I realize lots of people thought the Dune miniseries (well, what they've seen of it so far, anyway) sucked. I'm going to ignore that, brushing it off mainly as the idle whining of fanboys who can never be truly satisfied. The opening episode last night was spectacular, and I have high hopes for the following two.
Now, with that said, here's the meat of the comment. Finally, somebody has realized that Dune is not the whole story. In fact, Dune is not really even all that integral to the story of Dune, except for being interesting history and backstory for the real meat of what Dune is -- the story about Leto II, and the enormous sacrifices he made on behalf of humanity, accepting a fate some would say is worse than death to allow his people to continue to live.
My only hope now is that they see this through. No stopping at "Children". Take it all the way through Chapterhouse, and tell the full story. Then this fanboy will be truly impressed.
Too late - the legacy of Dune has already been blasphemed by a bunch of books with names like "God Emperor of Dune" and "Chapterhouse Dune."
Did you actually read through all six books, or what? The books dealing with Paul are simply backstory. They're interesting history that you'll need to understand the meat of the story, which is all about Leto II. Chapterhouse is pretty much a wrapping-up of everything. Oh, and a small nit that needs picking -- the quote was "blashpheme the legacy of Frank". Frank wrote the books you claim blaspheme the legacy of Dune. Two different concepts, there.
Anyway, if you must bash on a Dune book or two, complain about the new Dune: House * books. While not being bad books in and of themselves, compared to Frank's work, they're crap (oh, and they have some continuity problems, as well.
you'd have
ross.com, ross.ross.com, cool.ross.com, yeti.ross.com, irc.ross.com, sil.ross.com, silly.ross.com, and so on.
IMHO, one of the major problems that eat up so many domain names is people and businesses thinking they need a foo.com for everything, when bar.foo.com and xyzzy.foo.com would work just as well as bar.com and xyzzy.com. Movie labels are awful about this. Do we really need somemovie.com/.net/.org? Why can't we have somemovie.sony.com? or someothermovie.newline.com, and so on?
Other markets are just as bad, however. Why do we need a playstation.com, when playstation.sony.com would work just as well? And don't even get me started on crap like all the planet*.com sites. How about quake.planet.com rather than planetquake.com?
As much as I dislike themes (not that themability is bad, just that most themes are poorly-designed, pixmap-heavy, vomit-inducing eyesores), themes.org got this right. You want blackbox themes? Try blackbox.themes.org. You want themes for IRC clients? How about irc.themes.org? And so on.
If only dot-commers would pull their heads out their asses just long enough to see what the hell is going on, we might not have any need for new TLDs (well, yet, anyway).
While it may not be the same thing, you can download the Speech API from MS Research's web site (I don't feel like putting up a URL -- trust me, it's simple to find), which includes sample tools for voice control of the OS, voice dictation, and even screen reading. For the small amount I've used it, it works beautifully, though you'll certainly want a head-mount microphone. If I were blind, this would be a set of extremely useful tools. As it is, they're just toys for me to play with.
Still, it's fun to give your computers orders vocally and watch stuff happen.
That's bull. Yes, movies are 24fps, and NTSC video is 30fps (interleaved, for an effective 60Hz), and PAL video is 25fps. But it's very naive to just look at the numbers. What you're missing is important things like motion blur, which go a long way towards making motions seem smooth.
However, at 24fps and *with* motion blurring, movies still have a hell of a time doing a comfortable pan. Seriously, try paying attention next time you go to a movie. You'll never see a long pan, or if you do, you'll notice quite quickly that it's rather painful to watch.
Now, here's the kicker -- Most video games don't do any form of motion blur at all (some do a minor amount, like Homeworld, but the overhead neccessary to generate motion blur is generally more than it would be to just pump up the frame rate). So, without motion blur, the human eye needs a minimum of 24fps to see passably acceptable motion. However, that motion only really becomes comfortable around 60fps, or so.
As to the insignifigance of a sub-30fps drop, I urge you to actually play a game such as Quake3, where on-screen complexity and number of moving objects can get to large levels (play on a huge deathmatch map, with lots of players). You'll certainly notice that sudden drop in fps as you're aiming your railgun.
People have been blowing the "The eye can only see 24fps" bull for years. It's about time someone set them straight, even if it's only to say, "Oh, wait. It's not 24fps the eye can see. It's 72fps. Yeah, that's it." Of course, down the road, that'll change to, "Wait, it's not 72fps, but 100fps," and then "Wait, not 100fps, but 200fps," and so on. Seems to me the only reason for this is to make people not feel so bad about not having the latest and greatest video card out there.
