Hee hee... cool link, and I think "The Dancing Bunny Problem" clearly needs to go into the lexicon along with PEBKAC.
But I don't think it's true that botnets, etc., are inevitable for purely social/psychological reasons. If they were, we'd have botnets that ran on MacOS and Linux. There are certain simple, easy design decisions that you can make in an OS that will have a positive effect on security. MS decided that (in Unix terms) users would always be logged in as root by default. That was a bad design decision. There are many other bad design decisions built into Windows. MS doesn't even deny that these were bad decisions; they're trying to fix some of them with UAC, for example.
My mother in law got her Windows box horribly infested with viruses, to the point where it couldn't boot. She's actually a pretty smart lady. I helped her install Linux (dry run while she was visiting us here in California, then she did it on her own in Buffalo, with help over the phone). It's possible that she would try to download and run the dancing bunny. The thing is, she would find it difficult to download and run the dancing bunny on Linux. Because of the design of the OS, she would get dialog boxes along the way saying, "Are you sure you want to do this?" or "Please enter your administrator password." She would probably have time to bail out. On Windows, effectively running as root, the OS wouldn't have guaranteed her any of these chances to back out.
Re "A New Internet? The Old One is Putting Us in Jeopardy," by John Markoff
(Week in Review, Feb. 15, 2009):
Mr. Markoff both misstates and overstates the security problems faced by the
Internet as currently designed.
He never uses the word "Windows," but the virus outbreaks he describes are
almost entirely a Windows phenomenon, and due to the poor design of
that operating system. Microsoft's apologists have been saying for years
that this was only because Windows' market share made it the more attractive
target. But Apple's share of the desktop market has skyrocketed recently to
15% without any outbreaks of viruses targeting the Macintosh. And Microsoft
has never commanded more than about half of the server market; the other
half runs open-source operating systems such as Linux (used by Google)
and FreeBSD (Yahoo), on which viruses are essentially unknown.
Markoff says it's hard to prove your identity on the internet, and proposes
government regulation as a solution. But many people have been proving their
identities for years now using proven technologies like public-key
cryptography. The U.S. government played a negative role in the development
of these technologies by attempting to regulate their distribution
through export-control regulations originally intended for munitions.
That article is a multi-page annoyance, the grammar is bad and we already have flash-aware filesystems like jffs2.
As far as I can tell from some quick googling and checking on Wikipedia, jffs2 isn't much of a competitor at this point, e.g., it's apparently not really usable on flash chips bigger than 512 Mb. Maybe UBIFS or LogFS? None of them seem to be really mature.
Aw, come on. Sure, there were two "it's" mistakes in the article, but they both occurred in one sentence. The entire second sentence was completely free of "it's" mistakes. That represents real progress for slashdot. Actually, I believe slashdot's Central Committee For the Eradication of "It's" Mistakes (also known as COMINITS) has set a goal of reducing the average rate of "it's" mistakes to no more than three per sentence by the end of the current five-year plan. That means that production of "it's"-mistake-free sentences is actually 783% ahead of last year's production!
The number 1 issue is of course, would people want Apple closed binaries/frameworks on their Linux/*BSD?
Compare with Wine. The typical reason to use Wine is that you prefer Linux, but you don't have any choice because the app you need/want to use is Windows-only.
Since MacOS X isn't a monopoly, that doesn't occur very much with MacOS X. Nobody's bank tells them they can only access their account using MacOS X.
Ths typical Mac user likes MacOS because all the software and hardware is beautifully integrated and consistent, and everything Just Works. The last thing on earth that type of person is going to do is switch to Linux, but then insist on trying to run some kludged-together version of a Mac application that wasn't really designed to run on Linux.
Okay why can't we have an open standard to sync data with mobile devices?
Because your mobile devices are proprietary systems, and the companies who sold them to you don't want use every possible piece of functionality as a revenue stream.
If you want to synchronize your files between various devices, using open-source software, try unison. It's free, it's open source, it's fast, and IMO it's of very high quality. I use it to sync two desktops, a server, and an ARM-based network appliance (NSLU2). The key is that none of these are locked down systems sold to you by a cell phone company.
I -love- truecrypt but I wouldn't suggest it for a whole enterprise without being able to answer the question "How do I recover the key to this workstation when the employee dies unexpectedly of a heart attack?".
TrueCrypt's FAQ discusses this. See the question beginning "We use TrueCrypt in a corporate..."
if you don't know how much your site licenses cost, then you aren't in a position to influence future software purchasing decisions.
The OP said "influence," not "dictate." Clearly the OP isn't the top IT honcho at this university, but he/she doesn't claim to be.
