IBM, and BOTH machines acted the same way.
To say Win3.11 looked better than the same era MacOS is laughable. The second half of your point does nail it, that IBM clones were the whole driving force of MS success. This is not a technical thing, this was a business maneuver, in part banking on a decision by IBM to create a market a certain way as to not be the only ones who could participate.
Let's talk about a clone and an IBM, each one running a different flavor of linux, and different desktop GUI's, and then lets talk about support.
Ok... And as the person who choses to support a standard desktop from RHEL, you care about how a KDE Novell platform would be supported... Precisely why? Why would you care about the existence of solutions you don't chose any more than you care about them now as a Windows shop cares about Linux based solutions already?
Now it's about entrenchment and leveraging the monopoly every way they can. And they get away with it too because in part the government swings that way today and also in part because people are afraid that without a monopoly, things would change and they aren't comfortable predicting how.
1. Vision. This is precisely what a single distribution generally has. They pick a set of coherent packages that fit a vision. The problem a *lot* of critics of Linux is that any and all use of the platform should be consistent because choice is too hard. It's like complaining that motor vehicles are inconsistent in a free market. Audi has some car with DSG, and yet a competitor offers a stickshift, it's sheer madness for the users! A Ford is drastically different from a Harley, how in the world do they expect people to chose between them and drive the same roads? The MS situation has skewed the software market in an unfortunate way relative to other sane markets.
2. Marketing. I think your point was a tad confused. In the consumer space, I do think a coherent commercial marketing thing isn't there, regardless of the technical state of things. In the 'trouble ticket' scenario, that's more than covered by Novell and RedHat. If you are not going to pay a person to have the responsibility to muck around with the internals anyway, you don't need to care that they are there. It's like saying a bus with the engine compartment welded shut is popular among drivers because then the drivers feel they'll never get blamed. Just because they are paid to operate them, doesn't mean they are expected to be a mechanic just because they can reach the engine.
3. Again, what? In corporations with lots of money, they do follow what their highly paid geeks think. It's the whole job. I don't see exactly what you are getting at when it comes to blaming the choice if dissatisfied. They can blame the vendor either way, or the architect who made the wrong choice for their needs.
And what is the BS that Red Hat doesn't guarantee anything? They guarantee their commercial product. They don't guarantee 'everything', meaning third party support, but MS doesn't either. MS will not be jumping to your aid because AutoCAD crashes, so I'm not seeing what your point is.
Linux has several sets of leaders/vision/marketing. I know, this sounds like an oxymoron, but it is no more so than characterizing other fields as having multiple sets of these. There are multiple sets of people doing this in 'console gaming' that don't even try to get along. There are multiple sets in cars, in computing, in retailing, any industry. The problem is people think of Linux as a specific thing, when they should recognize it as a framework/environment that produces coherent visions. You can't claim RHEL is not the realization of a consistent vision, and that Ubuntu is not one either. Despite a lot of similarities, it is plain that they are not totally consistent to each other, but then again neither are Sony and Nintendo.
This truly is a huge reason why consumers should be jumping at linux. Being beholden to a single vendor throughout your lifecycle is not a good thing. The unique thing 'Linux' brings to the table is a relatively low barrier to jumping vendors. No, jumping from RHEL to SLES isn't trivial, but it's much more plausible than Windows to OSX or vice-versa. Choice in all parts of the stack has revolutionized the PC industry, the server industry, the computer component industry. Don't like Intel, get AMD. Don't like Lenovo, get an ASUS. Don't like your current Dell solution, get IBM next time. In all cases you can do that because the products are equivalently usable. Bringing that to the OS is a logical good thing for the consumer.
Games do that, *particularly* online ones, and suddenly you have rather drastic balance issues. A lot of FPS is dumbed pretty down (auto-aim like crazy) for the consoles, due to the inaccuracy of the control pads, and it would just be ludicrously easy if someone had a correctly behaved mouse doing the aiming. If they castrated the mouselook to be fair, it would be pointless too.
I don't have the console as I haven't seen a single exclusive to bother. I'm saying actually that a lot of the new games I dislike on consoles, though I wouldn't mind them on a desktop.
Hell, even Final Fantasy 13 isn't going to be exclusive, and their main line has been exclusive to the playstation since the ps1.