Remember that Microsoft also doesn't have the Japanese game designers that have made most of the games you'd commonly associate with console machines, but rather has US-based developers who are more often than not more familiar with PC games.
Izzat so, huh? Look at this link for a list of XBox developers, current as of 9/20. Just scanning the list shows a number of big Japanese names, including such hard-hitters as Capcom and Konami. Squaresoft is obviously missing, but that's because they focus on a single platform at a time and have already invested quite a bit of time in the PS2 (FFX and XI will both be PS2 titles -- after that, we'll see what happens). However, the Dreamcast, and the Genesis before it, are both proof positive that a console can survive without SquareSoft (and in the case of the Genesis, even have some good console RPGs, such as the Phantasy Star series).
Did I read that right? No policies against portscanning? Gah! Ban this ISP now! Any ISP that will not specifically disallow portscanning by its users must be blacklisted until they change said policy.
Then again, if the "no policies against it" refered to "allowing servers", please disregard the above paragraph (but in the future, try to be more-clear in your statements).
- get cable/dsl and set it up so that you have a static IP (even though cable/dsl uses dhcp you can generally hardcode your IP).
Using a cable connection for running a server is generally a Bad Idea (tm), considering
you typically must use DHCP, which means you won't have a static IP (no matter how satic that IP appears, it can go away quite easily at any time), and thus makes it hard to handle DNS for your domain and point it to an IP, and
Most cable providers restrict the running of services on their networks, unless you upgrade to a business plan. You certainly don't want your mail server to disappear some day simply because the administration caught on that you were Breaking their AUP.
- register a domain. Beg borrow or steal a dns server to use as the primary (gandi.net offers free dns hosting I think when you reg a domain).
Okay, now that's just plain wrong. You don't own the IP you're using -- your ISP does. Therefore, it's theirs to do with as they please, not yours. That means pointing domains to that IP, among other things. As well, you won't be able to do reverse DNS for your IP pointing to your domain unless you have your ISP's blessing. Try talking to your ISP before you go and do something silly like registering a domain to an IP owned by them. I think you'll find that 90% of all ISPs are quite willing to help out, and will typically even offer DNS services for free.
Please, people, try thinking before you follow advice like this.
Oh, bah. You bleeding-heart, tree-hugging hippy liberals are mainly what's wrong with the US these days. It's not the government's responsibility to provide for the welfare of its citizens, past providing basic rights (speech, religion, due process -- you know, the important things, not the "right" to have a full belly) and protecting those rights.
Seriously, rather than providing more welfare services, and thus raising taxes and lowering spendable income, why not reduce taxes so we can afford to feed our selves? It's a win-win situation -- citizens have more of their own money to spend on food, clothing, shelter, and the government gets to shed the responsibility of providing for the crackwhore baby factories.
See, this all really depends on how you define the IT market. IT = "Information Technology". To me, this means server admins, net admins, web devs, and others. Or in other words, not the true thrust of most CS curricula. Skilled programmers (yes, even those fresh out of college, given you have a decent set of work-related skills, hopefully honed through an internship or two, or a co-op) should have no problems finding a software development position these days. However, the CS student that is majoring in CS solely *because* the market appears to be insatiable (or did so a year or more ago) is going to have a tougher time of it, and rightly so.
So, unless you're one of those rare (in my experience) CS majors who aspire to nothing more than being a sysadmin, nettech, or other various support roles, your job outlook is still quite positive. (Note to sysadmins/nettechs/others: I'm not saying that your jobs are easy or useless or whatever other connotation you derive from the above, but simply that they typically don't require a Computer Science degree).
Then again, there's something to be said about continuing on through grad school to get a Master's or even a PhD. I'll probably do that myself, one of these days, but my BS in CS is currently "good enough".
Having been a computer instructor for a year and a half, I have found that there is good news when an authority screws up publicly. First, it shows than no one is an authority on everything.
Then again, Slashdot has never been an authority on anything...
Mr. McKenzie is a third-party economist with no interest in Microsoft other than economically (as in, the book was not funded by MS nor MS supporters, and he is not on their payroll in even the subtlest of ways). Yet, the books describes many contradictions in the DoJ's case, as well as describing the economic situation the software industry is in today. A very interesting read, as long as you keep a half-open mind.
For those of you that just brush this off as more Pro-MS FUD, I feel sorry for you. The book is a jewel of economic investigation, and sheds much needed light on the entire antitrust process, as well as the actual goals of the DoJ in this case (hint: Their goal isn't helping consumers).
A very good read. Too bad I didn't post this earlier.
