It sounds to me like he's talking about the Eolas suit against Microsoft in the first sentence. He leads off talking about how Microsoft has needed to defend its turf because of the nature of the OS business, says that RedHat would be crushed by a judgment of the same kind which Microsoft was handed (the fine was $500M), and then points out that the Eolas suit is not unique and that Microsoft faces suits like that quite frequently. That would seem to be how the sentence was meant given the lead-in, but maybe he intentionally phrased things ambiguously enough to be taken either way.
I agree with your interpretation. Anderer is saying that litigation has become commonplace in the IT industry. Some of these lawsuits (eg, Eolas vs Microsoft) might result in judgements on the order of $500 million. He believes Microsoft is defending themselves against 50 such suits right now and he doesn't think open-source companies would have the cash reserves to defend against similar nuisance lawsuits (esp. w.r.t patents). If open-source was to dominate the landscape then the industry would be decimated in mere months by the sheer number of $500 million judgements against developers.
He is basically saying "this is a game for the big boys and you open-sourcers are too puny to play in this arena, so go home and stop bothering us". Arrogant, true. But possibly correct. I don't personally believe his argument but I can see where he's coming from.
Of course, the common interpretation is that Anderer is threatening open-source companies. The last line is being intepreted to mean Microsoft is initiating 50 more lawsuits like SCO vs IBM. I don't buy that argument because Anderer would have to be stupid to voluntarily tell everybody about the conspiracy. Unfortunately because certain prominent people are stating that particular interpretation as if it were fact, all other interpretations are being ignored or shouted down.
How well did they rate "Using non-standard hardware." Including printers, laptops, and virtually unknown brands of cdroms, ethernet devices, win modems (Ok maybe this one doesn't matter much anymore), windows keyboards (how much function can you get out of those functions), etc.
An article doesn't have unlimited space. At some point they have to draw the line, cut some words to come under the limit, and call it a day. But I can try and answer some of your questions.
Printer support recently got very good. By installing CUPS, GS-ESP, HPIJS and FOOMATIC you receive high-quality printer drivers for several 1000 popular printers. By the way, I'm telling you those cryptic names for your own education, but the typical user just sees the "Printer" icon on their desktop. CUPS also has support for automatically detecting and installing USB and Plug and Play Parallel Port printers.
CDROMs these days all use the ATAPI standard, so there is no problem with "unknown brands". The bad old days of proprietary CDROM interfaces are long gone (and good riddance).
Ethernet is ridiculously well supported under Linux. Most Ethernet vendors seem to provide Linux drivers from their websites. Most of them also provide source code which gets quickly bundled into the next version of Linux.
The special keys on Windows keyboards have worked for a long time now (at least 5 years, probably more). The multimedia keys on modern keyboards also work automatically. KDE and GNOME both have graphical setup programs to allow the user to assign functions to those keys.
I'd like to know how well speed step performs in linux.
SpeedStep works beautifully with an interface for programmers exposed by the SYSFS module (you don't need to know that as a user). This SYSFS interface is standardised across all processors, not just Intel's SpeedStep. So the CPU frequency scaling also works on PowerPC and AMD chips. You as a user don't need to know this; any decent Linux distro will automatically install a daemon like "cpufreqd" which freq-steps the CPU as needed.
Can a laptop hibernate if it needs to?
Yes, of course, though support is spotty. I don't have hibernation on my laptop (iBook G4) but earlier versions of the iBook work (iBook G3). The Intel laptops usually have better luck.
On my i8k in Mandrake it allows either battery or external power. If I plug in / detach the cord, the computer freezes (although the battery does kick in because whatever was on screen stays on screen).
That's no good. So it's not perfect. Nobody ever claimed it was. Maybe it will get better in the next release.
I can only assume everything worked by the high ratings given. Also I can assume from the ratings that the system instantly recognizes when any new USB/Firewire/PCMCIA device is hotswapped, and in most cases can identify it. Right?
I have had good success plugging in Firewire disks and CDROM drives and having them automatically work. USB keys, mice and keyboards were also automatic. PCMCIA is also automatic though driver support is spotty. It can always identify the device, but whether it's supported is another matter.
I mean, I did see the word "excellent" so I think it should at least be equal with Windows.
You do realise that many people consider Linux to be better than Windows? I suppose if you come from a Windows background you might think Linux is a little "crappy" because it doesn't support your video card, or doesn't hibernate correctly on your laptop. But consider that some of us come from a UNIX background and we think Windows is crappy because... well... it isn't like UNIX!
