If you had a problem installing a program on Windows, would you blame Windows or the program?
That would depend upon the nature of the error.
However, you're missing the point. I'm not criticising the installation of a specific application, per se, just using it as an example of how the whole "package management" system is a fragile house of cards, constructed to fix a problem that shouldn't exist in the first place.
It's an example of one of those areas where the "Bazaar" fails.
If it's not in the repository, then.deb is double-click to install (on Debian-based system; I assume.rpm is double-click to install on Red Hat-based systems, and there's probably tools to make both double-click to install).jar is double click to install, and.bin is just "./program.bin" to install. Some.tar.gz programs (like Firefox) are just the program in zipped format, so you just extract it and run it. I've got plenty of programs on my system that are not in Ubuntu's repository, yet I haven't compiled any of them.
Again, you are labouring under the assumption all applications and the various support libraries they depend (frequently, specific versions thereof) exist in an appropriately prepackaged form. My whole complaint is about the problems that ensure when they _aren't_.
I'll concede that I don't really know a lot about running Linux in a business environment; I've only run it at home, so it might be more difficult to find and install programs that are more business oriented on more business oriented distros.
This aspect of the problem is the Linux communities' tendency to forsake stability for the latest and greatest. So distros that remain stable for reasonable periods of time (3 - 5 years) tend to be ignored because they're not new enough.
The programs you've installed were difficult, but not all programs are difficult to install, and I think on they are going to improve.
I'm not holding out hope, I must admit. Largely because there's not a lot of people who even recognise it as a problem, let alone want to come up with a solution.
Last I checked, each Windows program handles its own installation, its own updates (or not), and its own removal. There's often no consistency between software from the same company, much less in general.
This is because Windows (and OS X) programs are, almost to a unit, distributed in a self-contained fashion - all you need to make them work is whatever they come with and whatever version of Windows they say they support.
On Linux, installation with a package manager is identical for all software. Upgrades are handled automatically with the rest of the system. And all software removal is handled identically, from the same place.
And when the software isn't handled by the PM the whole house of cards is exposed for the fragile hack that it is.
One of those is a hack, and it's not the package manager. I'd love to hear why you think package managers are a hack. What are the deficiencies they're supposedly working around?
Firstly, the lack of any sort of consitent, reasonable level of base functionality across a reasonable number of Linux distributions. Secondly, the significant lack of interest in the Linux developer community for maintaining backwards compatibility.
When I install Program A, I shouldn't need to go out and track down specific versions of programs B and C, and support libraries D, E, F and G - nor should a "package manager" have to do so on my behalf - to make it work.
Package managers haven't been "an ugly hack" in quite a while either.
The whole damn _concept_ is little more than an ugly hack. Doesn't matter how much you polish that turd, it still stinks. It's still just working around fundamental stability and feature definciencies in the platform itself.
If you want to be taken seriously, you should probably try a "modern" distro other than LFS before making comments about Linux in general.
Your reading comprehension needs to improve. We don't use LFS. I said I had "done the whole LFS thing" in the past, I never said anything about using it today, or even recently.
(You also inadvertently highlight another problem with the Linux community - in a textbook example of circular reasoning, you need a "modern distro" to install $SOME_APPLICATION via the package manager, but if it can't handle it, it's because your distro isn't "modern" enough.)
I don't think you quite got what I was proposing. I was trying (probably poorly) to say that this hypothetical OS would be 100% usable out of the box. You'd get apps for it, and they would work just fine. If, and only if, you wanted to, you could add things to the OS like 3D desktop, and the apps would *still* work, because the OS interface hadn't changed for the applications, just the OS code would do something else with window calls, menu calls, etc. while still returning the same reactions to the code - option selected, menu visible, button pressed, and so on.
Why wouldn't it include the 3D desktop by default ? Who decides what "usable" is ? If it's "usable" without the extras, why does it need them ? How are users going to decide what "extras" they need(/want) to buy ?
You seem to be missing the two primary reasons for having a stable platform to target in the first place. Firstly, so developers know what functionality is _always_ present and, hence, what they can write their software to work with and, secondly, so end users can sit down in front of any machine running your platform and be able to use it, because it's always the same.
For the vast, vast majority of consumers, their computer is an appliance like their microware, TV or fridge. Windows and OS X cater to these people. For those who _really_ want to be able to tinker, and built their platform from scratch, there's Linux.
I get what you're saying, I just can't see how the system you're proposing differs in any significant amount to Windows and OS X as they exist today. By significant, I mean that you're basically saying "I want a platform just like OS X or Windows, but without $FEATURE". Where $FEATURE is something like the 3D desktop, or the media player, or the web browser - the one thing (or maybe a few things) that you, personally, don't think should be there by default. It's just like the definition of "bloat" - "everything that comes with the system that I don't personally want, is bloat". It's entirely subjective and has little - if any - technical or functional basis.
And Windows is any better? I can think of at least two times in the past when Windows systems were hit *hard* by trojans, not because of average users, but because a large number of system admins didn't feel like updating. If paid system admins don't want to do it, the average home user is screwed.
You're arguing the wrong discussion.
How many "average users" would need software outside of the 18,000+ packages in Debian's repository?
Thank you for reiterating my point. *IF* your application is in your distros package management system, then it's usually ok. It's when it isn't, that the nightmare starts.
I install a stunning amount of software, and I've only needed to go outside Debian's package system two or three times, and it was for software the "average user" almost certainly wouldn't be installing.
It shouldn't matter whether or not it's software "the "average user" almost certainly wouldn't be installing" - the problem is that "package managers" are an ugly hack to work around definciencies in the platform.
Hell, even when I compiled everything myself, the dependencies were nowhere near as bad as you make it out. Are you sure you're not confused with "DLL Hell" from Windows?
"DLL hell" hasn't been a problem on Windows for pushing a decade now. It's equal parts tragedy and irony, that the Linux community spent years criticising Windows about it, then reinvented and "improved" that wheel as well.
If your boss wants to hire somebody with a clue, who won't shit-can an entire day installing Linux and isn't baffled by Perl, can you have him call me? I've got a job, but I might be willing to $witch.
We've got more than enough cocksure developers ("and I've got a whole _network_ of Linux machines at home - I pulled most of them out of the dumpster for nothing !") here already, making my team's job harder than it needs to be, and wasting the company's money. We don't need any more.
If you're having problems installing a program on Linux, then the problem is the developer, not Linux.
Technically true, but practically irrelevant.
Most of the time installing things on Linux is at least as easy, if not easier, than installing on Windows. I've been using Linux (Ubuntu, to be specific) on my main computer for over a year and a half now, and I still have not found a need to learn how to recompile programs from CVS.
You seemed to have missed my main point, which was: if something is taken care of by $DISTRO_PACKAGE_MANAGEMENT, then you're usually ok (unless one of the inevitable dependencies is broken in some way) - but if it isn't, you're frequently in for a world of hurt.
The big problem here is that the distros that actually demonstrate a certain amount of stability, and hence be attractive to enterprises (primarily Red Hat and derivatives, like CentOS, and SuSe) are frequently ignored by Linux developers, who like to target the latest and greatest like Ubuntu and Fedora (the affliction is markedly less prevalent in the *BSD world - it's a shame companies like Oracle won't support them). The end result being either a) no packages at all; b) packages that rely on library versions newer than the ones shipped with the distro (recursively leading on to another whole world of hurt); or c) the compile-it-from source option (which typically results in the same situation as (b), only with even less stability).