For the record, Microsoft's "competitors" (ie, those who have brought this case against them, namely Sun, Oracle, and Netscape/AOL) also make political contributions, and in larger sums than Microsoft has. Microsoft has really been somewhat of a sleeping giant, in the political realm. They've had no real need to "bribe" officials before, and so they haven't let lose with any massive financial compensation for policy makers. You can bet that if MS has learned one thing from this whole ordeal, it's that to get things to go your way in Washington, DC, you simply need to spend more money than the other guys. And money is something MS has in abundance.
Not to mention the fact that that one lower court judge already has a grudge against Microsoft, having ruled against ehm prior to this case, yet losing to that ruling out to the appellate court. Kinda makes you step back and wonder whether or not Judge Jackson was the best judge to handle this case, especially in light of his other "transgressions" towards the end of his leg of the case...
And then of course comes the *other* stupidity they always engage in- denying things which are confirmed facts.
Uh... confirmed "facts"? By whom? I take it you didn't read Mainsoft's original press release too closely, did you? There was absolutely no confirmation in their that they were porting any Microsoft applications to Linux. Oh, sure, Mainsoft's win32 layer has a Linux target, but that doesn't mean Microsoft is going to port to anything more than Solaris and HP-UX, as they've done before.
Of course, if you can present the sources you used to "confirm" this "fact", perhaps it might lend a bit of credibility to your arguments.
I think you've totally missed the point of what Mainsoft does. They're not about porting the applications, so much as they are about porting the API. Certainly, if a decent win32 layer existed for UNIX (ignoring WINE, since it's always going to be at a disadvantage), the porting of applications would be as simple as you stated. However, before that can happen, you need to port the API. What's the easiest way to port an API? Why, to have the source code, of course.
Note that porting the API is essentially a one-time thing, but since the win32 API is huge, it can be a time-consuming process.
Just for clarification, could you specify what version of Windows you're speaking of? (For the record, when I say "Windows" in this comment, aside from the prior reference, I am refering to mainly Windows 2000, and possibly NT. Never Windows 95 or 98). NT and 2000, while not strictly multi-user in the UNIX sense of the phrase, do allow for multiple users and for a super user account (Administrator). Along with the super user account, NT and Windows 2000 have ACLs that have a much finer granularity than UNIX's standard permissions (yes, I know some unices have ACLs of a finer granularity than owner/group/world, read/write/execute. I'm ignoring those and many meaning "Linux" by my use of the word "UNIX").
That right there addresses numbers 2 and 3 in your "What's bad about Windows" list. As for number one, BSODs are typically caused by bad drivers, or bad third-party software (ie, stuff not under MS's control). I've had my own share of segfaults and such in various unices. This is not strictly a Windows problem. Granted, it seems empirically that Windows is more "unstable" than Linux, that's mainly due to the lower bar for entry -- Windows is easy to setup, but difficult to master. However, the "easy" part lures Joe Q. Randomuser, MCSE, to install NT. Of course, since he doesn't know what he's doing, and the system is too "easy" to require hiring an admin, he will generally fuck something up. Thus, the standard equation: stability = 90% administration, 10% hardware and OS.
Addressing number 4 in the list of "Bad" is simple -- Windows 2000 has a tab on the Task Manager to show all processes, with customizable columns. It can potentially show even more info than ps, and an "End Process" button is provided if you want to kill the processes. As well, the standard "Applications" tab still exists, but now it really only shows applications (as in, stuff with windows open).
For number 5, who said you have to write Win32 code? There are wrappers out there that make it nearly trivial to write windows code (sure, MFC is bloated, but there are other options). At the same time, win32 code is much cleaner and more sane than, for instance, X code. In my experience, writing win32 code is typically not much harder than writing gtk code, and (in my opinion) is more rewarding.
Anyway, I didn't reply to this to counter each "Bad" item you listed. On to other things.
I suggest you look into cLIeNUX (http://www.clienux.com) for a distribution of linux that is doing some very neat stuff for client users (what I assume you're wanting when you ask for a single-user version of unix). Aside from the fact that you're pretty much root, cLIeNUX has a new filesystem hierarchy (Dotted Standard Filesystem Hierarchy (DSFH), which "hides" directories using the dot notation (so/sbin becomes/.sbi, for example), and then links a more verbose name to those paths strictly for users (ie,/bin would become/.bi, and might be linked to/Applications). Makes it easy for internationalization), and an excellent help system.