Not entirely, they made a cash grab with 7 and 8 PC variants. Since 7 they've been so eye-candy demanding they required the storage of Sony's offerings. n64 cartridges precluded them, Gamecube was late and with smaller disks, xbox could have probably been sufficient for 10 and 12, but they were so late to the party. 13 was almost certainly going to be exclusive due in part to BD, but the MS bag of money probably just got too big to resist. Call it a conspiracy if you will, but Square/Enix all of a sudden got a bee in their bonnet about 360 releases. It might have been the mark of market decision of an exec, if not for the changing of Star Ocean to be 360 exclusive and the PS3 last remnant to be delayed to give the 360 variant top billing. To make that sort of decision takes some money bags changing hands.
I will say my chief gripe is that FPS developers have been dominating the PS3 scene too. The focus seems to be on things like Haze, Resistance, Orange Box, all this stuff. The 360 is chock full of them too. I don't mind a good FPS, but keep it an a desktop system, FPS and RTS I can't see playing any other way except with my good old mouse and keyboard.
Seriously, I have a Wii and a PS3 and I've found more interesting games for Wii than the PS3. However, some decent PS2 games have still been coming out.
Slashdot Mirror
I thought SuSE used to be download available, but only piecewise. In other words, you could do a live internet install or download packages individually, but from SuSE you would not have gotten a simple set of ISOs.
Anyway, now there is SLES and OpenSuSE, with the same relationship as RHEL and Fedora.
But you are right, both RH and SuSE can be downloaded for free with registration, with cutoff on updates. I wasn't even aware of this. It does ease some of the problems, but the simpler Ubuntu approach still seems more straightforward. A large chunk of commercial software pain is tracking entitlement, and coping with limited-period 'evaluations' don't exactly make that easier.
Well, SuSE's distribution was always aggravating to those who wanted it for free (no free ISO downloads back in the day meant it was hard to install for free). RH was amenable to first-party free distribution until RH9, after which they decided this was their way.
Namely, both SuSE and RHEL have a 'commercial-only' distribution with those enterprise sensibilities and a free 'first-party' offering that is ostensibly an enthusiast endeavor which really translates to recruiting enthusiasts as testers. They bank on trademark/copyright of text and images to keep clones from looking *too* much like their first-party offerings. CentOS is from a technical standpoint, a clone (plus some other stuff, but the clone-only behavior is default), but distinguishable enough to preclude Vendor and ISV support (both don't want to go the linux support path alone generally).
Meanwhile, here comes Canonical. They truly keep the distribution and support model independent. They have rapid release cycles, but denote a more 'enterprise-friendly' LTS cycle underscoring things. Regardless, the distribution is free to download and distribute. So clients can prototype and train and even do production as they feel comfortable with doing so without support, and then when they do need support, the contract is available without reinstall or other drastic measures. Suddenly, the mark of whether another party will support it or not is not keyed on the distribution, instead requiring a Canonical support contract to be in place.
I think SuSE/RH's approach is botching the market. I know of a *lot* of CentOS installs going in to places that might feel more comfortable with the option of purchasing a support contract. Knowing the strict distinction between RH and CentOS, Ubuntu will be very appealing to those places. The absolute identical nature of free training/development/prototyping systems with low support requirements and production use is also appealing.
Will be supported at launch? They mention handsets, but what of OBEX?
May shorten your life expectancy.
If you know your material well and can find prior art, politely point that out for the short term. If you can't, you simply must accept in this instance, you likely explicitly signed over rights to this.
Long term, if you are so fundamentally against a core procedure like this, and will likely antagonize and be at odds with your boss over it, you should look elsewhere for a situation that you agree with. It sounds to an extent callous to say "it's the company's", but non-decision-making people have this sort of action as the counter-leverage. If the top talent tends to hate software patents, patent loving companies would have hard time attracting what they need.
If your video driver crashes, then the lack of a blue screen wouldn't particularly be a significant improvement in this scenario either.
I.e. linux is mostly userspace for most video driver stuff, but when X crashes, the fact it can be sshed into would be of little comfort here either.
Android is in trouble wrt to fulfilling the hope of an 'open' platform. So far:
-We still know next to nothing about the current state of their development situation. The M5 SDK released all the way back in March is the latest hard technical resource people have without an NDA. Meanwhile, Google is refreshing the SDK for the cherry-picked few (http://groups.google.com/group/android-developers/browse_thread/thread/f031c33fe9e5b992).
For a platform trying to leverage a large development community, they sure are making it hard for those people.