For someone who really was the top manager in charge of IT at a university, my advice would be to start laying the groundwork over the next few years via the hiring process. Every time you interview someone for an IT position, make sure that one of your criteria is that the person has broad problem-solving skills, broad interests, and at least some academic qualifications in actual CS, as opposed to someone who just has MS certs. The school where I teach is pretty much the kind of MS monoculture the OP describes, except that there are some pockets of resistance in the fine arts division where they still use macs, despite intense pressure to standardize on Windows. The huge problem is that a lot of the IT staff doesn't know anything about Linux or OSS, and has never used Linux or OSS. Some of them are curious and willing to learn new things. Others are not. It doesn't help that the ratio of staff to boxen is on the low side, which tends to make them resist any complication such as supporting another OS. In this kind of environment, hiring even one tech who uses Ubuntu at home could make a huge difference.
I wouldn't mind picking up a machine at a garage sale with those specs for $50, but I'm not about to buy a new machine with those specs.
Come on. The Memory's $8 and the SSd is $10, and there's the Atom processor and display to count. That's much more than $50.
I didn't say that it ought to be produced and sold new for $50. I just said its specs were obsolete, so there was no way I'd buy it new. When you buy new hardware, you want it to be hardware that's going to be usable for a long time before it becomes obsolete. I'm much more willing to buy obsolete hardware if it's used and therefore very cheap.
I bet the netbook market has their attention. I can walk into a Target, Best Buy, or Wal-Mart and purchase a sub $300 netbook loaded with Linux. That's damn near the cost of Vista Ultimate -- sans computer.
Well, yeah, but your analysis a little off in some ways.
I own one of the WalMart gPC machines, and it's worked fine. But you typically can't walk into WalMart and buy them. You pretty much have to order them online, and therefore very few people know about them.
Re Target, my wife happened to go to Target today, where she noticed that they had the Eee PC out on a retail display. The Windows and Linux versions were side by side, with the Linux one $50 cheaper. However, the hardware they're selling in Target is an extremely out of date configuration with only a 4 Gb hard disk and 512 Mb of ram. (According to the WP article, it's a configuration that is only being sold in Target.) I wouldn't mind picking up a machine at a garage sale with those specs for $50, but I'm not about to buy a new machine with those specs.
So realistically, you really can't walk in to Target or WalMart and buy a decent, modern netbook -- not quite yet.
And while it may be true that some versions of Windows cost several hundred dollars, the cost of Windows to the OEM for low-end machines is as low as $34, so your comparison is somewhat misleading.
There are several other goofy things about the summary.
For a variety of reasons, this move is almost certainly targeted at Ubuntu Linux's desktop success.
As you pointed out, there's no evidence that this is specifically about Ubuntu. The other goofy thing in this sentence is the reference to "desktop success," which makes it sound as though Ubuntu is already a successful competitor, and MS is responding to that. Now I use Ubuntu, love Ubuntu, and I think it's great that companies like Asus and HP are shipping machines with Ubuntu preinstalled, but as far as anyone can tell, Linux's share of the desktop is still stalled at about 1%. Asus and OLPC are actually no longer exclusively tied to Linux. I think it's much more reasonable to interpret this as a move to fight against a competitor that is currently not successful at all in any quantitative sense, but threatens to become successful (i.e., start growing beyond 1% market share) in the future.
With the Mac, not Linux, apparently eating into Microsoft's Windows market share, what is it about desktop Linux, and specifically Ubuntu, that has Microsoft spooked?
There's no evidence to back up the part about "specifically Ubuntu."
Reader christian.einfeldt notes Microsoft's acknowledgment of the FOSS threat to their business model within SEC filings, and suggests that this job posting could instead be about maintaining Internet Explorer's market share lead against Firefox.
Uh, except that that MS job announcement specifically refers to desktop Windows. They're clearly advertising for a position for someone to be a cheerleader for Windows versus Linux, to head off any hypothetical future erosion of their market share to Linux.
We continue to watch the evolution of open source software development and distribution, and continue to differentiate our products from competitive products including those based on open source software. We believe that Microsoft's share of server units grew modestly in fiscal 2004, while Linux distributions rose slightly faster on an absolute basis.
The SEC filing refers only to servers when it comes to competition from OSS. That's because the server market, not the desktop market, is where MS currently has to compete with Linux and BSD on relatively even terms. So the final paragraph of the summary brings up two points that are unrelated to each other (browser versus OS) and unrelated to the job position (which is about desktop, not server).
You're really discussing three separate things here: (1) autosave, (2) keeping all the old versions around forever by default, and (3) eliminating the ability to give someone else a file without also giving them access to all the old versions. You can have 1 without 2 and 3 (commonly implemented in word processors), or 1 and 2 without 3 (as in VMS).
Autosave is bad in the situation where the cat walks across your keyboard while you're making tea. Autosave is good in the situation where the power goes out unexpectedly.