As for the problems with the GUI (X), the only sane solution is to get rid of it! Seriously, X is what's holding linux (and other unices) back, in the gui realm. Why is it that Windows and Mac OS tend to have applications with a consistent look&feel? It's because there aren't several hundred window managers (I'm not making that number up). There is one system that handles rendering, and though you can change shells with Windows, the widgets and such all look pretty much the same. Granted, the current designs for Windows, Mac OS, BeOS, etc don't allow for X's network transparency, but that wouldn't really be too hard to do (look at Terminal Server). What X needs is a standard widget set (none of this Xt, athena, motif, tk, gtk+, qt, etc crap), at the very least. on the widget level, you can enforce look&feel, and keyboard accelerators. Of course, that's never going to happen.
my > $0.02
Re:Scary obviously means 2 YEARS (or 1.5 or 3)
on
New Doom Details
·
· Score: 1
no no no. "Two Weeks" is the amount of time Quake2 took. Yes, I realize Quake2 took longer than that, but that was the official Id line. "When will Quake2 be released?" "Two weeks". It's the same thing as Quake3 Arena's "When it's done". Apparently now, it's going to be "Two weeks (or 1.5 or 3)". This is simply Id's long standing tradition of taking as much time as they need to build their technology demos.
Finally! Somebody on Slashdot with a brain! Even ignoring the meat of the argument (which is excellent, and sums up the entire case quite nicely), at least somebody besides myself has enough presence of mind to notice that splitting up Microsoft will not solve the "problem" of MS having a monopoly. What I don't understand is how relatively sane people can in one sentence say that Microsoft has a monopoly on the desktop market, and then in the next advocate splitting the company up as a "solution" to the monopoly "problem". Every break-up suggestion I've seen looks something like -- three companies, focusing on operating systems, applications, and internet presence, respectively. How does that at all solve the "problem" of MS having an OS monopoly of the desktop market? It doesn't.
From my point of view, there's really only one way to "break" Microsoft's supposed monopoly, and that's by encouraging competitor operating systems (and no, Linux is still not a suitable replacement for Windows on the desktop. As much as I love linux, it's just not well-suited for desktop work, yet). At worst, this may require some government intervention to keep MS from destroying or assimilating the competitors, but I think that simply restricting MS's licensinge policies as stated above would be enough to give competitors the leverage they need.
Please, people. Think before you go around advocating something like breaking up Microsoft just because you "hate" the company, especially when such a solution does nothing to solve the problems you're all so enflamed about.
PS: The statement "nobody here likes Microsoft" is insanely wrong. It's just that most people are afraid of expressing such a view because the 1337 bandwagon-jumpers will flame them to death with shit like "L1NUX R00LZ, D00D! MIRCO$OFT SUX0RZZ!!".
Just a few comments. First off, Microsoft, Apple, and other companies with a vested interest in UI design have sunk many dollars into R&D and testing for their UI designs, and as much as you may not like them, they *do* work for the majority of users. In fact, Microsoft (through Microsoft Press) has published several books on UI design and style. UI design is not something that is taken lightly at MSFT.
Second, the component object model (COM) used by Windows makes sense -- it's a language-independent, binary-compatible specification for writing software objects that can be reused by any application that can see the interface (and by using type libraries, which are nearly always embedded in COM objects, it's possible to get the interfaces you need without having even a header file). From what I gathered from the article, that's where the major "borrowing" has occured -- in the form of GNOME's dependence upon CORBA. The ability to use other developers' objects in pretty much any, even if no source is provided, is a real God-send, and is an idea that linux (and unix in general) can benefit from.
Finally, to address the issue of "real" transparency -- if it's used properly, it can make quite a bit of difference in the user experience, from subtle things such as properly-translucent dragging icons to more garish displays like fading menus, translucent windows, and more. Personally, I think it's a Good Thing (tm) to have alpha-blending transparency in a modern UI, since it can easily enhance the effectiveness of the UI and it's rather trivial to do these days (making the assumption that everybody has video cards that can handle alpha-blending -- and those that don't should upgrade or turn off the effect).
Of course, since I've not done any research in UI design, I'm probably not qualified to comment. Then again, I'd bet dollars to donuts that neither are you (hey, that's the beauty of forums such as this -- unqualified people can comment to their heart's content without any restrictions:).
Slashdot Mirror
Thufir Hawat, the old guy often seen wearing purple and a silly hat.
I realize lots of people thought the Dune miniseries (well, what they've seen of it so far, anyway) sucked. I'm going to ignore that, brushing it off mainly as the idle whining of fanboys who can never be truly satisfied. The opening episode last night was spectacular, and I have high hopes for the following two.