This particular example is a non-trivial perl program. As such, it depends on the obligatory 50-odd tiny little perl modules that need to be installed, either via your distro's package management system (if you're lucky enough for them to be present), CPAN (which basically does the same thing, only kludgier) or manual installation (the last, desparate hope).
The fundamental problem is that the Linux community is largely driven by Engineers, like Linus (those, at least, I can excuse) and immature brats who think that there's nothing wrong whatsoever with insisting on using the latest versions of libraries (which, by the way, commonly aren't even API-compatible, let alone binary-compatible, with older versions) to write their software. Perl developers, in particular, seem to make it their life's mission to track down the most obscure modules possible to make their software dependent on.
Linux is a _nightmare_ of cross-dependencies and ugly hacks to work around them, all in the name of "choice". It's astounding that the OSS community spent years watching and (quite justifiably) criticising Windows for "DLL hell" in the early-to-mid-90s, then promptly went out and did basically the same thing, only making it worse in the process.
Just the other day, I had a friend ask me, "Why is my computer rebooting?" (Xp BSOD with only a 64K dump. Goes by pretty fast.) Once I explained it was "just normal Windows," they asked me, "How can I get Linux?"
Note to potential Linux advocates: lying in order to push your agenda is _not_ a good way to "advocate Linux".
The market for helmets is separate from the market for motorcycles. The market for speedometers is a subset of the market for motorcycles, unless it is a replacement part. No one sold motorcycle speedometers before there was a market for motorcycles. I'm sure you'll find some way to argue against that point, but it is not because you're an idiot that can't understand the concept. It is because you're a stubborn fool who does not want to understand the concept so they are intentionally being obtuse.
No, it's because I think you're wrong.
To pick an example, when "everyone else" is including a browser with their OS, how can the "browser market" *NOT* be considered a "subset" of the "OS market" ? A browser is useless without an OS to run it on. Some OSes were "bundling" a browser before Microsoft did.
Similarly with network stacks, GUIs, text editors, media players, memory management and advanced CPU scheduling (to pick a few other things you have previously argued Microsoft should be barred from including in Windows). Like Windows, most OSes include examples of at least some of these functionalities, some have done so for longer than Windows, some have not done so since their inception. *Clearly* this is currently considered functionality that can easily be identified as a "subset" of the OS market. Similarly, clearly, the definition of what these "subsets" are has changed over time. That they can *also* be replaced with third-party tools is completely and utterly irrelevant to the issue of what should be considered part of an operating system's standard featureset.
And you _still_ haven't explained why Microsoft are significantly improving Windows with every release, despite your insistence that "monopolies" do not do such a thing.
Have fun with your irrational beliefs you try to support with logic, instead of opinions formed from a logical consideration of the facts. It is sad that our educational system has so failed to teach logic, to the point that it is a way to justify belief, instead of make decisions or form opinions.
My "beliefs" are quite rational an well supported by logic. They just happen to be different from yours.
No shit? Linux From Scratch is hard to maintain? I'm shocked! Shocked! Did you really just say that Linux From Scatch was hard to maintain, so you stopped using Linux?
No.
I don't think Linux From Scratch is your problem here.
I never said it was.
I'm sure you'll point out some reason the average user is too stupid to do that, but it's a hell of a lot easier than LFS.
The issue is not that they are "too stupid" to do that, it is that they neither want to, nor should have to.
Why should I, as a user, have to worry about libraries?
You shouldn't. which was my whole point. "Package managers" are a massive kludge, not a solution.
I shouldn't. And with a distro like Debian or SuSe, I don't. I open Synaptic, click on the application I want, click "Apply", and the application is installed along with any necessary libraries. Oh, and it'll automatically get updated along with the rest of the system.
Yep, the kludge usually works fine if the software you want is maintained in the repositories. Tough luck if it isn't, though.
Try doing that on Mac or Windows.
I don't need to, because it's almost universally unnecessary. Typical Windows and Mac software either only uses the functionality built into the respective systems, or ship with all of it included.
As a developer, I still don't see your point. It makes very little difference to me if I'm using the API built into the OS, or a third party library. In one case I'll have to add a line to the build scripts. Big fuckin' deal.
If you're the poor end user who has to try and resolve the dependency hell that "third party library" usually produces, it *is* a "big fucking deal". Combined with the typical Linux developer's disinterest in maintaining any sort of decent binary compatibility or API stability ("because they can just grab the latest code from CVS and recompile") and you have one of the primary reasons Linux is having such difficulties penetrating past the tech-savvy userbase. Companies like Red Hat and SuSe have made heroic efforts to try and mitigate the problem, but most of the community simply isn't interested.
Having just wasted an entire day trying to configure a new RT server, this massive problem Linux has is fresh in my mind, since dealing with the typical clusterfuck of cascading dependencies inherent to any non-trivial perl program is something I've only just recently finished doing. However, I don't expect another idiotic "developer" like yourself[0] to understand why an entire day of lost productivity wasted on something that should be completely fucking trivial, without kludges like "package managers" isn't a "big deal".
[0] And I know exactly what you're like, because we've got a whole group of people like you here with much the same attitude, who think using some third-party library just for the hell of it is A-OK, since "it's just another line in a config file" - or, in our case, just another bit of software that needs to be rolled out to every desktop. No fucking concept whatsoever about actually _running_ an IT infrastructure.
I find that hard to believe. A new OS release that somehow makes old machines slower, and new machines faster?
Sure. Think things like better kernel locking and threading (good for multicpu systems, not so good for single CPU systems because of the extra overhead) or caching algorithms than can be tuned with the assumption of greater minimum measures of RAM.
On high-end (for the time) hardware, Windows 2000 was faster than NT 4.0. On low-end hardware, NT 4 was faster. The same pattern was repeated with XP and is being repeated with Vista (with the video, as you note).
It's relatively unusual for a newer OS to be faster than a prior release on low end hardware. It's not at all uncommon for the newer release to be faster on high-end hardware. A Linux 2.6 kernel, for example, runs vastly better on a 16 CPU machine than a 2.0 kernel would (if it runs at all) - but a 2.0.x kernel will run much faster on a 4MB RAM 386 than a 2.6.x will (assuming it would run at all ?).
As I said, OS X getting faster on lower-end hardware is very much an anomaly - and it's an anomaly that exists largely because when something's as slow as OS X was on release, there's not really anywhere to go but up. Today, still, OS X is sluggish even on relatively powerful hardware (which has led me to believe its problems are more fundamental than simple tuning can fix).
Well, maybe that's a signal we're looking at things incorrectly, then. Why not build a stable core - multitasking, networking, application sandboxing, list management, basic graphics with user-settable bitmaps and/or polygonal models -- the rest of the usual suspects like disk io and USB -- and then let the user decide if they want, for instance, to add a 3d desktop with voice and haptic features, widgets, zooming, 400 language compatibility (OSX carries a crapload of language stuff to your drive it doesn't really need to, for instance) and drivers for every printer ever known to man?
Primarily because the vast, vast majority of consumers lack the knowledge and, more importantly, the will, to do so.
Heck, *I* have zero interest in doing that sort of thing these days, and it wasn't that long ago I did the whole Linux-from-scratch thing, just for the hell of it. I'm more than happy to sacrifice some (dirt cheap) disk space and processor time, to save myself the effort of putting the whole thing together myself and subsequently having to keep it maintained. This is precisely the same reason I don't use Linux on my desktop - because it's more work to get everything going and keep it that way.
That almost sounds like a linux release, but the key thing missing in all linux versions is a stable and always-there set of GUI tools so applications can run on the OS itself.