Always bear in mind, if you're looking at Linux as a free version of Windows, you will always be disappointed. If you're looking at Linux as a free version of UNIX, you will probably be pleasantly surprised. If you're looking at Linux with no preconceptions then, in my limited experience with such people, you will be happy with the results.
Prior to MS DOS, every operating system was sold by a hardware manufacturer, and they wouldn't sell the OS without a computer. But Microsoft changed that. With MS DOS, it was possible for computers from two different manufacturers to run the same application without porting or recompiling.
MS DOS allowed Compaq to clone the IBM PC, which introduced real price competetion into the world of PC hardware, and eventually gave us the fast cheap machines we all use.
If IBM hadn't sub-contracted out their OS work to another company, computer technology wouldn't have advanced nearly as fast. (That company didn't necessarily have to be Microsoft, anyone could've done it, but Bill Gates lucked out.
Your reasoning is correct but one niggly fact was wrong. You talk about OSs only being sold by the hardware manufacturer and you say "Microsoft changed that". But CP/M predates MS-DOS by a very long time and it was available on many personal computers from many vendors. You could even get CP/M for the Commodore 128.
While I agree with you that divorcing hardware from the software was important for the growth of the IT industry, Microsoft wasn't the first company to do it. Even UNIX could be properly seen as divorcing hardware from software; you could run UNIX on dozens of different minis (not PCs) well before Microsoft even existed.
Loneliness is the 1st problem for the senior citizens here in Europe. We don't need robots to assist them we need human beings to keep them company. I thinkt hat being surrounded by machines is even more depressing than being all alone, I'd feel totally worthless if I was given to a robot to take care of me.
These aren't robots to keep them company. They're robots to keep them clean and robots to help them with everyday tasks. Get a grip on the situation and get off your soapbox.
I suppose in your world we shouldn't allow old people to drive cars because then they'd be lonely. Instead, we should have 6 fit young men carrying each old person around on a litter.
Instead of impersonal cooking machines, like microwaves, we can just hire teams of people to breathe heavily on the food until it's cooked. Heavens forbid that old people use a "technical solution" to cook their food.
I wonder what you might say if I explained to you the concept of a phone; a "technical solution" that allows for *greater* human interaction. Probably your head would explode as you tried to reconcile the paradox.
Stop being such a drongo. Robots to keep the elderly clean is a great thing. It means they'll receive better care, at an affordable price, and they can clean themselves when they want to rather than when the overworked nurse is available. A nurse, by the way, who could actually improve the quality of their patients lives if they weren't wasting their valuable time giving sponge baths.
Sat in a meeting with a Senior VP who was trying to convince everyone that you could replace a 8 way 400 MHz Sun E4500 with 1 PC. His reason? Because 8x400MHz = 3200MHz, so all you would need is one to two 3 GHz processor intel system.
It's often amazing how otherwise intelligent people can be unbelievably stupid.
For example, I have a friend who got top marks in Computer Science, went on to do Honours, and he is now doing his Masters in his spare time. He's one of the brightest people I know. He can code like a natural, can design like a professional, and holds down an important and well paid job as a J2EE designer.
One day he was describing to me how a PIII running Linux is way better than a Sun E10k. I thought he was joking at first. No, he was being serious. Apparently he'd ran the same database software and J2EE application on both systems and the Linux system smoked the E10k in performance.
After a little prodding I find out his domain on the E10k was only 4 CPUs and was memory starved.
After a little more prodding I find out that the database software they were using was single-threaded.
Yet pointing these facts out didn't convince him then of his mistake, and to this day he is convinced that E10ks are gutless wonders (which is true but not for the reasons he thinks).
So I learnt that day that anybody can make a mistake. The important lesson to learn is that the idiot might be yourself. I make an effort these days to be more humble when someone tells me I'm wrong; they might just be right.
In other words, SCO is claiming that $1million+ "worth"
of licenses were sold. So that's $1mill/$699 = 1400+ licenses, or $1mill/$1399 = 700+ licenses. SCO's own quarterly says only $20k
income from licensing this quarter. It's possible the EV1 payments are
in stages, or won't appear on SCO's financials until next quarter,
but it's also possible that EV1 only paid $20k for their licenses.