In a version control system, it's useful to have ways for the user to tell the system that the project is in a sane state at a certain point (e.g., git commit), and also to have ways for the user to say that certain states of the project are especially good/complete/debugged (e.g., git tag). The good thing about explicit saves, as opposed to autosaves, is that they perform this function. On the other hand, the user may have other mechanisms available for performing this function. For instance, I use a file synchronizer (unison) to sync up a certain set of machines that mirror my files. Usually I do a sync when I'm at a good, sane stopping point in a certain task. Suppose I'm using a word-processor, and the cat walks across the keyboard while I'm making tea, causing my entire novel to be erased and then autosaved. No big deal. I just sync up with the mirror machine and grab the undamaged version. I think MacOS's Time Machine provides similar "oh shit" functionality.
When a user is used to explicit saves, it could be argued that autosaves are bad because they're inconsistent with the way the user's other apps work. On the other hand, we seem to be seeing a proliferation of systems these days that are essentially version control systems reimplemented in different ways -- built into file systems, built into applications, etc. E.g., if I'm editing file foo using emacs, I could potentially have three files: foo, foo~, and #foo#, which sort of represent three levels of version control. This is good in the sense that it's consistent with the usual explicit-save paradigm, but bad in the sense that it's inconsistent with the way other applications handle it.
I also bought some Fry's GQ boxes, but they Lindows installed.
Yeah, I think they switched from ThizLinux to Lindows at some point. Too bad Fry's doesn't sell them anymore. They really were, uh, great quality in my experience:-) Apparently Fry's were just getting way too many returns, and it wasn't profitable for them. Since my machines ran great year after year, and are all still running, I assume the returns weren't due to hardware problems, but were from people who didn't understand that they weren't getting a machine with Windows installed.
Ubuntu is simply the sane thing to put on a desktop machine these days, especially for users who may not already be familiar with Linux.
It was always really frustrating to me in the past to see hardware companies selling machines with Linux preinstalled, but with some crappy version of Linux that was bound to create a bad impression of Linux in general.
Back when Fry's was selling Great Quality boxes for as low as $180, I bought several of them. They had something called ThizLinux on them, which was apparently a distro that GQ created themselves. No documentation for ThizLinux came with the machine, and googling for ThizLinux turned up a Chinese-language web site with no English translation. The printed docs that came with the machines were actually 90% information on how to wipe ThizLinux off your hard disk and install Windows. The impression any user would get from this was probably that Linux was crap, and nobody really wanted it.
Same deal with the Everex gPC, which I reviewed a while back. This may be a little unfair, because what I bought from them was a beta of their gOS distro, and now they have a newer version out, but basically it sucked, and I very quickly decided to replace it with Ubuntu. IMO it was just foolish of Everex to put out their own distro. I think they were imagining that by making it look slick (and a lot like MacOS) they would attract users. But in reality it worked so poorly that I think they were shooting themselves in the foot.
Xfce is your friend.
I use Xubuntu. Plain, clear, simple and *fast*. 8.10 runs out of the box everything on my ThinkPad laptop including Bluetooth. Get it.
I teach physics labs in a room with 7 school-supplied Windows machines and 5 Linux boxes I got at garage sales, Good Will, etc. At various times I've had Gnome, KDE, and Xfce on various machines. As in the video, my students often don't realize that they're not running Windows. There can also be kind of a converse effect, however, where they won't even try to use a Linux box because they think it's going to be hard, or different, or not as good. In other words, you could have a machine that really had Windows on it, but if you put a post-it on the side saying "Linux," they'd say, "I tried that one, and it was too hard."
Xfce does indeed perform somewhat better than Gnome and KDE on low-end hardware, while still giving people a fairly familiar user interface based on a screen littered with little icons. Xubuntu is what I have on those machines, and it generally works fine. (The only problem I've had is that on machines with low-resolution monitors, sometimes part of an application's window is hidden behind the tray at the bottom of the screen, and I haven't been able to figure out how to reveal it. On KDE the same problem happens, but I think you can hold down the alt key while dragging the window.) I think different people have different tolerances for unresponsiveness. Gnome has always felt unacceptably slow to me on every machine I've ever tried it on.
Personally, I've never understood why people like icons. I prefer fluxbox, which is very fast, and doesn't make my computer screen look like someone took the laundry out of the dryer and threw it on the kitchen floor.
Except for the fact that you don't have to use C++ in order to use Qt.
The PerlQt interface hasn't been updated since 2003, so I think it's dead. The python bindings aren't free on Windows, which would dull my enthusiasm, since I thought a lot of the allure of Qt was supposed to be that it's cross-platform. There are no Ruby bindings on the list you linked to. Of the remaining languages on the list, Pascal isn't widely used these days, PHP isn't a language I'd have thought of using for a GUI, and Ada isn't anything I'd use unless I was writing software for the US military.
Sorry folks, Linus essentially conceded this just yesterday. There will never be a 'year of the Linux desktop' because there will never be a single Linux desktop. Nobody seems to want it - or even to want to try to get as close as possible. Not the various distros, not Linus, not a hell of a lot of Linux fans.
Of course there's never going to be a single, universal Linux distro. It's the nature of open source that you can't just tell someone to stop using Slackware. What makes you think that's a problem? Ubuntu is the universal default for anyone who wants a desktop system that Just Works.