Close. More important than the "set of GUI tools" is a standard, stable, "set of libraries" (I use the term "libraries", but I basically mean a stable, defined set of basic functionalities that will _always_ be present in a known form). This is a _huge_ feature than OS X (and Windows) has over Linux.
Hardware resources are _cheap_. My time - and developers' time - is _expensive_. Sacrificing hardware resources to get better software, quicker, is a more than reasonable tradeoff and, ultimately, the whole point of computers in the first place.
The point of software [like this] is not to use as little hardware resources as possible. The point of software is to make my life as easy as it possibly can and the hardware resources be damned.
Why would they accept an OS that gets slower with every release?
Because a) in many cases it isn't true (the higher end your hardware, the less true it is) and b) in the cases where it isn't, it's quite normal behaviour (eg: more recent versions of Linux are slower on low-end machines than older ones).
The only OS in recent memory that has improved in performance on low end machines with new releases is OS X, which has far, far more to do with how dismally slow it was at initial release and compiler improvements, than it does with Apple's OS development.
No, they are "forced" to buy Windows because there are no realistic alternatives.
Indeed. Precisely the reason I was "forced" to buy my motorbike.
If there was only one motorcycle you could buy that would get you where you needed to go and you run a business selling something other than motorcycles (like helmets) that the only motorcycle dealer has tied to the purchase of motorcycles, then you should be able to sue them.
My motorcycle came with numerous features that I was "forced" to have bundled, including a trip computer, a pillion seat and luggage.
Fortunately it also came with several things I wanted - mirrors, headlights and a muffler - for no extra charge, so that made up for it.
Please stay on topic and compare apples to apples here.
I am. Your implication is that Microsoft has bundled various features into Windows that consumers did not want and/or did not derive any benefit from. My point - as illustrated by the somewhat ridiculous example above - is that your implication is baseless, since whether the "extra features" provide value is _entirely_ dependent on perspective.
Almost every (I'd say every, but I'm sure you'll manage to find some obscure example and use it to proclaim complete and utter victory) piece of functionality "bundled" into Windows over the years has been in direct response to either competitors doing the same or customer demand. Moreover, they are quite relevant to the purpose of Windows as a product and the market it was targeting.
In general the product that wins a given sale is the one that best meets the customer's needs. By leveraging a monopoly I can create artificial problems with my competitor's products in that market. For example, I might give away a "free" helmet with every bike. In a market where neither helmets nor motorcycles are monopolized, this is perfectly fine. In a market where motorcycles are monopolized this means everyone was just forced to buy a helmet, even if it not the color, or style, or does not meet the safety requirements. Some will use it anyway. They bought an inferior product for them, because the monopolist artificially influenced the market, effectively doubling the price of competitor's helmets.
Doubling ? 2*0 = 0, I think your maths needs some work.
Your example here is still inaccurate, however, because it implies there are no alternatives to Windows when there are. They might not be as good (although I'd argue Ubuntu is pretty damn close), but that is no more the fault of Microsoft than all the other helmets on the market not being as good as the one bundled with the bike is the fault of the motorcycle manufacturer.
MS are allowed to make products better, they just aren't allowed to tie them to Windows.
Ie: Microsoft aren't allowed to make Windows better.
For example, they can invent a new protocol (exchange) and have it talk to their server, but they can't keep that protocol secret so other server manufacturers can't use that feature of their monopolized desktop.
Nor can they be allowed to have trade secrets, it seems...
They can make the best media player on the planet, they just can't bundle it with Windows insuring it will be everywhere, when they refuse to bundle competitor's media players as well.
Right. Because customers would clearly benefit from having a thousand different media players installed on their PCs by default. Hey, it would make Windows just like Linux !
This idea is so ridiculous I'm surprised someone of your intelligence would even suggest it. It's the kind of thing that gets littered throughout Slashdot comments by fifteen-year-olds who, when they make it, are thinking "Microsoft should bundle WMP and $MY_FAVOURITE_MEDIA_PLAYERS with Windows, because then I wouldn't have to go and download them !".
How many different media players are out there ? Who is going to keep track of _all_ of them to ensure they are all bundled into *and work wit
This is an empty assertion. Assuming the only interest of sellers in a capitalist marketplace is to make as much money as possible, you could just as easily assert the goal of all sellers is to become dictator for life.
You could, but given that changes the scope of the discussion from economics to politics, it's hardly in context.
(Or you could take the perspective that in capitalism, a vendor with a genuine monopoly _is_ a dictator.)
How does expecting companies to obey the laws and not take funds by leveraging a monopoly differ?
Because a) monopoly status is not objectively defined and b) typically a company connot know if it is "abusing monopoly" until _after_ it has done.
There are qualitative differences between the concepts of anti-trust law, and the concepts of fraud and property laws. Do not try to conflate them, as it just makes you look dishonest.
A capitalist seller can make boatloads of money selling into a competitive marketplace.
Not as much as they can make as a monopoly, which is the point.
Illegally leveraging a monopoly merely lets them do so without operating within the bounds of a free market.
And, hence, without the downsides that brings. Again, the point.
If the same model of TV is for sale for $50 at one vendor and an identical TV is for sale for $2000 from a different vendor, it is in the consumer's best interest to buy it for $50. Does that mean capitalism is working? What if the $50 TV is being sold by the government who subsidized it with tax dollars collected from everyone? It does not change what is in the consumer's best interest. What if the TV was stolen? Does that mean capitalism is working? Your assertion is thus proven false.
I should not need to specifically say "when all else is equal" to stop you playing pointless semantic games.
What is best in the long term for society (innovation, competition, lower prices) is not the same thing as what is in the best interests of an individual purchaser and even if it is, that may not be apparent to the purchaser. Either way, proper decision making for society is undermined by the monopoly.
Again, you are straying from context. Capitalism has nothing to do with "the good of society", it is about the efficient production and distribution of goods and services. It is an inherently amoral system.
Did you see the final feature set for Vista? The lion's share of the new "features" do not benefit consumers, they benefit MS.
"Benefitting consumers" and "benefitting Microsoft" are inseperable. Microsoft cannot (feasibly) "benefit" consumers without giving them a product they want to buy and, hence, benefitting themselves.
The "lions share" of the new features in Vista have direct and demonstrable benefits to users.
Built in DRM allows them to move into media markets. Built in XPS allows them to move into the PDF/PS tool market, Defender allows them to move into the antivirus software market, indexed search allows them to take market from Google's toolbar, IE improvements allows them to illegally take share from Firefox, etc., etc.
Built in DRM allows users to consume content they would otherwise be unable to. Built in XPS allows users to create and transfer print-ready documents without relying on (or having to pay for) third party software. Defender assists users in identifying and removing malware, without relying on (or having to pay for) third-party software. Indexed search allows users to find their data quicker and easier than previous versions. IE improvements make browsing the web easier and safer.
*All* of these improvements in Vista give direct and demonstrable benefit to users.
There are comparatively few improvements in Vista that are not designed to attack something outside the desktop OS market.
With the bias and perspective you have demonstrated so far, I'm astounded you concede there are _any_ improvements in V
And for the more enlightened users, I believe it mostly comes down to DRM. Even many average users have some idea that DRM is overtly anti-consumer. One doesn't have to know specifically how DRM is implemented in Vista. Anyone who's had the displeasure of experiencing how it's been "working" in a current environment knows enough.
You do need to know when you're going to criticise its implications, as you have tried to do.