Now, with that said, here's the meat of the comment. Finally, somebody has realized that Dune is not the whole story. In fact, Dune is not really even all that integral to the story of Dune, except for being interesting history and backstory for the real meat of what Dune is -- the story about Leto II, and the enormous sacrifices he made on behalf of humanity, accepting a fate some would say is worse than death to allow his people to continue to live.
My only hope now is that they see this through. No stopping at "Children". Take it all the way through Chapterhouse, and tell the full story. Then this fanboy will be truly impressed.
Too late - the legacy of Dune has already been blasphemed by a bunch of books with names like "God Emperor of Dune" and "Chapterhouse Dune."
Did you actually read through all six books, or what? The books dealing with Paul are simply backstory. They're interesting history that you'll need to understand the meat of the story, which is all about Leto II. Chapterhouse is pretty much a wrapping-up of everything. Oh, and a small nit that needs picking -- the quote was "blashpheme the legacy of Frank". Frank wrote the books you claim blaspheme the legacy of Dune. Two different concepts, there.
Anyway, if you must bash on a Dune book or two, complain about the new Dune: House * books. While not being bad books in and of themselves, compared to Frank's work, they're crap (oh, and they have some continuity problems, as well.
And how is that much different than using subdomains? Rather than
ross.com ross.ross ross.cool ross.yeti ross.irc ross.sil ross.silly
you'd have ross.com, ross.ross.com, cool.ross.com, yeti.ross.com, irc.ross.com, sil.ross.com, silly.ross.com, and so on.
IMHO, one of the major problems that eat up so many domain names is people and businesses thinking they need a foo.com for everything, when bar.foo.com and xyzzy.foo.com would work just as well as bar.com and xyzzy.com. Movie labels are awful about this. Do we really need somemovie.com/.net/.org? Why can't we have somemovie.sony.com? or someothermovie.newline.com, and so on?
Other markets are just as bad, however. Why do we need a playstation.com, when playstation.sony.com would work just as well? And don't even get me started on crap like all the planet*.com sites. How about quake.planet.com rather than planetquake.com?
As much as I dislike themes (not that themability is bad, just that most themes are poorly-designed, pixmap-heavy, vomit-inducing eyesores), themes.org got this right. You want blackbox themes? Try blackbox.themes.org. You want themes for IRC clients? How about irc.themes.org? And so on.
If only dot-commers would pull their heads out their asses just long enough to see what the hell is going on, we might not have any need for new TLDs (well, yet, anyway).
Just my $0.02.
While it may not be the same thing, you can download the Speech API from MS Research's web site (I don't feel like putting up a URL -- trust me, it's simple to find), which includes sample tools for voice control of the OS, voice dictation, and even screen reading. For the small amount I've used it, it works beautifully, though you'll certainly want a head-mount microphone. If I were blind, this would be a set of extremely useful tools. As it is, they're just toys for me to play with.
Still, it's fun to give your computers orders vocally and watch stuff happen.
That's bull. Yes, movies are 24fps, and NTSC video is 30fps (interleaved, for an effective 60Hz), and PAL video is 25fps. But it's very naive to just look at the numbers. What you're missing is important things like motion blur, which go a long way towards making motions seem smooth.
However, at 24fps and *with* motion blurring, movies still have a hell of a time doing a comfortable pan. Seriously, try paying attention next time you go to a movie. You'll never see a long pan, or if you do, you'll notice quite quickly that it's rather painful to watch.
Now, here's the kicker -- Most video games don't do any form of motion blur at all (some do a minor amount, like Homeworld, but the overhead neccessary to generate motion blur is generally more than it would be to just pump up the frame rate). So, without motion blur, the human eye needs a minimum of 24fps to see passably acceptable motion. However, that motion only really becomes comfortable around 60fps, or so.
As to the insignifigance of a sub-30fps drop, I urge you to actually play a game such as Quake3, where on-screen complexity and number of moving objects can get to large levels (play on a huge deathmatch map, with lots of players). You'll certainly notice that sudden drop in fps as you're aiming your railgun.
People have been blowing the "The eye can only see 24fps" bull for years. It's about time someone set them straight, even if it's only to say, "Oh, wait. It's not 24fps the eye can see. It's 72fps. Yeah, that's it." Of course, down the road, that'll change to, "Wait, it's not 72fps, but 100fps," and then "Wait, not 100fps, but 200fps," and so on. Seems to me the only reason for this is to make people not feel so bad about not having the latest and greatest video card out there.
Remember that Microsoft also doesn't have the Japanese game designers that have made most of the games you'd commonly associate with console machines, but rather has US-based developers who are more often than not more familiar with PC games.