Incompatible distros and a chaotic development cycle are non-starters as far as mainstream desktops are concerned. ISV's won't target you - ISV's can't target you.
Does anyone know why Linus didn't require a copyright assignment?
The FSF requires it for GNU, but it's a lot of work. Laws differ in different countries, etc., so it's really not a simple thing to set up. The Linux kernel has a lot of contributors who found one line in a driver that they needed to fix, so they submitted one-line patches. Those people probably wouldn't have gone to the trouble of filling out all the paperwork; they would have just applied the patch on their own systems, and nobody else would have gotten the benefit of their patch.
With an open-source project, the copyright ownership isn't even normally such a big deal (unless you get sued by SCO). With MS, it's a whole different ballgame. If they want to demand copyright ownership of code that goes into Windows, it's something they have to negotiate with outside parties whose code they're paying for, and those parties may decide they don't want to do the deal on those terms, may jack up the price, etc.
Fourth sentence of article: "[...]Windows will never become an open source project in the same vein as Linux[...]"
Sixth sentence of article: "[...]I'll concede that some Windows source code probably will never see the light of day."
I think what he really wants to say is that the cost of Windows has to approach zero. That's completely different from being open source. It's the classic "free as in speech" (or as in freedom) versus "free as in beer."
I think it should be fairly obvious that MS can't open-source the whole OS. For one thing, I doubt that they own the copyright of every single line of code in Windows, and they've surely had to license a gazillion patents, make deals involving trade secrets, etc. Look at the situation with Linux and GPL 2 versus GPL 3 -- even if Linus changed his mind and wanted to make it GPL 3, it can't happen, because you won't get thousands of programmers to agree. With Windows it's bound to be even more complex.
Okay, so let's imagine that the price of Windows becomes zero dollars. So what? Then the US would be like China, just another country where everybody runs Windows and nobody pays for it. You'd still have banks telling you their web interface only works with IE. You'd still have people with hard disks full of documents that are in proprietary formats, preventing them from switching to Linux. Things like video encoders and color management would still be patent encumbered. The main effect would probably be to boost MS's market share, and that would probably allow them not just to sell more copies of Office, etc., but to abuse their monopoly more effectively for competitive advantage. That's essentially what the author of the article is talking about by the time he gets to pages 2 and 3.
And is anyone under the illusion that every version of Windows would cost zero dollars? No way. They'd very carefully set up a tiered system of price-differentiated versions of Windows in order to maximize their profits. Then it's like drug dealing: the first hit is free. This is what they're already doing in the third world, turning a blind eye to pirated versions of Windows because it helps to make those countries dependent on MS. The article says preinstalled Win XP is about $34 worth of the price of a new computer, and $34 is close enough to zero that I'd say that we're essentially already in that regime.
Except MS would probably develop a netbook version of Windows with a trimmed down version of Office; if only to maintain their dominance of the desktop. While it wouldn't make as much money as a full blown Windows license it would be valuable as a strategic move to protect their higher margin offerings.
The thing is that if netbooks got down to $100, MS's profit margin would have to be essentially zero. A world where users can freely choose Windows versus Linux, and where MS makes zero profit even on the users who choose Windows, is not a good world in which to be MS.
As long as netbooks are viewed as less-capable or lite versions of real notebooks that strategy will work. Hardware manufacturers have a vested interest in keeping that distinction as well in order to protect their higher margin laptops.
Dell has a vested interest in maintaining that distinction. Asus doesn't. Asus has been drifting closer and closer to the mainstream notebook market.
The main thing I like about OS PDF readers is that in 50 years you will still be able to open PDF's with the OS reader or its descendants, Adobe is unlikely to still be around.
I don't follow you. It's true that PDF is an open format that's been used for a huge number of documents all over the world, so I think it's extremely likely that people will still have easy ability to read PDF documents far into the future. But that's an argument in favor of PDF as an open standard, not an argument in favor of using a particular piece of software to read your PFDs. I have no idea what software people will be using to read PDFs in 2059. It may be open-source or proprietary, or it may be some people using one and some using the other, the same as today. It may be software written in the 2050's, or it may be old software running in an emulator that makes it think it's on an ancient x86 system.
Personally, I've never had a problem with Adobe Reader on any platform, and this site seems to be blatantly against it.
The site is a directory of open-source pdf readers. AR isn't open source.
Even if you don't care about open source, there are serious problems with AR:
It's too slow for me to be willing to use it as a browser plugin.
By default, it will execute javascript that's embedded in pdf files. This is both a privacy (people can track readers) and a security issue. After the first buffer-overflow exploit was announced, I kept it on my system. After the second buffer-overflow exploit, I deleted it. (If you want somewhat more security, disable JS: go to Edit, Preferences, JavaScript, and uncheck "Enable Acrobat JavaScript".)
The main functionality that AR has that isn't available in competing open-source plugins is all functionality related to DRM. I get along fine with Evince.