The DRM in Vista does _nothing_ unless you have DRM-encumbered media - and even if you *have* DRM-encumbered media, Vista doesn't do anything that other platforms (be they computer or standalone) will also do.
For "enlightened users", the DRM in Vista should be a complete and utter non-issue, because they should be able to figure out that it's either a) irrelevant or b) inevitable (depending on what they're trying to do).
The PC has traditionally been been associated with open architecture and freedom for the user to do with it whatever he or she wishes.
The PC has traditionally been cheap and pervasive. The "open architecture" you speak of is something the vast bulk of users are only vaguely aware of, if they are at all - and it's most certainly not something the PC was initially designed to have.
This is a very important concept to many people, regardless of what those dependent upon MS for their liveliehoods would like to believe.
It may be a very important concept to many people, but it sure as hell isn't an important concept to most of them. Most people's criteria for choosing a PC are a) does it run the software I want and b) can I afford it.
If an "open architecture" were as important as you say, Apple's sales would be going downhill, not improving (and OS X wouldn't be attracting anything close to the proportion of "advanced users" that it is).
Aero is a visual upgrade; it should be an add-on, similar to Plus! was for Win98. Because it's not, IMO, a upgrade; it doesn't actually do anything to improve the ability to do anything.
Aero is vastly more than a "visual upgrade". Indeed, the "visual" part of Vista's new display system is the _least_ significant aspect.
Security & Safety: or "making sure you can't use the software and media you bought 3 years ago". a joke. Win2k with Tiny Personal Firewall 2.0.14 and maybe zone alarm is much more secure.
False.
But is it not obvious that this is not really an upgrade?
Vista is a _massive_ upgrade. It rivals Apple's update of NeXTSTEP to OS X in scope. The entire network and audio stacks have been rewritten. The display system is completely new and unrivalled in its capabilities. While the changes to "security" are largely in the UI aspect, the amount of work done behind the scenes to mitigate the impact on legacy software (eg: virtualising the Registry) is significant. The work done in the kernel and lower system levels regarding threading, locking, memory management and component dependency isolation is worth a whole x.0 update on its own. Improvements to the remote managability and deployment capabilities are also large.
Vista is, unquestionably, the biggest update to Windows (from both the technical and non-technical perspective) since Windows 3.1 -> Windows 95. That it's being met with relative disinterest is a much bigger compliment to Windows XP (and 2000) than it is criticism of Vista. Vista could have been released with half as much work done and _still_ have easily justified a completely new Windows release.
With every little thing that gets folded in to the system and thereby run by the draconian DRM/trusted computing system, you are eroding your own freedoms to watch what you want, create what you want, read what you want.
Hm, you seem to think Gutmann's posts and rebuttals are slashdot comments. I'm referring to the article A Cost Analysis of Windows Vista Content Protection. This is the work of a notable security expert, who has analysed and written a report on the AACS specification and its implementation in Vista. This is not some evangelist slashdot post, it is a serious paper, heavily based in fact by a noted security expert. It's become extremely popular.
I'm aware of who Gutman is, I've read the paper several times and it's far more alarmist FUD than fact (as is most of the commentary you have written). The rebuttal from Microsoft covers the most egrarious and inaccurate statements, but I'm not going to bother repeating them, because if you don't believe the developer of the Operating System then I - a random Slashdot poster - am sure as hell not going to change your mind.
(If one were to make all the worst-case assumptions about the potentially negative implications of open source code and write a paper about how bad Linux was because of them, then Slashdot readers would (rightfully, albeit for the wrong reasons) decry it as baseless FUD and tear its author to shreds. Yet, when someone does the same thing about Vista, they are cheered on as some sort of saint.)
So, I repeat, to the parent & everyone else who even begins to consider "upgrading" to Vista: Is there ANY feature of Vista that will improve my ability to do ANYTHING AT ALL I currently do on my DRM-free Win2k machine?
This is not a repeat, it is a different question. "What's new ?" is not the same as "what's new that benefits me ?"
Make no mistake: Monopoly isn't a feature of capitalism, it's a failure. Free markets do not work without government to do things like: provide courts to enforce contracts etc, and intervene when a market failure occurs, such as with monopoly.
Monopolies (well, a single monopoly) are the only logical conclusion of an unregulated free market - at least so long as scarcity exists (and if it didn't, well, there wouldn't be any need for capitalism ).
(Of course, you couldn't still call it a "free market" by the time that happened, but it would be inevitable.)
The big question is how does someone patent someone else's idea by accident after admitting they had copied it.
"Someone" doesn't. However, Microsoft is not "someone", they are a collections of tens of thousands of "someones". As anyone who has even a passing acquaintance with groups of people will be able to tell you, even the most organised ones frequently have instances of miscommunication and confusions.
The surprising thing is not that it happened, it's that it hasn't happened more frequently.
Please educate yourself on monopolies before trying to argue this subject. Monopolies are anti-capitalist. They allow a company to break the capitalist system by undermining the benefits it normally provides, which is why almost every country in the world restricts their action. The basic idea of the capitalist method is that you can get more innovation by appealing to people's greed, using competition. Maybe three companies all duplicate the same research, development, and production facilities. That's pretty wasteful. But, because all of them are motivated by the market (greed) to deliver the best product to customers, they make good decisions and in the end consumers get a choice among several superior products instead of one inferior one.
The irony being, of course, that the ultimate objective of any seller in a capitalist marketplace is to become a monopoly.
As a result, the most innovative product does not win the market and consumers get an inferior product, despite the fact that they are still making the correct decisions for their own self-interest.
By definition, if consumers are making the correct decisions for their own self-interest, capitalism is working. This holds true even when they are making decisions that you - or other "we know betters" like you - would not make.
Unless, of course, you assert there is some way of defining, objectively, what "the most innovative product" is (what's the capitalist word for God ?).
"The most innovative" (or "best", if you prefer) product never wins, because "best" is a subjective, relative metric. No one product can ever be the "best", because no product will ever be everything to everyone - ie: perfect.
(And even if such a product ever was invented, the concept of anti-trust would be compelled to make it illegal.)
If you disagree with me that is fine, but please do not bother to argue this unless you actually understand how monopolies and capitalism interact.
If Microsoft is a monopoly, why do they continually improve their products (in measurable, objective ways), especially with features the vast majority of their customers are unaware of ?
Yes, they slowed innovation to a crawl and turned one of the fastest growing and most promising chunks of that market into a slow moving market that has held back related fields of study for a decade. We're easily 5 or 10 years behind where we could be if our anti-trust law was enforced.
Why don't you just say we're 73.85493% worse off ? It would be about as valid.
What exactly HAS Microsoft done for the tech industry?
Facilitated the commoditisation of computers.
(You might (well, will) argue that "anyone could have done it". I will not disagree, however, the historical fact remains that it was Microsoft who did.)
Tell me, just what has microsoft done for the computer industry?
I studied MS extensively for a political science thesis. My conclusion is that MS is evil.
So if you call Microsoft "evil", what do you call, say, Tobacco corporations ? You know, to communicate a bit of scale ? Reallyreallyreallyreallyreallyreallyreallyreallyre allyreallyreallyEvil ?
Slashdot Mirror
If you had a problem installing a program on Windows, would you blame Windows or the program?
That would depend upon the nature of the error.
However, you're missing the point. I'm not criticising the installation of a specific application, per se, just using it as an example of how the whole "package management" system is a fragile house of cards, constructed to fix a problem that shouldn't exist in the first place.
It's an example of one of those areas where the "Bazaar" fails.