But SCO is spinning this to sound much more impressive. EV1 was the patsy here; they thought they were getting a great deal, but they were just another pawn in SCO's (Microsoft's?) smear campaign against Linux.
You rely on the fact that "ideas can be duplicated without effort": please remember that while duplication of ideas costs near zero, the development of those ideas can cost quite a lot.
Yes, I am aware of that, but my point was to illustrate that "resource allocation" is an inappropriate way to describe patents. Ideas are not a resource to be allocated.
case-by-case would be too unworkable:
They already evalute each patent on a case-by-case basis, so I hardly think it would be unworkable.
All that and here's the bag of chips: The Debian "Software Update" (apt) updates EVERY APPLICATION ON THE SYSTEM. Compare to OS X, where I get my Apple updates through SWU and have to download and install Adobe updates, new builds of BZFlag, new versions of Quicksilver, etceteras myself.
I've always wondered why Apple (and Microsoft for that matter) haven't partnered with major application vendors to supply application patches and updates through the Software Update feature.
I can imagine some of the reasons - too hard, too many variables, not our problem - but surely if the free software guys can work it out, it's not an insurmountable problem.
Downloading from the net is not illegal. Putting stuff on the net you don't have rights to is - but the downloaders aren't doing that. They are just accessing publicly available information.
Utterly wrong.
The downloaders are making a copy. They send the request to the server, initiating the copy and the transfer. The downloaders are responsible for making the copy. If they make a copy of copyrighted material where the license doesn't permit copying, then they have infringed on copyright.
The downloader could perhaps argue that they weren't aware the copyrighted material was licensed in such a way to prevent copies being made from the webserver. Depending on the circumstances, they might even convince a judge of their innocence. But it doesn't seem likely if the filename is "Windows2000SourceCode.zip" in the "warez/0" directory.
If you need a real world analogy to help you understand this point. If you buy a television from the back of a truck, and the asking price is 1/10th what it costs in the store, but you can see that it's second hand, and the seller wants cash-only, and he won't give you a receipt, then you might have difficulty explaining to the local magistrate that you didn't realise the goods were stolen.
In a day and age where automation has made neither food nor labor scarce society shouldn't be running in a mode where everybody is compelled to try to be king-of-the-hill just to life a modest life. It seems like we've turned into a society where either you are one of the few people who reap the benefits of modern technology (ie you own the capital), or you are one of the people displaced by it (ie it you work at Walmart).
You sound exactly like Karl Marx;-)
He argued that the master-slave society could only give way to communism once there were sufficient resources to sustain everybody's basic needs.
He also said that the "haves" would fight to prevent the "have-nots" from sharing in the wealth. Looking at the world around us today, it is like Karl Marx had a window into the future.
I disagree with a lot of what Karl Marx wrote - I think he ignores the fundamental problem of human greed, as well as the ever-increasing lower limit of our basic "needs" - but it's always a treat to read The Communist Manifesto and realise how right he was about so many things. Marx has a a better hit-rate for his predictions than Nostradamus!
The patent system awards the first to invent (or, outside of the US, the first to file), and it's just tough luck if someone beats you to it. That's life in a competitive system. If you don't like it, change society or live in a commune where another resource allocation method is used.
The problem is that ideas aren't a finite resource to be allocated like wheat or coal or milk. Ideas can be duplicated without effort. I'm reminded of this quote:
"He who receives an idea from me, receives instructions himself without lessening mine; as he who lights his taper at mine, receives light without darkening me." -- Thomas Jefferson 1813
I'd also like to express my distate for the "if you don't love the USA then go to Russia" style of argument. It's intellectually bankrupt.
Read the US constitution and find that the patent system is there for the people too - that includes you and I as creative individuals. The bargain is a _limited_ monopoly for 20 years. 20 years if not a bad payoff for the time and effort and risks involved in pursuing inventions.
That would be OK if the invention was actually innovative such that the 20 years was justified. Unfortunately a number of patents are granted that are completely without merit - they are either unoriginal, obvious, or trivial - and as such they are undeserving of a 20 year monopoly.
I think the best reform for patents would be variable periods for the government granted monopoly. Say, between 18 months and 20 years, decided by merit on a case-by-case basis.
Wow, after seeing the MPAA issue hundreds of subpoenas, it somehow didn't occur to me that DVDCCA didn't actually sue 321 studios, but a company named 321 Studios.
It was actually only 78 studios, but some of those companies had 8x speed burners.