Seriously. Asus kept promising low price, but what they kept delivering was higher performance. The article says, According to Asus, the PC 1000HE will be available "soon" for approximately $400. Following the pattern they've always followed, that means it will actually sell for $800. What would have been really revolutionary would have been a $100 laptop -- but OLPC screwed up, and Asus decided to head up instead of down. If Asus had actually followed through on their original plans to deliver these things at low prices, we'd be seeing the imminent death of Microsoft. As it is, there may be some downward price pressure on Windows, but not enough to make MS surrender a whole sector of the market and allow Linux to move beyond the 1% share of the desktop where it's been stagnating for years now.
Interesting. I didn't know that you could get the three digest-format magazines electronically.
But even without including those, there are quite a few electronic-only magazines that pay professional rates of $.05/word or more. The Science Fiction Writers of America maintains a list of approved pro markets. The main criterion is that they have to pay 5 cents a word, but they also won't list them unless they have a regular publication schedule and a decent circulation. The following is a list of pro markets for SF (not fantasy or horror) short stories, omitting publications whose main focus isn't fiction:
I don't think there's any big difference in quality between the electronic and print markets. I write SF, and the list of magazines that I've managed to sell to is about evenly distributed between the print and electronic markets. The main differences aren't differences between electronic and print, they're differences in style. JBU specializes in action-adventure and military SF. Asimov's does mostly character-based SF. Strange Horizons publishes a lot of fiction that isn't as commercially oriented. There are also lots of differences in terms of business model. JBU is electronic, with a subscription model; they let you read the first half of each story without a subscription, but you have to pony up to read the rest. Strange Horizons is a nonprofit foundation, free to read online. Card's won't let you read anything without a subscription.
Slashdot Mirror
Hee hee ... cool link, and I think "The Dancing Bunny Problem" clearly needs to go into the lexicon along with PEBKAC.
But I don't think it's true that botnets, etc., are inevitable for purely social/psychological reasons. If they were, we'd have botnets that ran on MacOS and Linux. There are certain simple, easy design decisions that you can make in an OS that will have a positive effect on security. MS decided that (in Unix terms) users would always be logged in as root by default. That was a bad design decision. There are many other bad design decisions built into Windows. MS doesn't even deny that these were bad decisions; they're trying to fix some of them with UAC, for example.
My mother in law got her Windows box horribly infested with viruses, to the point where it couldn't boot. She's actually a pretty smart lady. I helped her install Linux (dry run while she was visiting us here in California, then she did it on her own in Buffalo, with help over the phone). It's possible that she would try to download and run the dancing bunny. The thing is, she would find it difficult to download and run the dancing bunny on Linux. Because of the design of the OS, she would get dialog boxes along the way saying, "Are you sure you want to do this?" or "Please enter your administrator password." She would probably have time to bail out. On Windows, effectively running as root, the OS wouldn't have guaranteed her any of these chances to back out.
To the Editor:
Re "A New Internet? The Old One is Putting Us in Jeopardy," by John Markoff (Week in Review, Feb. 15, 2009):
Mr. Markoff both misstates and overstates the security problems faced by the Internet as currently designed.
He never uses the word "Windows," but the virus outbreaks he describes are almost entirely a Windows phenomenon, and due to the poor design of that operating system. Microsoft's apologists have been saying for years that this was only because Windows' market share made it the more attractive target. But Apple's share of the desktop market has skyrocketed recently to 15% without any outbreaks of viruses targeting the Macintosh. And Microsoft has never commanded more than about half of the server market; the other half runs open-source operating systems such as Linux (used by Google) and FreeBSD (Yahoo), on which viruses are essentially unknown.
Markoff says it's hard to prove your identity on the internet, and proposes government regulation as a solution. But many people have been proving their identities for years now using proven technologies like public-key cryptography. The U.S. government played a negative role in the development of these technologies by attempting to regulate their distribution through export-control regulations originally intended for munitions.
As far as I can tell from some quick googling and checking on Wikipedia, jffs2 isn't much of a competitor at this point, e.g., it's apparently not really usable on flash chips bigger than 512 Mb. Maybe UBIFS or LogFS? None of them seem to be really mature.
Aw, come on. Sure, there were two "it's" mistakes in the article, but they both occurred in one sentence. The entire second sentence was completely free of "it's" mistakes. That represents real progress for slashdot. Actually, I believe slashdot's Central Committee For the Eradication of "It's" Mistakes (also known as COMINITS) has set a goal of reducing the average rate of "it's" mistakes to no more than three per sentence by the end of the current five-year plan. That means that production of "it's"-mistake-free sentences is actually 783% ahead of last year's production!
Compare with Wine. The typical reason to use Wine is that you prefer Linux, but you don't have any choice because the app you need/want to use is Windows-only.
Since MacOS X isn't a monopoly, that doesn't occur very much with MacOS X. Nobody's bank tells them they can only access their account using MacOS X.