If it's not in the repository, then .deb is double-click to install (on Debian-based system; I assume .rpm is double-click to install on Red Hat-based systems, and there's probably tools to make both double-click to install) .jar is double click to install, and .bin is just "./program.bin" to install. Some .tar.gz programs (like Firefox) are just the program in zipped format, so you just extract it and run it. I've got plenty of programs on my system that are not in Ubuntu's repository, yet I haven't compiled any of them.
Again, you are labouring under the assumption all applications and the various support libraries they depend (frequently, specific versions thereof) exist in an appropriately prepackaged form. My whole complaint is about the problems that ensure when they _aren't_.
I'll concede that I don't really know a lot about running Linux in a business environment; I've only run it at home, so it might be more difficult to find and install programs that are more business oriented on more business oriented distros.
This aspect of the problem is the Linux communities' tendency to forsake stability for the latest and greatest. So distros that remain stable for reasonable periods of time (3 - 5 years) tend to be ignored because they're not new enough.
The programs you've installed were difficult, but not all programs are difficult to install, and I think on they are going to improve.
I'm not holding out hope, I must admit. Largely because there's not a lot of people who even recognise it as a problem, let alone want to come up with a solution.
You didn't answer my question...
That's because it's a straw man.
Last I checked, each Windows program handles its own installation, its own updates (or not), and its own removal. There's often no consistency between software from the same company, much less in general.
This is because Windows (and OS X) programs are, almost to a unit, distributed in a self-contained fashion - all you need to make them work is whatever they come with and whatever version of Windows they say they support.
On Linux, installation with a package manager is identical for all software. Upgrades are handled automatically with the rest of the system. And all software removal is handled identically, from the same place.
And when the software isn't handled by the PM the whole house of cards is exposed for the fragile hack that it is.
One of those is a hack, and it's not the package manager. I'd love to hear why you think package managers are a hack. What are the deficiencies they're supposedly working around?
Firstly, the lack of any sort of consitent, reasonable level of base functionality across a reasonable number of Linux distributions. Secondly, the significant lack of interest in the Linux developer community for maintaining backwards compatibility.
When I install Program A, I shouldn't need to go out and track down specific versions of programs B and C, and support libraries D, E, F and G - nor should a "package manager" have to do so on my behalf - to make it work.
Package managers haven't been "an ugly hack" in quite a while either.
The whole damn _concept_ is little more than an ugly hack. Doesn't matter how much you polish that turd, it still stinks. It's still just working around fundamental stability and feature definciencies in the platform itself.
If you want to be taken seriously, you should probably try a "modern" distro other than LFS before making comments about Linux in general.
Your reading comprehension needs to improve. We don't use LFS. I said I had "done the whole LFS thing" in the past, I never said anything about using it today, or even recently.
(You also inadvertently highlight another problem with the Linux community - in a textbook example of circular reasoning, you need a "modern distro" to install $SOME_APPLICATION via the package manager, but if it can't handle it, it's because your distro isn't "modern" enough.)
I don't think you quite got what I was proposing. I was trying (probably poorly) to say that this hypothetical OS would be 100% usable out of the box. You'd get apps for it, and they would work just fine. If, and only if, you wanted to, you could add things to the OS like 3D desktop, and the apps would *still* work, because the OS interface hadn't changed for the applications, just the OS code would do something else with window calls, menu calls, etc. while still returning the same reactions to the code - option selected, menu visible, button pressed, and so on.
Why wouldn't it include the 3D desktop by default ? Who decides what "usable" is ? If it's "usable" without the extras, why does it need them ? How are users going to decide what "extras" they need(/want) to buy ?
You seem to be missing the two primary reasons for having a stable platform to target in the first place. Firstly, so developers know what functionality is _always_ present and, hence, what they can write their software to work with and, secondly, so end users can sit down in front of any machine running your platform and be able to use it, because it's always the same.
For the vast, vast majority of consumers, their computer is an appliance like their microware, TV or fridge. Windows and OS X cater to these people. For those who _really_ want to be able to tinker, and built their platform from scratch, there's Linux.
I get what you're saying, I just can't see how the system you're proposing differs in any significant amount to Windows and OS X as they exist today. By significant, I mean that you're basically saying "I want a platform just like OS X or Windows, but without $FEATURE". Where $FEATURE is something like the 3D desktop, or the media player, or the web browser - the one thing (or maybe a few things) that you, personally, don't think should be there by default. It's just like the definition of "bloat" - "everything that comes with the system that I don't personally want, is bloat". It's entirely subjective and has little - if any - technical or functional basis.
And Windows is any better? I can think of at least two times in the past when Windows systems were hit *hard* by trojans, not because of average users, but because a large number of system admins didn't feel like updating. If paid system admins don't want to do it, the average home user is screwed.
You're arguing the wrong discussion.
How many "average users" would need software outside of the 18,000+ packages in Debian's repository?
Thank you for reiterating my point. *IF* your application is in your distros package management system, then it's usually ok. It's when it isn't, that the nightmare starts.
I install a stunning amount of software, and I've only needed to go outside Debian's package system two or three times, and it was for software the "average user" almost certainly wouldn't be installing.
It shouldn't matter whether or not it's software "the "average user" almost certainly wouldn't be installing" - the problem is that "package managers" are an ugly hack to work around definciencies in the platform.
Hell, even when I compiled everything myself, the dependencies were nowhere near as bad as you make it out. Are you sure you're not confused with "DLL Hell" from Windows?
"DLL hell" hasn't been a problem on Windows for pushing a decade now. It's equal parts tragedy and irony, that the Linux community spent years criticising Windows about it, then reinvented and "improved" that wheel as well.
If your boss wants to hire somebody with a clue, who won't shit-can an entire day installing Linux and isn't baffled by Perl, can you have him call me? I've got a job, but I might be willing to $witch.
We've got more than enough cocksure developers ("and I've got a whole _network_ of Linux machines at home - I pulled most of them out of the dumpster for nothing !") here already, making my team's job harder than it needs to be, and wasting the company's money. We don't need any more.
If you're having problems installing a program on Linux, then the problem is the developer, not Linux.
Technically true, but practically irrelevant.
Most of the time installing things on Linux is at least as easy, if not easier, than installing on Windows. I've been using Linux (Ubuntu, to be specific) on my main computer for over a year and a half now, and I still have not found a need to learn how to recompile programs from CVS.
You seemed to have missed my main point, which was: if something is taken care of by $DISTRO_PACKAGE_MANAGEMENT, then you're usually ok (unless one of the inevitable dependencies is broken in some way) - but if it isn't, you're frequently in for a world of hurt.
The big problem here is that the distros that actually demonstrate a certain amount of stability, and hence be attractive to enterprises (primarily Red Hat and derivatives, like CentOS, and SuSe) are frequently ignored by Linux developers, who like to target the latest and greatest like Ubuntu and Fedora (the affliction is markedly less prevalent in the *BSD world - it's a shame companies like Oracle won't support them). The end result being either a) no packages at all; b) packages that rely on library versions newer than the ones shipped with the distro (recursively leading on to another whole world of hurt); or c) the compile-it-from source option (which typically results in the same situation as (b), only with even less stability).
This particular example is a non-trivial perl program. As such, it depends on the obligatory 50-odd tiny little perl modules that need to be installed, either via your distro's package management system (if you're lucky enough for them to be present), CPAN (which basically does the same thing, only kludgier) or manual installation (the last, desparate hope).