Izzat so, huh? Look at this link for a list of XBox developers, current as of 9/20. Just scanning the list shows a number of big Japanese names, including such hard-hitters as Capcom and Konami. Squaresoft is obviously missing, but that's because they focus on a single platform at a time and have already invested quite a bit of time in the PS2 (FFX and XI will both be PS2 titles -- after that, we'll see what happens). However, the Dreamcast, and the Genesis before it, are both proof positive that a console can survive without SquareSoft (and in the case of the Genesis, even have some good console RPGs, such as the Phantasy Star series).
Did I read that right? No policies against portscanning? Gah! Ban this ISP now! Any ISP that will not specifically disallow portscanning by its users must be blacklisted until they change said policy.
Then again, if the "no policies against it" refered to "allowing servers", please disregard the above paragraph (but in the future, try to be more-clear in your statements).
Thank you
- get cable/dsl and set it up so that you have a static IP (even though cable/dsl uses dhcp you can generally hardcode your IP).
Using a cable connection for running a server is generally a Bad Idea (tm), considering
- register a domain. Beg borrow or steal a dns server to use as the primary (gandi.net offers free dns hosting I think when you reg a domain).
Okay, now that's just plain wrong. You don't own the IP you're using -- your ISP does. Therefore, it's theirs to do with as they please, not yours. That means pointing domains to that IP, among other things. As well, you won't be able to do reverse DNS for your IP pointing to your domain unless you have your ISP's blessing. Try talking to your ISP before you go and do something silly like registering a domain to an IP owned by them. I think you'll find that 90% of all ISPs are quite willing to help out, and will typically even offer DNS services for free.
Please, people, try thinking before you follow advice like this.
Oh, bah. You bleeding-heart, tree-hugging hippy liberals are mainly what's wrong with the US these days. It's not the government's responsibility to provide for the welfare of its citizens, past providing basic rights (speech, religion, due process -- you know, the important things, not the "right" to have a full belly) and protecting those rights.
Seriously, rather than providing more welfare services, and thus raising taxes and lowering spendable income, why not reduce taxes so we can afford to feed our selves? It's a win-win situation -- citizens have more of their own money to spend on food, clothing, shelter, and the government gets to shed the responsibility of providing for the crackwhore baby factories.
Then again, that's just one man's opinion.
See, this all really depends on how you define the IT market. IT = "Information Technology". To me, this means server admins, net admins, web devs, and others. Or in other words, not the true thrust of most CS curricula. Skilled programmers (yes, even those fresh out of college, given you have a decent set of work-related skills, hopefully honed through an internship or two, or a co-op) should have no problems finding a software development position these days. However, the CS student that is majoring in CS solely *because* the market appears to be insatiable (or did so a year or more ago) is going to have a tougher time of it, and rightly so.
So, unless you're one of those rare (in my experience) CS majors who aspire to nothing more than being a sysadmin, nettech, or other various support roles, your job outlook is still quite positive. (Note to sysadmins/nettechs/others: I'm not saying that your jobs are easy or useless or whatever other connotation you derive from the above, but simply that they typically don't require a Computer Science degree).
Then again, there's something to be said about continuing on through grad school to get a Master's or even a PhD. I'll probably do that myself, one of these days, but my BS in CS is currently "good enough".
Erm, Microsoft is almost 40,000 people, worldwide.
Back on-topic: As has already been pointed out, the editorial was aimed at the Ma&Pa shops, not at large OEMs like Dell or Compaq or such.
banner ads? What are those? Oh, you must mean that blank rectangle above the actual content. Gotta love squid and adzap!
Having been a computer instructor for a year and a half, I have found that there is good news when an authority screws up publicly. First, it shows than no one is an authority on everything.
...
Then again, Slashdot has never been an authority on anything
For those that care (which seems to be precious few people, here, but oh well), a *much* better piece of literature that is not neccessarily Pro-Microsoft, but definitely Anti-Antitrust, is Trust on Trial: How the Microsoft Case is Reframing the Rules of Competition by Richard B. McKenzie.
Mr. McKenzie is a third-party economist with no interest in Microsoft other than economically (as in, the book was not funded by MS nor MS supporters, and he is not on their payroll in even the subtlest of ways). Yet, the books describes many contradictions in the DoJ's case, as well as describing the economic situation the software industry is in today. A very interesting read, as long as you keep a half-open mind.
For those of you that just brush this off as more Pro-MS FUD, I feel sorry for you. The book is a jewel of economic investigation, and sheds much needed light on the entire antitrust process, as well as the actual goals of the DoJ in this case (hint: Their goal isn't helping consumers).