Ths typical Mac user likes MacOS because all the software and hardware is beautifully integrated and consistent, and everything Just Works. The last thing on earth that type of person is going to do is switch to Linux, but then insist on trying to run some kludged-together version of a Mac application that wasn't really designed to run on Linux.
Because your mobile devices are proprietary systems, and the companies who sold them to you don't want use every possible piece of functionality as a revenue stream.
If you want to synchronize your files between various devices, using open-source software, try unison. It's free, it's open source, it's fast, and IMO it's of very high quality. I use it to sync two desktops, a server, and an ARM-based network appliance (NSLU2). The key is that none of these are locked down systems sold to you by a cell phone company.
TrueCrypt's FAQ discusses this. See the question beginning "We use TrueCrypt in a corporate..."
This latter-day sequel is good: Flatterland: Like Flatland, Only More So.
The OP said "influence," not "dictate." Clearly the OP isn't the top IT honcho at this university, but he/she doesn't claim to be.
For someone who really was the top manager in charge of IT at a university, my advice would be to start laying the groundwork over the next few years via the hiring process. Every time you interview someone for an IT position, make sure that one of your criteria is that the person has broad problem-solving skills, broad interests, and at least some academic qualifications in actual CS, as opposed to someone who just has MS certs. The school where I teach is pretty much the kind of MS monoculture the OP describes, except that there are some pockets of resistance in the fine arts division where they still use macs, despite intense pressure to standardize on Windows. The huge problem is that a lot of the IT staff doesn't know anything about Linux or OSS, and has never used Linux or OSS. Some of them are curious and willing to learn new things. Others are not. It doesn't help that the ratio of staff to boxen is on the low side, which tends to make them resist any complication such as supporting another OS. In this kind of environment, hiring even one tech who uses Ubuntu at home could make a huge difference.
I didn't say that it ought to be produced and sold new for $50. I just said its specs were obsolete, so there was no way I'd buy it new. When you buy new hardware, you want it to be hardware that's going to be usable for a long time before it becomes obsolete. I'm much more willing to buy obsolete hardware if it's used and therefore very cheap.
Well, yeah, but your analysis a little off in some ways.
I own one of the WalMart gPC machines, and it's worked fine. But you typically can't walk into WalMart and buy them. You pretty much have to order them online, and therefore very few people know about them.
Re Target, my wife happened to go to Target today, where she noticed that they had the Eee PC out on a retail display. The Windows and Linux versions were side by side, with the Linux one $50 cheaper. However, the hardware they're selling in Target is an extremely out of date configuration with only a 4 Gb hard disk and 512 Mb of ram. (According to the WP article, it's a configuration that is only being sold in Target.) I wouldn't mind picking up a machine at a garage sale with those specs for $50, but I'm not about to buy a new machine with those specs.
So realistically, you really can't walk in to Target or WalMart and buy a decent, modern netbook -- not quite yet.
And while it may be true that some versions of Windows cost several hundred dollars, the cost of Windows to the OEM for low-end machines is as low as $34, so your comparison is somewhat misleading.
There are several other goofy things about the summary.
As you pointed out, there's no evidence that this is specifically about Ubuntu. The other goofy thing in this sentence is the reference to "desktop success," which makes it sound as though Ubuntu is already a successful competitor, and MS is responding to that. Now I use Ubuntu, love Ubuntu, and I think it's great that companies like Asus and HP are shipping machines with Ubuntu preinstalled, but as far as anyone can tell, Linux's share of the desktop is still stalled at about 1%. Asus and OLPC are actually no longer exclusively tied to Linux. I think it's much more reasonable to interpret this as a move to fight against a competitor that is currently not successful at all in any quantitative sense, but threatens to become successful (i.e., start growing beyond 1% market share) in the future.
There's no evidence to back up the part about "specifically Ubuntu."
Uh, except that that MS job announcement specifically refers to desktop Windows. They're clearly advertising for a position for someone to be a cheerleader for Windows versus Linux, to head off any hypothetical future erosion of their market share to Linux.
The SEC filing refers only to servers when it comes to competition from OSS. That's because the server market, not the desktop market, is where MS currently has to compete with Linux and BSD on relatively even terms. So the final paragraph of the summary brings up two points that are unrelated to each other (browser versus OS) and unrelated to the job position (which is about desktop, not server).
You're really discussing three separate things here: (1) autosave, (2) keeping all the old versions around forever by default, and (3) eliminating the ability to give someone else a file without also giving them access to all the old versions. You can have 1 without 2 and 3 (commonly implemented in word processors), or 1 and 2 without 3 (as in VMS).
Autosave is bad in the situation where the cat walks across your keyboard while you're making tea. Autosave is good in the situation where the power goes out unexpectedly.