The fundamental problem is that the Linux community is largely driven by Engineers, like Linus (those, at least, I can excuse) and immature brats who think that there's nothing wrong whatsoever with insisting on using the latest versions of libraries (which, by the way, commonly aren't even API-compatible, let alone binary-compatible, with older versions) to write their software. Perl developers, in particular, seem to make it their life's mission to track down the most obscure modules possible to make their software dependent on.
Linux is a _nightmare_ of cross-dependencies and ugly hacks to work around them, all in the name of "choice". It's astounding that the OSS community spent years watching and (quite justifiably) criticising Windows for "DLL hell" in the early-to-mid-90s, then promptly went out and did basically the same thing, only making it worse in the process.
Just the other day, I had a friend ask me, "Why is my computer rebooting?" (Xp BSOD with only a 64K dump. Goes by pretty fast.) Once I explained it was "just normal Windows," they asked me, "How can I get Linux?"
Note to potential Linux advocates: lying in order to push your agenda is _not_ a good way to "advocate Linux".
The market for helmets is separate from the market for motorcycles. The market for speedometers is a subset of the market for motorcycles, unless it is a replacement part. No one sold motorcycle speedometers before there was a market for motorcycles. I'm sure you'll find some way to argue against that point, but it is not because you're an idiot that can't understand the concept. It is because you're a stubborn fool who does not want to understand the concept so they are intentionally being obtuse.
No, it's because I think you're wrong.
To pick an example, when "everyone else" is including a browser with their OS, how can the "browser market" *NOT* be considered a "subset" of the "OS market" ? A browser is useless without an OS to run it on. Some OSes were "bundling" a browser before Microsoft did.
Similarly with network stacks, GUIs, text editors, media players, memory management and advanced CPU scheduling (to pick a few other things you have previously argued Microsoft should be barred from including in Windows). Like Windows, most OSes include examples of at least some of these functionalities, some have done so for longer than Windows, some have not done so since their inception. *Clearly* this is currently considered functionality that can easily be identified as a "subset" of the OS market. Similarly, clearly, the definition of what these "subsets" are has changed over time. That they can *also* be replaced with third-party tools is completely and utterly irrelevant to the issue of what should be considered part of an operating system's standard featureset.
And you _still_ haven't explained why Microsoft are significantly improving Windows with every release, despite your insistence that "monopolies" do not do such a thing.
Have fun with your irrational beliefs you try to support with logic, instead of opinions formed from a logical consideration of the facts. It is sad that our educational system has so failed to teach logic, to the point that it is a way to justify belief, instead of make decisions or form opinions.
My "beliefs" are quite rational an well supported by logic. They just happen to be different from yours.
No shit? Linux From Scratch is hard to maintain? I'm shocked! Shocked! Did you really just say that Linux From Scatch was hard to maintain, so you stopped using Linux?
No.
I don't think Linux From Scratch is your problem here.
I never said it was.
I'm sure you'll point out some reason the average user is too stupid to do that, but it's a hell of a lot easier than LFS.
The issue is not that they are "too stupid" to do that, it is that they neither want to, nor should have to.
Why should I, as a user, have to worry about libraries?
You shouldn't. which was my whole point. "Package managers" are a massive kludge, not a solution.
I shouldn't. And with a distro like Debian or SuSe, I don't. I open Synaptic, click on the application I want, click "Apply", and the application is installed along with any necessary libraries. Oh, and it'll automatically get updated along with the rest of the system.
Yep, the kludge usually works fine if the software you want is maintained in the repositories. Tough luck if it isn't, though.
Try doing that on Mac or Windows.
I don't need to, because it's almost universally unnecessary. Typical Windows and Mac software either only uses the functionality built into the respective systems, or ship with all of it included.
As a developer, I still don't see your point. It makes very little difference to me if I'm using the API built into the OS, or a third party library. In one case I'll have to add a line to the build scripts. Big fuckin' deal.
If you're the poor end user who has to try and resolve the dependency hell that "third party library" usually produces, it *is* a "big fucking deal". Combined with the typical Linux developer's disinterest in maintaining any sort of decent binary compatibility or API stability ("because they can just grab the latest code from CVS and recompile") and you have one of the primary reasons Linux is having such difficulties penetrating past the tech-savvy userbase. Companies like Red Hat and SuSe have made heroic efforts to try and mitigate the problem, but most of the community simply isn't interested.
Having just wasted an entire day trying to configure a new RT server, this massive problem Linux has is fresh in my mind, since dealing with the typical clusterfuck of cascading dependencies inherent to any non-trivial perl program is something I've only just recently finished doing. However, I don't expect another idiotic "developer" like yourself[0] to understand why an entire day of lost productivity wasted on something that should be completely fucking trivial, without kludges like "package managers" isn't a "big deal".
[0] And I know exactly what you're like, because we've got a whole group of people like you here with much the same attitude, who think using some third-party library just for the hell of it is A-OK, since "it's just another line in a config file" - or, in our case, just another bit of software that needs to be rolled out to every desktop. No fucking concept whatsoever about actually _running_ an IT infrastructure.
I find that hard to believe. A new OS release that somehow makes old machines slower, and new machines faster?
Sure. Think things like better kernel locking and threading (good for multicpu systems, not so good for single CPU systems because of the extra overhead) or caching algorithms than can be tuned with the assumption of greater minimum measures of RAM.
On high-end (for the time) hardware, Windows 2000 was faster than NT 4.0. On low-end hardware, NT 4 was faster. The same pattern was repeated with XP and is being repeated with Vista (with the video, as you note).
It's relatively unusual for a newer OS to be faster than a prior release on low end hardware. It's not at all uncommon for the newer release to be faster on high-end hardware. A Linux 2.6 kernel, for example, runs vastly better on a 16 CPU machine than a 2.0 kernel would (if it runs at all) - but a 2.0.x kernel will run much faster on a 4MB RAM 386 than a 2.6.x will (assuming it would run at all ?).
As I said, OS X getting faster on lower-end hardware is very much an anomaly - and it's an anomaly that exists largely because when something's as slow as OS X was on release, there's not really anywhere to go but up. Today, still, OS X is sluggish even on relatively powerful hardware (which has led me to believe its problems are more fundamental than simple tuning can fix).
Well, maybe that's a signal we're looking at things incorrectly, then. Why not build a stable core - multitasking, networking, application sandboxing, list management, basic graphics with user-settable bitmaps and/or polygonal models -- the rest of the usual suspects like disk io and USB -- and then let the user decide if they want, for instance, to add a 3d desktop with voice and haptic features, widgets, zooming, 400 language compatibility (OSX carries a crapload of language stuff to your drive it doesn't really need to, for instance) and drivers for every printer ever known to man?
Primarily because the vast, vast majority of consumers lack the knowledge and, more importantly, the will, to do so.
Heck, *I* have zero interest in doing that sort of thing these days, and it wasn't that long ago I did the whole Linux-from-scratch thing, just for the hell of it. I'm more than happy to sacrifice some (dirt cheap) disk space and processor time, to save myself the effort of putting the whole thing together myself and subsequently having to keep it maintained. This is precisely the same reason I don't use Linux on my desktop - because it's more work to get everything going and keep it that way.
That almost sounds like a linux release, but the key thing missing in all linux versions is a stable and always-there set of GUI tools so applications can run on the OS itself.
Close. More important than the "set of GUI tools" is a standard, stable, "set of libraries" (I use the term "libraries", but I basically mean a stable, defined set of basic functionalities that will _always_ be present in a known form). This is a _huge_ feature than OS X (and Windows) has over Linux.