A very good read. Too bad I didn't post this earlier.
For the record, Microsoft's "competitors" (ie, those who have brought this case against them, namely Sun, Oracle, and Netscape/AOL) also make political contributions, and in larger sums than Microsoft has. Microsoft has really been somewhat of a sleeping giant, in the political realm. They've had no real need to "bribe" officials before, and so they haven't let lose with any massive financial compensation for policy makers. You can bet that if MS has learned one thing from this whole ordeal, it's that to get things to go your way in Washington, DC, you simply need to spend more money than the other guys. And money is something MS has in abundance.
...
Draw your own conclusions
Not to mention the fact that that one lower court judge already has a grudge against Microsoft, having ruled against ehm prior to this case, yet losing to that ruling out to the appellate court. Kinda makes you step back and wonder whether or not Judge Jackson was the best judge to handle this case, especially in light of his other "transgressions" towards the end of his leg of the case ...
Hrm. Interesting thought ...
The Microsoft main campus and the Nintendo main campus are within a mile of each other.
Food for thought
And then of course comes the *other* stupidity they always engage in- denying things which are confirmed facts.
... confirmed "facts"? By whom? I take it you didn't read Mainsoft's original press release too closely, did you? There was absolutely no confirmation in their that they were porting any Microsoft applications to Linux. Oh, sure, Mainsoft's win32 layer has a Linux target, but that doesn't mean Microsoft is going to port to anything more than Solaris and HP-UX, as they've done before.
Uh
Of course, if you can present the sources you used to "confirm" this "fact", perhaps it might lend a bit of credibility to your arguments.
I think you've totally missed the point of what Mainsoft does. They're not about porting the applications, so much as they are about porting the API. Certainly, if a decent win32 layer existed for UNIX (ignoring WINE, since it's always going to be at a disadvantage), the porting of applications would be as simple as you stated. However, before that can happen, you need to port the API. What's the easiest way to port an API? Why, to have the source code, of course.
Note that porting the API is essentially a one-time thing, but since the win32 API is huge, it can be a time-consuming process.
Just for clarification, could you specify what version of Windows you're speaking of? (For the record, when I say "Windows" in this comment, aside from the prior reference, I am refering to mainly Windows 2000, and possibly NT. Never Windows 95 or 98). NT and 2000, while not strictly multi-user in the UNIX sense of the phrase, do allow for multiple users and for a super user account (Administrator). Along with the super user account, NT and Windows 2000 have ACLs that have a much finer granularity than UNIX's standard permissions (yes, I know some unices have ACLs of a finer granularity than owner/group/world, read/write/execute. I'm ignoring those and many meaning "Linux" by my use of the word "UNIX").
/sbin becomes /.sbi, for example), and then links a more verbose name to those paths strictly for users (ie, /bin would become /.bi, and might be linked to /Applications). Makes it easy for internationalization), and an excellent help system.
That right there addresses numbers 2 and 3 in your "What's bad about Windows" list. As for number one, BSODs are typically caused by bad drivers, or bad third-party software (ie, stuff not under MS's control). I've had my own share of segfaults and such in various unices. This is not strictly a Windows problem. Granted, it seems empirically that Windows is more "unstable" than Linux, that's mainly due to the lower bar for entry -- Windows is easy to setup, but difficult to master. However, the "easy" part lures Joe Q. Randomuser, MCSE, to install NT. Of course, since he doesn't know what he's doing, and the system is too "easy" to require hiring an admin, he will generally fuck something up. Thus, the standard equation: stability = 90% administration, 10% hardware and OS.
Addressing number 4 in the list of "Bad" is simple -- Windows 2000 has a tab on the Task Manager to show all processes, with customizable columns. It can potentially show even more info than ps, and an "End Process" button is provided if you want to kill the processes. As well, the standard "Applications" tab still exists, but now it really only shows applications (as in, stuff with windows open).
For number 5, who said you have to write Win32 code? There are wrappers out there that make it nearly trivial to write windows code (sure, MFC is bloated, but there are other options). At the same time, win32 code is much cleaner and more sane than, for instance, X code. In my experience, writing win32 code is typically not much harder than writing gtk code, and (in my opinion) is more rewarding.
Anyway, I didn't reply to this to counter each "Bad" item you listed. On to other things.