In a version control system, it's useful to have ways for the user to tell the system that the project is in a sane state at a certain point (e.g., git commit), and also to have ways for the user to say that certain states of the project are especially good/complete/debugged (e.g., git tag). The good thing about explicit saves, as opposed to autosaves, is that they perform this function. On the other hand, the user may have other mechanisms available for performing this function. For instance, I use a file synchronizer (unison) to sync up a certain set of machines that mirror my files. Usually I do a sync when I'm at a good, sane stopping point in a certain task. Suppose I'm using a word-processor, and the cat walks across the keyboard while I'm making tea, causing my entire novel to be erased and then autosaved. No big deal. I just sync up with the mirror machine and grab the undamaged version. I think MacOS's Time Machine provides similar "oh shit" functionality.
When a user is used to explicit saves, it could be argued that autosaves are bad because they're inconsistent with the way the user's other apps work. On the other hand, we seem to be seeing a proliferation of systems these days that are essentially version control systems reimplemented in different ways -- built into file systems, built into applications, etc. E.g., if I'm editing file foo using emacs, I could potentially have three files: foo, foo~, and #foo#, which sort of represent three levels of version control. This is good in the sense that it's consistent with the usual explicit-save paradigm, but bad in the sense that it's inconsistent with the way other applications handle it.
Yeah, I think they switched from ThizLinux to Lindows at some point. Too bad Fry's doesn't sell them anymore. They really were, uh, great quality in my experience :-) Apparently Fry's were just getting way too many returns, and it wasn't profitable for them. Since my machines ran great year after year, and are all still running, I assume the returns weren't due to hardware problems, but were from people who didn't understand that they weren't getting a machine with Windows installed.
Ubuntu is simply the sane thing to put on a desktop machine these days, especially for users who may not already be familiar with Linux.
It was always really frustrating to me in the past to see hardware companies selling machines with Linux preinstalled, but with some crappy version of Linux that was bound to create a bad impression of Linux in general.
Back when Fry's was selling Great Quality boxes for as low as $180, I bought several of them. They had something called ThizLinux on them, which was apparently a distro that GQ created themselves. No documentation for ThizLinux came with the machine, and googling for ThizLinux turned up a Chinese-language web site with no English translation. The printed docs that came with the machines were actually 90% information on how to wipe ThizLinux off your hard disk and install Windows. The impression any user would get from this was probably that Linux was crap, and nobody really wanted it.
Same deal with the Everex gPC, which I reviewed a while back. This may be a little unfair, because what I bought from them was a beta of their gOS distro, and now they have a newer version out, but basically it sucked, and I very quickly decided to replace it with Ubuntu. IMO it was just foolish of Everex to put out their own distro. I think they were imagining that by making it look slick (and a lot like MacOS) they would attract users. But in reality it worked so poorly that I think they were shooting themselves in the foot.
I teach physics labs in a room with 7 school-supplied Windows machines and 5 Linux boxes I got at garage sales, Good Will, etc. At various times I've had Gnome, KDE, and Xfce on various machines. As in the video, my students often don't realize that they're not running Windows. There can also be kind of a converse effect, however, where they won't even try to use a Linux box because they think it's going to be hard, or different, or not as good. In other words, you could have a machine that really had Windows on it, but if you put a post-it on the side saying "Linux," they'd say, "I tried that one, and it was too hard."
Xfce does indeed perform somewhat better than Gnome and KDE on low-end hardware, while still giving people a fairly familiar user interface based on a screen littered with little icons. Xubuntu is what I have on those machines, and it generally works fine. (The only problem I've had is that on machines with low-resolution monitors, sometimes part of an application's window is hidden behind the tray at the bottom of the screen, and I haven't been able to figure out how to reveal it. On KDE the same problem happens, but I think you can hold down the alt key while dragging the window.) I think different people have different tolerances for unresponsiveness. Gnome has always felt unacceptably slow to me on every machine I've ever tried it on.
Personally, I've never understood why people like icons. I prefer fluxbox, which is very fast, and doesn't make my computer screen look like someone took the laundry out of the dryer and threw it on the kitchen floor.
The PerlQt interface hasn't been updated since 2003, so I think it's dead. The python bindings aren't free on Windows, which would dull my enthusiasm, since I thought a lot of the allure of Qt was supposed to be that it's cross-platform. There are no Ruby bindings on the list you linked to. Of the remaining languages on the list, Pascal isn't widely used these days, PHP isn't a language I'd have thought of using for a GUI, and Ada isn't anything I'd use unless I was writing software for the US military.
Of course there's never going to be a single, universal Linux distro. It's the nature of open source that you can't just tell someone to stop using Slackware. What makes you think that's a problem? Ubuntu is the universal default for anyone who wants a desktop system that Just Works.
Sounds like you haven't heard of the Linux Standard Base.
The FSF requires it for GNU, but it's a lot of work. Laws differ in different countries, etc., so it's really not a simple thing to set up. The Linux kernel has a lot of contributors who found one line in a driver that they needed to fix, so they submitted one-line patches. Those people probably wouldn't have gone to the trouble of filling out all the paperwork; they would have just applied the patch on their own systems, and nobody else would have gotten the benefit of their patch.