Supposition: Mac users actually buy software, Linux users demand stuff for free. Every platform has its user quirks. I think Linux's is that they all want everything gratis.
In the past 10 years I have spent over $400 on Linux distributions and $1500 on Linux software.
That's just for personal use. I have spent easily 10x that for business purposes.
This is the man who dismissed internet gaming as a fad,
I only wish he was right. I don't know if I'm the only one, but I DETEST Internet gaming. All the online games are full of cheats and weenies and fu><0r j00 d00d morons. The servers are always full, or down, or slow, or whatever. But it's never a smooth experience. I'm constantly frustrated when trying to connect. Plus with half the online games charging a monthly service, it's simply too expensive for a casual gamer like me.
I'd like to think I could just say "pfeh, I don't play them, but they don't bother me". But it's not true. Internet games are sucking up all the developer resources. Every second game has Internet connectivity. That's wasted effort on a feature I'll never use, but I still have to pay for.
I do like LAN games. LAN parties are great. I also like party games (ie, 4 of you crowd around the same telly). But Internet gaming? No thanks. Been there. Done that. Wasn't interested in the t-shirt.
You're lucky. You've got yourself a half decent text editor there. Imagine if you'd taken 200 gigs of compressed source, 22 days of compiling, and found that you only ended up with emacs.
Don't be silly. The entire EMACS code won't fit into a mere 200 gig of compressed source.
Sadly, there are still very few games for Linux and Mac,
Even worse, the very few games there are have fallen badly into disrepair.
I bought pretty much every game released by Hyperion, Loki, Sirtech and Vicarious. Nearly 2 dozen titles. Half of them don't work on a modern (Debian/unstable) x86 installation. I found 5 of the Loki games were ported to PPC; only 1 of those still works (Heroes 3). This isn't a Linux specific problem. The few DOS games I own (eg, X-Wing) don't work on any modern Microsoft platform.
To contrast, the open-source commercial games (Doom, Quake, Quake2, Starcon2) work flawlessly on all the Linux platforms I've tried. Just goes to show that with enough interest, any game can stay ported to the latest platform, but you need source code!
For Linux gaming I've decided to giveup on closed source. Either the game comes with source - I'll still pay for it - or I'll just stick with what I've got. The open source games aren't very good (with a few notable exceptions like nethack) but I refuse to purchase another closed source game that lasts less than a year before bit-rotting.
PS: I actually get a fair bit of mileage out of emulators (eg, dosbox). But I know that won't appeal to most gamers.
Are there third-party vendors recompiling Windows, Office etc? I haven't seen this, although it could be true. My perception is that a given release of a Microsoft product will contain the same binaries no matter which vendor distributes/installs it. That allows for much easier verification that there has been no tampering with the binaries.
You don't need to recompile Windows to change its behaviour. You just need to install certain binaries. For example, the GINA authentication DLLs, or the LiteStep explorer replacement.
And viruses don't need to recompile a binary to infect it. They simply modify the PE header then piggyback at the end of the file. This was the case before Windows even existed.
I disagree that it's "easier" to verify that binaries haven't been tampered with. Microsoft releases many versions of their DLLs and they do not provide MD5 sums. You could compare against a known good source but how would you know if the difference was because of deceit or because of an upgrade? You'd need access to all valid versions of Windows, including every intermediate version, including non-Microsoft versions that are bundled with third party software (eg, DirectX betas are often bundled with third party games, but similar bundles occur for non-gaming software).
Microsoft tried using digital signatures as an improvement over MD5 sums, but their vetting processes were flawed. You might remember the ActiveX control that was signed despite its only action being to reboot the computer. Microsoft revoked the signature but only after the author brought their attention to it. Who says that there aren't trojans in existing ActiveX controls? They don't even have to be intentional; the flaws in Microsoft code used by viruses aren't intentional.
Finally, malware already exists despite there being no access to Windows source code. Scripting languages in Windows are powerful enough to cause the same disruption as corrupted binaries. So a perfect set of uncorrupted binaries proves nothing.
Does that make sense?
I hear what you're saying but I disagree that it makes any practical difference. The example the author provided - a rogue contractor delivering trojaned software - can happen with closed source. The author claimed that open-source makes trojans easier to deploy, but I see no evidence to support such an outlandish claim, and my own understanding of the issues leads me to disagree with such claims.
You need to trust your staff and your vendors. That is the bottom line.