-The bits we do have come without platform source. There seems to be a good chance Google might keep their middleware closed-source. Otherwise, why be so secretive about it even today?
So far we've seen promise of being open falling through to date, we've seen the supposed source of strength of android (the community), hamstringed by Google's own actions. I've seen promises of 'once the phone is on the market, we'll make good!', but I fail to see why they can't allow the SDK to be in public hands because of that excuse.
Then we have LiMo, which so far looks not to be user-centered, and more cell-phone manufacturer centered, so I'd not expect that to change the world significantly.
The *only* platform that so far in spirit lives up to those promises is OpenMoko. Unfortunately, the 'best' platform for it (FreeRunner) is a tad underpowered technically.
I blame to an extent the love affair with abstraction layers. Applied correctly, the concept is good, but try to frontend too much and you end up with a layer that doesn't make things easier and adds complexity. Too many times I see an abstraction layer constructed that only frontends one technology, in the name of presenting only the 'necessary' bits. Over time, they realize why the underlying technology had those bits and why they need them. At the end, they end up with a layer that has to represent 99% of the technology underneath and didn't let them move from one underlying technology to another.
Generalize that to dozens of layers in some cases. That is a common cause that I see where software projects increase their complexity to unreasonable degrees without payoff.
I worked on one project with many many lines of code, many abstraction layers, so that ultimately the underlying interface would be obfuscated into XML-RPC calls. Meanwhile, I also worked on something at the same time that didn't bother trying to follow that and just coded. This was the 'legacy' application that would be sunset. After a couple of years, the legacy was still going strong and in fact was able to fix problems that were a challenge for the 'new' project. Functionality of the new never caught up, and couldn't be as agile. The new project had a headcount of 40 and a large budget. The old project had 2 people.
The new project has been canceled, and the 'legacy' strategy continued, as ultimately, KISS wins. Complexity for the sake of 'simplicity' is a dumb idea.
I said two Power8XCell processers can do 200 gflops double precision, therefore one can do about 100.
For people to use it to develop an understanding that makes you comfortable about recommending purchase of QS blade servers, which have none of those limitations of the PS3.
Howerver, I still want a good PS3 Myth frontend. I already have one that is on an AMD PC, but I would love to have one fewer device in the mix.
If they didn't know how to develop for it, they wouldn't have gotten to #1 supercomputing slot. IBM knows how.
Now, with this, it's a move to drum up interest in the architecture. IBM has server-class systems with enterprise support for CBE systems (QS22 being the current generation). However, drumming up general community interest in Cell benefits them in many ways.
Businesses contemplating Cell can have developers evaluate the architecture on the cheap. They won't be able to have a high speed interconnect or decent RAM, or even respectable double-precision performance. They'll have just enough to get a feel for how coding for the platform would be. It's IBM's dream scenario, a crippled product that people buy that can serve as an evaluation platform for customers.
IBM has come out with an amazing performance chip and managed to get it into a readily available package people can play with. Whether IBM knew it or not, the PS3 capability to run Linux is a boon to their QS server stragety.
A two processor Xeon 3.0 ghz quad core system would have an xhpl-relevant Rpeak of about 96 Gflops. Two cell processors (of the Poer8XCelli varaint, not the ones in PS3) has an Rpeak of 200 Gigaflops. The power consumption of those are about even. So the processor achieves over twice the aggregate performance within a comparable power envelope. *THAT* is why it's interesting.
See the #1 Top500 system. Not significantly more power usage than Intel systems, but blows the Intel ones out of the water.
The thing about code is, a Cell system can run the same C code other platforms can. Not blazingly fast, mind you, but it isn't like you *must* retool everything. In order to get the 2x boosts in certain code loops, yes, you would have to, but a lot of that is done already.
Hah!
Though typically pointless to reply, I'll clarify what I really thought.
I'm not a fan of Apple, I'd just rather run Linux. I think the would-be customers of Apple clones explicitly disapproved of by Apple should forget OSX and run something else. Buying an Apple is a valid choice for those who accept Apple's model, but if you really dislike the whole strategy of Apple's platform design, don't send them your money to use in proliferating such a platform. OSX loses a lot of the 'it just works' charms when exposed to a more open ecosystem anyway.
I would like to see development attention on fighting OSX to instead think of improving a GNUstep based platform to compete.
If you really want the OS, then it just makes sense to acquire it in a fashion Apple wouldn't mind.
Do I believe in draconian EULA limitations? No. Is it perfectly legitimate to install OSX on a system not blessed by apple? I think it should be.