Hardware resources are _cheap_. My time - and developers' time - is _expensive_. Sacrificing hardware resources to get better software, quicker, is a more than reasonable tradeoff and, ultimately, the whole point of computers in the first place.
The point of software [like this] is not to use as little hardware resources as possible. The point of software is to make my life as easy as it possibly can and the hardware resources be damned.
Why would they accept an OS that gets slower with every release?
Because a) in many cases it isn't true (the higher end your hardware, the less true it is) and b) in the cases where it isn't, it's quite normal behaviour (eg: more recent versions of Linux are slower on low-end machines than older ones).
The only OS in recent memory that has improved in performance on low end machines with new releases is OS X, which has far, far more to do with how dismally slow it was at initial release and compiler improvements, than it does with Apple's OS development.
The relative disinterest with which Vista is being greeted, is far more of a compliment to Windows XP and 2000 than it is criticism of Vista.
No, they are "forced" to buy Windows because there are no realistic alternatives.
Indeed. Precisely the reason I was "forced" to buy my motorbike.
If there was only one motorcycle you could buy that would get you where you needed to go and you run a business selling something other than motorcycles (like helmets) that the only motorcycle dealer has tied to the purchase of motorcycles, then you should be able to sue them.
My motorcycle came with numerous features that I was "forced" to have bundled, including a trip computer, a pillion seat and luggage.
Fortunately it also came with several things I wanted - mirrors, headlights and a muffler - for no extra charge, so that made up for it.
Please stay on topic and compare apples to apples here.
I am. Your implication is that Microsoft has bundled various features into Windows that consumers did not want and/or did not derive any benefit from. My point - as illustrated by the somewhat ridiculous example above - is that your implication is baseless, since whether the "extra features" provide value is _entirely_ dependent on perspective.
Almost every (I'd say every, but I'm sure you'll manage to find some obscure example and use it to proclaim complete and utter victory) piece of functionality "bundled" into Windows over the years has been in direct response to either competitors doing the same or customer demand. Moreover, they are quite relevant to the purpose of Windows as a product and the market it was targeting.
In general the product that wins a given sale is the one that best meets the customer's needs. By leveraging a monopoly I can create artificial problems with my competitor's products in that market. For example, I might give away a "free" helmet with every bike. In a market where neither helmets nor motorcycles are monopolized, this is perfectly fine. In a market where motorcycles are monopolized this means everyone was just forced to buy a helmet, even if it not the color, or style, or does not meet the safety requirements. Some will use it anyway. They bought an inferior product for them, because the monopolist artificially influenced the market, effectively doubling the price of competitor's helmets.
Doubling ? 2*0 = 0, I think your maths needs some work.
Your example here is still inaccurate, however, because it implies there are no alternatives to Windows when there are. They might not be as good (although I'd argue Ubuntu is pretty damn close), but that is no more the fault of Microsoft than all the other helmets on the market not being as good as the one bundled with the bike is the fault of the motorcycle manufacturer.
MS are allowed to make products better, they just aren't allowed to tie them to Windows.
Ie: Microsoft aren't allowed to make Windows better.
For example, they can invent a new protocol (exchange) and have it talk to their server, but they can't keep that protocol secret so other server manufacturers can't use that feature of their monopolized desktop.
Nor can they be allowed to have trade secrets, it seems...
They can make the best media player on the planet, they just can't bundle it with Windows insuring it will be everywhere, when they refuse to bundle competitor's media players as well.
Right. Because customers would clearly benefit from having a thousand different media players installed on their PCs by default. Hey, it would make Windows just like Linux !
This idea is so ridiculous I'm surprised someone of your intelligence would even suggest it. It's the kind of thing that gets littered throughout Slashdot comments by fifteen-year-olds who, when they make it, are thinking "Microsoft should bundle WMP and $MY_FAVOURITE_MEDIA_PLAYERS with Windows, because then I wouldn't have to go and download them !".
How many different media players are out there ? Who is going to keep track of _all_ of them to ensure they are all bundled into *and work wit
This is an empty assertion. Assuming the only interest of sellers in a capitalist marketplace is to make as much money as possible, you could just as easily assert the goal of all sellers is to become dictator for life.
You could, but given that changes the scope of the discussion from economics to politics, it's hardly in context.
(Or you could take the perspective that in capitalism, a vendor with a genuine monopoly _is_ a dictator.)
How does expecting companies to obey the laws and not take funds by leveraging a monopoly differ?
Because a) monopoly status is not objectively defined and b) typically a company connot know if it is "abusing monopoly" until _after_ it has done.
There are qualitative differences between the concepts of anti-trust law, and the concepts of fraud and property laws. Do not try to conflate them, as it just makes you look dishonest.
A capitalist seller can make boatloads of money selling into a competitive marketplace.
Not as much as they can make as a monopoly, which is the point.
Illegally leveraging a monopoly merely lets them do so without operating within the bounds of a free market.
And, hence, without the downsides that brings. Again, the point.
If the same model of TV is for sale for $50 at one vendor and an identical TV is for sale for $2000 from a different vendor, it is in the consumer's best interest to buy it for $50. Does that mean capitalism is working? What if the $50 TV is being sold by the government who subsidized it with tax dollars collected from everyone? It does not change what is in the consumer's best interest. What if the TV was stolen? Does that mean capitalism is working? Your assertion is thus proven false.
I should not need to specifically say "when all else is equal" to stop you playing pointless semantic games.
What is best in the long term for society (innovation, competition, lower prices) is not the same thing as what is in the best interests of an individual purchaser and even if it is, that may not be apparent to the purchaser. Either way, proper decision making for society is undermined by the monopoly.
Again, you are straying from context. Capitalism has nothing to do with "the good of society", it is about the efficient production and distribution of goods and services. It is an inherently amoral system.
Did you see the final feature set for Vista? The lion's share of the new "features" do not benefit consumers, they benefit MS.
"Benefitting consumers" and "benefitting Microsoft" are inseperable. Microsoft cannot (feasibly) "benefit" consumers without giving them a product they want to buy and, hence, benefitting themselves.
The "lions share" of the new features in Vista have direct and demonstrable benefits to users.
Built in DRM allows them to move into media markets. Built in XPS allows them to move into the PDF/PS tool market, Defender allows them to move into the antivirus software market, indexed search allows them to take market from Google's toolbar, IE improvements allows them to illegally take share from Firefox, etc., etc.
Built in DRM allows users to consume content they would otherwise be unable to. Built in XPS allows users to create and transfer print-ready documents without relying on (or having to pay for) third party software. Defender assists users in identifying and removing malware, without relying on (or having to pay for) third-party software. Indexed search allows users to find their data quicker and easier than previous versions. IE improvements make browsing the web easier and safer.
*All* of these improvements in Vista give direct and demonstrable benefit to users.
There are comparatively few improvements in Vista that are not designed to attack something outside the desktop OS market.
With the bias and perspective you have demonstrated so far, I'm astounded you concede there are _any_ improvements in V
And for the more enlightened users, I believe it mostly comes down to DRM. Even many average users have some idea that DRM is overtly anti-consumer. One doesn't have to know specifically how DRM is implemented in Vista. Anyone who's had the displeasure of experiencing how it's been "working" in a current environment knows enough.
You do need to know when you're going to criticise its implications, as you have tried to do.
The DRM in Vista does _nothing_ unless you have DRM-encumbered media - and even if you *have* DRM-encumbered media, Vista doesn't do anything that other platforms (be they computer or standalone) will also do.