I suggest you look into cLIeNUX (http://www.clienux.com) for a distribution of linux that is doing some very neat stuff for client users (what I assume you're wanting when you ask for a single-user version of unix). Aside from the fact that you're pretty much root, cLIeNUX has a new filesystem hierarchy (Dotted Standard Filesystem Hierarchy (DSFH), which "hides" directories using the dot notation (so
As for the problems with the GUI (X), the only sane solution is to get rid of it! Seriously, X is what's holding linux (and other unices) back, in the gui realm. Why is it that Windows and Mac OS tend to have applications with a consistent look&feel? It's because there aren't several hundred window managers (I'm not making that number up). There is one system that handles rendering, and though you can change shells with Windows, the widgets and such all look pretty much the same. Granted, the current designs for Windows, Mac OS, BeOS, etc don't allow for X's network transparency, but that wouldn't really be too hard to do (look at Terminal Server). What X needs is a standard widget set (none of this Xt, athena, motif, tk, gtk+, qt, etc crap), at the very least. on the widget level, you can enforce look&feel, and keyboard accelerators. Of course, that's never going to happen.
my > $0.02
no no no. "Two Weeks" is the amount of time Quake2 took. Yes, I realize Quake2 took longer than that, but that was the official Id line. "When will Quake2 be released?" "Two weeks". It's the same thing as Quake3 Arena's "When it's done". Apparently now, it's going to be "Two weeks (or 1.5 or 3)". This is simply Id's long standing tradition of taking as much time as they need to build their technology demos.
Finally! Somebody on Slashdot with a brain! Even ignoring the meat of the argument (which is excellent, and sums up the entire case quite nicely), at least somebody besides myself has enough presence of mind to notice that splitting up Microsoft will not solve the "problem" of MS having a monopoly. What I don't understand is how relatively sane people can in one sentence say that Microsoft has a monopoly on the desktop market, and then in the next advocate splitting the company up as a "solution" to the monopoly "problem". Every break-up suggestion I've seen looks something like -- three companies, focusing on operating systems, applications, and internet presence, respectively. How does that at all solve the "problem" of MS having an OS monopoly of the desktop market? It doesn't.
From my point of view, there's really only one way to "break" Microsoft's supposed monopoly, and that's by encouraging competitor operating systems (and no, Linux is still not a suitable replacement for Windows on the desktop. As much as I love linux, it's just not well-suited for desktop work, yet). At worst, this may require some government intervention to keep MS from destroying or assimilating the competitors, but I think that simply restricting MS's licensinge policies as stated above would be enough to give competitors the leverage they need.
Please, people. Think before you go around advocating something like breaking up Microsoft just because you "hate" the company, especially when such a solution does nothing to solve the problems you're all so enflamed about.
PS: The statement "nobody here likes Microsoft" is insanely wrong. It's just that most people are afraid of expressing such a view because the 1337 bandwagon-jumpers will flame them to death with shit like "L1NUX R00LZ, D00D! MIRCO$OFT SUX0RZZ!!".
Yeah, that's exactly what linux needs -- a series of subpar games using 4-year-old technology. Sounds like a winner to me!
Seriously, though. The Tomb Raider games aren't much better than the Deer Hunter games. Why burden Linux's still-fledgling gaming abilities with crap?
Just a few comments. First off, Microsoft, Apple, and other companies with a vested interest in UI design have sunk many dollars into R&D and testing for their UI designs, and as much as you may not like them, they *do* work for the majority of users. In fact, Microsoft (through Microsoft Press) has published several books on UI design and style. UI design is not something that is taken lightly at MSFT.
:).
Second, the component object model (COM) used by Windows makes sense -- it's a language-independent, binary-compatible specification for writing software objects that can be reused by any application that can see the interface (and by using type libraries, which are nearly always embedded in COM objects, it's possible to get the interfaces you need without having even a header file). From what I gathered from the article, that's where the major "borrowing" has occured -- in the form of GNOME's dependence upon CORBA. The ability to use other developers' objects in pretty much any, even if no source is provided, is a real God-send, and is an idea that linux (and unix in general) can benefit from.
Finally, to address the issue of "real" transparency -- if it's used properly, it can make quite a bit of difference in the user experience, from subtle things such as properly-translucent dragging icons to more garish displays like fading menus, translucent windows, and more. Personally, I think it's a Good Thing (tm) to have alpha-blending transparency in a modern UI, since it can easily enhance the effectiveness of the UI and it's rather trivial to do these days (making the assumption that everybody has video cards that can handle alpha-blending -- and those that don't should upgrade or turn off the effect).
Of course, since I've not done any research in UI design, I'm probably not qualified to comment. Then again, I'd bet dollars to donuts that neither are you (hey, that's the beauty of forums such as this -- unqualified people can comment to their heart's content without any restrictions