However, practically speaking, it's a bad path. Just as you have the right to use the software as you see fit, Apple has the right (even if a bad idea) to withhold updates/support from such people. If an OSX update fixes important security issues *and* kills the loophole you use to run the OS, Apple is not required to split it out. Best case, you always wait for a community that is potentially untrustworthy to bless patches and rerelease. However, Apple can sue those people for illegal redistribution.
If you don't want to play by their rules, why give them money to explicitly fight with them over how to use the software?
I'm not personally particularly interested in this instance, but I do think this long after a creation, it's really not helping the general advancement of society to have a company milking it long after the creator is dead. Everyone reasonably responsible for the idea is gone and their descendants have earned probably as much as could be reasonable without revamping it with new and novel twists. Scrabble becoming a generic would save us from a confusing namespace of tons of different names for the same thing. Imagine if the names chess and checkers were still protected.
Now, I'll preface this with a disclaimer that I avoid Fedora generally. I got reminded of why during a recent attempt to use it and follow it, it really punishes the users with inconsistant updates even after release.
That said, RPM dependencies are no more convoluted than deb dependencies. The difference is that originally, RH distros had only the rpm command and debian out of the gate recognized the need for both dpkg *and* apt. RPM distributions each have at least one repository management strategy now (YaST, yum, etc etc). So dependency hell is not one of their worst problems (though I do prefer some apt defaults more than yum, I won't say yum isn't up to the task).
But judging soley from the price of a single share is incomplete data. If I had a company at $10,000 dollars a share, but only 10 shares, it'd be a very low valued company (whether deservedly so or not). Many companies that are popularly tracked would have split on an ascent like Google's, but they chose not to. A better, but imperfect measure would be market cap.
That said, Google's market cap is 150 billion. CBS is at 12 billion or so. One analogy I've been fond of thinking of is comparing television advertising to internet advertising. It's imperfect, but still. The business models are different, as CBS has a higher demanding expense I would guess (TV show production/content creation/distribution I would guess more expensive than even Google's indexing of content from other people), but still over a magnitude more valuable does strike me as steep...
Ugh. MS's commercial success had little to do with technical merit and everything to do with sane and strategically superior business positioning.
People can bash MS all they want, but still, the bottom line is, if their product didn't work (and in the early days, better than anything else), it wouldn't be as popular.
In the early days, the key is at what price a total platform could be acquired for, and that it worked at least most of the time, even if not the best.
"you can get whatever flavor desktop you want, don't like gnome, get kde". .... Yeah, I want to support THAT mess with > 1000 desktops under my umbrella
Just because the world has choice doesn't mean you have to relay that to your organization. Do you pick a single source hardware vendor? How can you possibly do that when people in the world can buy something different than what you choose? The answer is you don't give a rat's ass about the general choice, because your organization picked one. People gripe and override your choice, well, tough for them if they call for help. MS itself has choices inherent. If you chose IE, and users went with Firefox, even in windows, how would this be different?
Macintosh had a GUI second (GEOS on the COMMODeORE was first), and it cost so much to make the hardware and software, nobody bought that crap (LISA). Then Mac himself came out, and it looked like shit. Windows solved both, in that you could have a clone, or a real IBM, and BOTH machines acted the same way.
To say Win3.11 looked better than the same era MacOS is laughable. The second half of your point does nail it, that IBM clones were the whole driving force of MS success. This is not a technical thing, this was a business maneuver, in part banking on a decision by IBM to create a market a certain way as to not be the only ones who could participate.
Let's talk about a clone and an IBM, each one running a different flavor of linux, and different desktop GUI's, and then lets talk about support.
Ok... And as the person who choses to support a standard desktop from RHEL, you care about how a KDE Novell platform would be supported... Precisely why? Why would you care about the existence of solutions you don't chose any more than you care about them now as a Windows shop cares about Linux based solutions already?
Now it's about entrenchment and leveraging the monopoly every way they can. And they get away with it too because in part the government swings that way today and also in part because people are afraid that without a monopoly, things would change and they aren't comfortable predicting how.
1. Vision. This is precisely what a single distribution generally has. They pick a set of coherent packages that fit a vision. The problem a *lot* of critics of Linux is that any and all use of the platform should be consistent because choice is too hard. It's like complaining that motor vehicles are inconsistent in a free market. Audi has some car with DSG, and yet a competitor offers a stickshift, it's sheer madness for the users! A Ford is drastically different from a Harley, how in the world do they expect people to chose between them and drive the same roads? The MS situation has skewed the software market in an unfortunate way relative to other sane markets.