For "enlightened users", the DRM in Vista should be a complete and utter non-issue, because they should be able to figure out that it's either a) irrelevant or b) inevitable (depending on what they're trying to do).
The PC has traditionally been been associated with open architecture and freedom for the user to do with it whatever he or she wishes.
The PC has traditionally been cheap and pervasive. The "open architecture" you speak of is something the vast bulk of users are only vaguely aware of, if they are at all - and it's most certainly not something the PC was initially designed to have.
This is a very important concept to many people, regardless of what those dependent upon MS for their liveliehoods would like to believe.
It may be a very important concept to many people, but it sure as hell isn't an important concept to most of them. Most people's criteria for choosing a PC are a) does it run the software I want and b) can I afford it.
If an "open architecture" were as important as you say, Apple's sales would be going downhill, not improving (and OS X wouldn't be attracting anything close to the proportion of "advanced users" that it is).
Besides, Roman roads didn't only work with Roman chariots. Roman aqueducts didn't only dispense water finally through Roman-designed faucets. etc etc.
Last I checked, Windows still runs non-Microsoft software.
Aero is a visual upgrade; it should be an add-on, similar to Plus! was for Win98. Because it's not, IMO, a upgrade; it doesn't actually do anything to improve the ability to do anything.
Aero is vastly more than a "visual upgrade". Indeed, the "visual" part of Vista's new display system is the _least_ significant aspect.
Security & Safety: or "making sure you can't use the software and media you bought 3 years ago". a joke. Win2k with Tiny Personal Firewall 2.0.14 and maybe zone alarm is much more secure.
False.
But is it not obvious that this is not really an upgrade?
Vista is a _massive_ upgrade. It rivals Apple's update of NeXTSTEP to OS X in scope. The entire network and audio stacks have been rewritten. The display system is completely new and unrivalled in its capabilities. While the changes to "security" are largely in the UI aspect, the amount of work done behind the scenes to mitigate the impact on legacy software (eg: virtualising the Registry) is significant. The work done in the kernel and lower system levels regarding threading, locking, memory management and component dependency isolation is worth a whole x.0 update on its own. Improvements to the remote managability and deployment capabilities are also large.
Vista is, unquestionably, the biggest update to Windows (from both the technical and non-technical perspective) since Windows 3.1 -> Windows 95. That it's being met with relative disinterest is a much bigger compliment to Windows XP (and 2000) than it is criticism of Vista. Vista could have been released with half as much work done and _still_ have easily justified a completely new Windows release.
With every little thing that gets folded in to the system and thereby run by the draconian DRM/trusted computing system, you are eroding your own freedoms to watch what you want, create what you want, read what you want.
You have no idea how the DRM in Vista works.
"Upgrade" has the implication that something is being improved; my position is that Vista is a downgrade.
And it is wrong, as demonstrated by the list of improvements.
Hm, you seem to think Gutmann's posts and rebuttals are slashdot comments. I'm referring to the article A Cost Analysis of Windows Vista Content Protection. This is the work of a notable security expert, who has analysed and written a report on the AACS specification and its implementation in Vista. This is not some evangelist slashdot post, it is a serious paper, heavily based in fact by a noted security expert. It's become extremely popular.
I'm aware of who Gutman is, I've read the paper several times and it's far more alarmist FUD than fact (as is most of the commentary you have written). The rebuttal from Microsoft covers the most egrarious and inaccurate statements, but I'm not going to bother repeating them, because if you don't believe the developer of the Operating System then I - a random Slashdot poster - am sure as hell not going to change your mind.
(If one were to make all the worst-case assumptions about the potentially negative implications of open source code and write a paper about how bad Linux was because of them, then Slashdot readers would (rightfully, albeit for the wrong reasons) decry it as baseless FUD and tear its author to shreds. Yet, when someone does the same thing about Vista, they are cheered on as some sort of saint.)
huh? what, exactly, are the features of Vista that make it a "upgrade" from WinXP, or better yet, Win2k?
Here's a list to get you started.
So, I repeat, to the parent & everyone else who even begins to consider "upgrading" to Vista: Is there ANY feature of Vista that will improve my ability to do ANYTHING AT ALL I currently do on my DRM-free Win2k machine?
This is not a repeat, it is a different question. "What's new ?" is not the same as "what's new that benefits me ?"
Make no mistake: Monopoly isn't a feature of capitalism, it's a failure. Free markets do not work without government to do things like: provide courts to enforce contracts etc, and intervene when a market failure occurs, such as with monopoly.
Monopolies (well, a single monopoly) are the only logical conclusion of an unregulated free market - at least so long as scarcity exists (and if it didn't, well, there wouldn't be any need for capitalism ).
(Of course, you couldn't still call it a "free market" by the time that happened, but it would be inevitable.)
The big question is how does someone patent someone else's idea by accident after admitting they had copied it.
"Someone" doesn't. However, Microsoft is not "someone", they are a collections of tens of thousands of "someones". As anyone who has even a passing acquaintance with groups of people will be able to tell you, even the most organised ones frequently have instances of miscommunication and confusions.
The surprising thing is not that it happened, it's that it hasn't happened more frequently.
Please educate yourself on monopolies before trying to argue this subject. Monopolies are anti-capitalist. They allow a company to break the capitalist system by undermining the benefits it normally provides, which is why almost every country in the world restricts their action. The basic idea of the capitalist method is that you can get more innovation by appealing to people's greed, using competition. Maybe three companies all duplicate the same research, development, and production facilities. That's pretty wasteful. But, because all of them are motivated by the market (greed) to deliver the best product to customers, they make good decisions and in the end consumers get a choice among several superior products instead of one inferior one.
The irony being, of course, that the ultimate objective of any seller in a capitalist marketplace is to become a monopoly.
As a result, the most innovative product does not win the market and consumers get an inferior product, despite the fact that they are still making the correct decisions for their own self-interest.
By definition, if consumers are making the correct decisions for their own self-interest, capitalism is working. This holds true even when they are making decisions that you - or other "we know betters" like you - would not make.
Unless, of course, you assert there is some way of defining, objectively, what "the most innovative product" is (what's the capitalist word for God ?).
"The most innovative" (or "best", if you prefer) product never wins, because "best" is a subjective, relative metric. No one product can ever be the "best", because no product will ever be everything to everyone - ie: perfect.
(And even if such a product ever was invented, the concept of anti-trust would be compelled to make it illegal.)
If you disagree with me that is fine, but please do not bother to argue this unless you actually understand how monopolies and capitalism interact.
If Microsoft is a monopoly, why do they continually improve their products (in measurable, objective ways), especially with features the vast majority of their customers are unaware of ?
Yes, they slowed innovation to a crawl and turned one of the fastest growing and most promising chunks of that market into a slow moving market that has held back related fields of study for a decade. We're easily 5 or 10 years behind where we could be if our anti-trust law was enforced.
Why don't you just say we're 73.85493% worse off ? It would be about as valid.
What exactly HAS Microsoft done for the tech industry?
Facilitated the commoditisation of computers.
(You might (well, will) argue that "anyone could have done it". I will not disagree, however, the historical fact remains that it was Microsoft who did.)
Tell me, just what has microsoft done for the computer industry?
Indeed. What *have* the Romans ever done for us ?
I studied MS extensively for a political science thesis. My conclusion is that MS is evil.
So if you call Microsoft "evil", what do you call, say, Tobacco corporations ? You know, to communicate a bit of scale ? Reallyreallyreallyreallyreallyreallyreallyreallyre allyreallyreallyEvil ?