I don't think you quite understood his scenario. Let's say Vendor X gets a contract to provide a government agency with 800 desktop computers, with Linux, OpenOffice, etc. Meeting a bunch of carefully written specs from that agency's IT department. Vendor X takes Fedora or Gentoo or Debian and customizes it, complete with a "Foo Agency" splash screen, encrypted disk partitions, escrowed bypass for crypto, etc.
How do we know they didn't plant malware in OpenOffice? What geeks will have access to this binary? Geeks won't even know this mini-distro exists. How much do you know about the Linux being used by Burlington Coat Factory, for example?
"Let's say Vendor X gets a contract to provide a government agency with 800 desktop computers, with Windows, Office, etc. Meeting a bunch of carefully written specs from that agency's IT department. Vendor X takes Windows XP and customizes it, complete with a "Foo Agency" splash screen, encrypted disk partitions, escrowed bypass for crypto, etc.
"How do we know they didn't plant malware in Windows? What geeks will have access to this binary? Geeks won't even know this mini-distro exists. "
The problem with your example, and with the article that preceded this thread, is that it discusses problems that are common to both open and closed source. The real question is "how can we trust contractors to not screw us". Blaming open-source is disingenuous.
Over the past decade or so I've tried Linux on and off a half-dozen times. Every time, I've gone back to Windows, which blows goats but at least lets me get my g.d. work done instead of having to continually f*** with obscure configuration files.
Whenever I use Windows I find it a frustrating experience, having to deal with obscure registry settings and drivers and service packs.
FreeBSD feels, to me, like it was designed. Linux always feels like it just accumulated by accident.
Comparing FreeBSD to Linux is like comparing a Toyota Corolla to a V8 engine. Try comparing FreeBSD to a distribution like Suse, Red Hat or Mandrake.
My experience is that FreeBSD is no better or worse than any of the community driven distros like Gentoo or Debian. Seeing as the majority of userspace is the same (XFree86, OpenOffice and GNOME) that's really no surprise. It's strange to claim FreeBSD is "designed" whereas Linux is not, because most of the software in FreeBSD is accumulated in exactly the same way that it is accumulated in every Linux distribution.
HarleyPig asks: "I'm using my tax return to buy a laptop. I don't want to know which laptop to get (that's a religious discussion I'd like to avoid). What peripherals do you find yourself wishing you'd bought, or have ended up buying? I know I'll need a mouse, extra cabling, extra batteries and some kind of case to hold and carry around the laptop. What else should I consider putting in my list of stuff to buy with a laptop?"
In my opinion... don't buy many peripherals. The problem with buying lots of peripherals is that your ~2kg laptop suddenly needs a bag the size of a small suitcase weighing in at 10kg or more to carry around all the junk you "need".
Take the mouse, for example. You've already decided to get one. Why? The trackpad really is not that bad. The mouse needs a flat surface to work on, which ties you to a desk. I've found I can work the trackpad fairly well, after just a few days practise.
Cables? Why? If you're going to be in a place where you need a cable, there's likely to be one there already. Take for example people who carry a serial cable around so they can "plug-in" to a Cisco router while onsite. Have you ever seen a comms room without a serial cable? I haven't. So why carry the cable? Same for Ethernet. If you simply *must* have a backup cable, put it in the boot of your car.
Extra batteries are only useful if you keep them charged. In my experience, nobody is that organised. They'll have two batteries and one of them is always flat. So when their live battery goes flat... well... they're in the same boat as the rest of us with only one battery.
I managed to avoid getting a computer bag for the laptop. A leather briefcase costs the same and looks heaps more professional. Modern cases have fairly soft interiors and it's not like I will be throwing this thing around, anyway.
One thing I would buy as an additional extra is a second power pack. It's useful to have your home setup permanently "wired" with the second supply in the bag.
Step 1) Leak their source
Step 2) Sue Onen Source developers down the road because obviously they have studied the MS leaked source.
Step 3)... Ya, I'm sure you know what goes here.
Ok but seriously, I'm not touching it. The last thing I need is Microsoft saying that I somehow owe something to them.
I see a lot of people saying this - even PJ@Groklaw - but I don't think it's realistic.
If I read the FreeBSD source code then I'm not obligated to make my contributions to Linux under the BSD license.
If I work for Sun writing Solaris drivers then I'm not "tainted" for life from working on Linux. And yes, all Solaris developers receive a full copy of the source code.