With an open-source project, the copyright ownership isn't even normally such a big deal (unless you get sued by SCO). With MS, it's a whole different ballgame. If they want to demand copyright ownership of code that goes into Windows, it's something they have to negotiate with outside parties whose code they're paying for, and those parties may decide they don't want to do the deal on those terms, may jack up the price, etc.
title of article: Why Windows Must Go Open Source
Fourth sentence of article: "[...]Windows will never become an open source project in the same vein as Linux[...]"
Sixth sentence of article: "[...]I'll concede that some Windows source code probably will never see the light of day."
I think what he really wants to say is that the cost of Windows has to approach zero. That's completely different from being open source. It's the classic "free as in speech" (or as in freedom) versus "free as in beer."
I think it should be fairly obvious that MS can't open-source the whole OS. For one thing, I doubt that they own the copyright of every single line of code in Windows, and they've surely had to license a gazillion patents, make deals involving trade secrets, etc. Look at the situation with Linux and GPL 2 versus GPL 3 -- even if Linus changed his mind and wanted to make it GPL 3, it can't happen, because you won't get thousands of programmers to agree. With Windows it's bound to be even more complex.
Okay, so let's imagine that the price of Windows becomes zero dollars. So what? Then the US would be like China, just another country where everybody runs Windows and nobody pays for it. You'd still have banks telling you their web interface only works with IE. You'd still have people with hard disks full of documents that are in proprietary formats, preventing them from switching to Linux. Things like video encoders and color management would still be patent encumbered. The main effect would probably be to boost MS's market share, and that would probably allow them not just to sell more copies of Office, etc., but to abuse their monopoly more effectively for competitive advantage. That's essentially what the author of the article is talking about by the time he gets to pages 2 and 3.
And is anyone under the illusion that every version of Windows would cost zero dollars? No way. They'd very carefully set up a tiered system of price-differentiated versions of Windows in order to maximize their profits. Then it's like drug dealing: the first hit is free. This is what they're already doing in the third world, turning a blind eye to pirated versions of Windows because it helps to make those countries dependent on MS. The article says preinstalled Win XP is about $34 worth of the price of a new computer, and $34 is close enough to zero that I'd say that we're essentially already in that regime.
The thing is that if netbooks got down to $100, MS's profit margin would have to be essentially zero. A world where users can freely choose Windows versus Linux, and where MS makes zero profit even on the users who choose Windows, is not a good world in which to be MS.
Dell has a vested interest in maintaining that distinction. Asus doesn't. Asus has been drifting closer and closer to the mainstream notebook market.
I don't follow you. It's true that PDF is an open format that's been used for a huge number of documents all over the world, so I think it's extremely likely that people will still have easy ability to read PDF documents far into the future. But that's an argument in favor of PDF as an open standard, not an argument in favor of using a particular piece of software to read your PFDs. I have no idea what software people will be using to read PDFs in 2059. It may be open-source or proprietary, or it may be some people using one and some using the other, the same as today. It may be software written in the 2050's, or it may be old software running in an emulator that makes it think it's on an ancient x86 system.
The site is a directory of open-source pdf readers. AR isn't open source.
Even if you don't care about open source, there are serious problems with AR:
The main functionality that AR has that isn't available in competing open-source plugins is all functionality related to DRM. I get along fine with Evince.
Seriously. Asus kept promising low price, but what they kept delivering was higher performance. The article says, According to Asus, the PC 1000HE will be available "soon" for approximately $400. Following the pattern they've always followed, that means it will actually sell for $800. What would have been really revolutionary would have been a $100 laptop -- but OLPC screwed up, and Asus decided to head up instead of down. If Asus had actually followed through on their original plans to deliver these things at low prices, we'd be seeing the imminent death of Microsoft. As it is, there may be some downward price pressure on Windows, but not enough to make MS surrender a whole sector of the market and allow Linux to move beyond the 1% share of the desktop where it's been stagnating for years now.
Interesting. I didn't know that you could get the three digest-format magazines electronically.
But even without including those, there are quite a few electronic-only magazines that pay professional rates of $.05/word or more. The Science Fiction Writers of America maintains a list of approved pro markets. The main criterion is that they have to pay 5 cents a word, but they also won't list them unless they have a regular publication schedule and a decent circulation. The following is a list of pro markets for SF (not fantasy or horror) short stories, omitting publications whose main focus isn't fiction:
I don't think there's any big difference in quality between the electronic and print markets. I write SF, and the list of magazines that I've managed to sell to is about evenly distributed between the print and electronic markets. The main differences aren't differences between electronic and print, they're differences in style. JBU specializes in action-adventure and military SF. Asimov's does mostly character-based SF. Strange Horizons publishes a lot of fiction that isn't as commercially oriented. There are also lots of differences in terms of business model. JBU is electronic, with a subscription model; they let you read the first half of each story without a subscription, but you have to pony up to read the rest. Strange Horizons is a nonprofit foundation, free to read online. Card's won't let you read anything without a subscription.