2. Marketing. I think your point was a tad confused. In the consumer space, I do think a coherent commercial marketing thing isn't there, regardless of the technical state of things. In the 'trouble ticket' scenario, that's more than covered by Novell and RedHat. If you are not going to pay a person to have the responsibility to muck around with the internals anyway, you don't need to care that they are there. It's like saying a bus with the engine compartment welded shut is popular among drivers because then the drivers feel they'll never get blamed. Just because they are paid to operate them, doesn't mean they are expected to be a mechanic just because they can reach the engine.
3. Again, what? In corporations with lots of money, they do follow what their highly paid geeks think. It's the whole job. I don't see exactly what you are getting at when it comes to blaming the choice if dissatisfied. They can blame the vendor either way, or the architect who made the wrong choice for their needs.
And what is the BS that Red Hat doesn't guarantee anything? They guarantee their commercial product. They don't guarantee 'everything', meaning third party support, but MS doesn't either. MS will not be jumping to your aid because AutoCAD crashes, so I'm not seeing what your point is.
Linux has several sets of leaders/vision/marketing. I know, this sounds like an oxymoron, but it is no more so than characterizing other fields as having multiple sets of these. There are multiple sets of people doing this in 'console gaming' that don't even try to get along. There are multiple sets in cars, in computing, in retailing, any industry. The problem is people think of Linux as a specific thing, when they should recognize it as a framework/environment that produces coherent visions. You can't claim RHEL is not the realization of a consistent vision, and that Ubuntu is not one either. Despite a lot of similarities, it is plain that they are not totally consistent to each other, but then again neither are Sony and Nintendo.
This truly is a huge reason why consumers should be jumping at linux. Being beholden to a single vendor throughout your lifecycle is not a good thing. The unique thing 'Linux' brings to the table is a relatively low barrier to jumping vendors. No, jumping from RHEL to SLES isn't trivial, but it's much more plausible than Windows to OSX or vice-versa. Choice in all parts of the stack has revolutionized the PC industry, the server industry, the computer component industry. Don't like Intel, get AMD. Don't like Lenovo, get an ASUS. Don't like your current Dell solution, get IBM next time. In all cases you can do that because the products are equivalently usable. Bringing that to the OS is a logical good thing for the consumer.
Games do that, *particularly* online ones, and suddenly you have rather drastic balance issues. A lot of FPS is dumbed pretty down (auto-aim like crazy) for the consoles, due to the inaccuracy of the control pads, and it would just be ludicrously easy if someone had a correctly behaved mouse doing the aiming. If they castrated the mouselook to be fair, it would be pointless too.
I don't have the console as I haven't seen a single exclusive to bother. I'm saying actually that a lot of the new games I dislike on consoles, though I wouldn't mind them on a desktop.
Hell, even Final Fantasy 13 isn't going to be exclusive, and their main line has been exclusive to the playstation since the ps1.
Not entirely, they made a cash grab with 7 and 8 PC variants. Since 7 they've been so eye-candy demanding they required the storage of Sony's offerings. n64 cartridges precluded them, Gamecube was late and with smaller disks, xbox could have probably been sufficient for 10 and 12, but they were so late to the party. 13 was almost certainly going to be exclusive due in part to BD, but the MS bag of money probably just got too big to resist. Call it a conspiracy if you will, but Square/Enix all of a sudden got a bee in their bonnet about 360 releases. It might have been the mark of market decision of an exec, if not for the changing of Star Ocean to be 360 exclusive and the PS3 last remnant to be delayed to give the 360 variant top billing. To make that sort of decision takes some money bags changing hands.
I will say my chief gripe is that FPS developers have been dominating the PS3 scene too. The focus seems to be on things like Haze, Resistance, Orange Box, all this stuff. The 360 is chock full of them too. I don't mind a good FPS, but keep it an a desktop system, FPS and RTS I can't see playing any other way except with my good old mouse and keyboard.
If you think that's bad, try the PS3...
Seriously, I have a Wii and a PS3 and I've found more interesting games for Wii than the PS3. However, some decent PS2 games have still been coming out.
And the technical tradeoffs. However, the fact remains the ability to execute x86 and x86-64 opcodes is of vital importance practically speaking.