Merely seeing code doesn't mean you can't work on other software. Though it does make it more difficult to prove your innocence, if the issue of code theft ever comes up. But if your code is obviously different from the Microsoft code then what is the problem? I can't imagine that merely viewing Microsoft code would mean that Microsoft owns you for the rest of your programming life. Otherwise noone could ever move between software companies, and that happens all the time!
For example, Digital doesn't own Windows just because Dave Cutler jumped ship. Borland doesn't own Microsoft Visual C++ just because Microsoft stole half the Borland developers. If seeing code was such a big deal then the entire industry is already corrupted. Why does it suddenly matter if Microsoft's code is involved?
Admittedly, as I said earlier, it's harder to prove innocence if you have looked at the code. But surely it's not cut-and-dry "you looked, you forfeit all your own code". Surely not.
Slashdot Mirror
I agree with your interpretation. Anderer is saying that litigation has become commonplace in the IT industry. Some of these lawsuits (eg, Eolas vs Microsoft) might result in judgements on the order of $500 million. He believes Microsoft is defending themselves against 50 such suits right now and he doesn't think open-source companies would have the cash reserves to defend against similar nuisance lawsuits (esp. w.r.t patents). If open-source was to dominate the landscape then the industry would be decimated in mere months by the sheer number of $500 million judgements against developers.
He is basically saying "this is a game for the big boys and you open-sourcers are too puny to play in this arena, so go home and stop bothering us". Arrogant, true. But possibly correct. I don't personally believe his argument but I can see where he's coming from.
Of course, the common interpretation is that Anderer is threatening open-source companies. The last line is being intepreted to mean Microsoft is initiating 50 more lawsuits like SCO vs IBM. I don't buy that argument because Anderer would have to be stupid to voluntarily tell everybody about the conspiracy. Unfortunately because certain prominent people are stating that particular interpretation as if it were fact, all other interpretations are being ignored or shouted down.
An article doesn't have unlimited space. At some point they have to draw the line, cut some words to come under the limit, and call it a day. But I can try and answer some of your questions.
Printer support recently got very good. By installing CUPS, GS-ESP, HPIJS and FOOMATIC you receive high-quality printer drivers for several 1000 popular printers. By the way, I'm telling you those cryptic names for your own education, but the typical user just sees the "Printer" icon on their desktop. CUPS also has support for automatically detecting and installing USB and Plug and Play Parallel Port printers.
CDROMs these days all use the ATAPI standard, so there is no problem with "unknown brands". The bad old days of proprietary CDROM interfaces are long gone (and good riddance).
Ethernet is ridiculously well supported under Linux. Most Ethernet vendors seem to provide Linux drivers from their websites. Most of them also provide source code which gets quickly bundled into the next version of Linux.
The special keys on Windows keyboards have worked for a long time now (at least 5 years, probably more). The multimedia keys on modern keyboards also work automatically. KDE and GNOME both have graphical setup programs to allow the user to assign functions to those keys.
SpeedStep works beautifully with an interface for programmers exposed by the SYSFS module (you don't need to know that as a user). This SYSFS interface is standardised across all processors, not just Intel's SpeedStep. So the CPU frequency scaling also works on PowerPC and AMD chips. You as a user don't need to know this; any decent Linux distro will automatically install a daemon like "cpufreqd" which freq-steps the CPU as needed.
Yes, of course, though support is spotty. I don't have hibernation on my laptop (iBook G4) but earlier versions of the iBook work (iBook G3). The Intel laptops usually have better luck.
That's no good. So it's not perfect. Nobody ever claimed it was. Maybe it will get better in the next release.
I have had good success plugging in Firewire disks and CDROM drives and having them automatically work. USB keys, mice and keyboards were also automatic. PCMCIA is also automatic though driver support is spotty. It can always identify the device, but whether it's supported is another matter.
You do realise that many people consider Linux to be better than Windows? I suppose if you come from a Windows background you might think Linux is a little "crappy" because it doesn't support your video card, or doesn't hibernate correctly on your laptop. But consider that some of us come from a UNIX background and we think Windows is crappy because... well... it isn't like UNIX!
Always bear in mind, if you're looking at Linux as a free version of Windows, you will always be disappointed. If you're looking at Linux as a free version of UNIX, you will probably be pleasantly surprised. If you're looking at Linux with no preconceptions then, in my limited experience with such people, you will be happy with the results.
Your reasoning is correct but one niggly fact was wrong. You talk about OSs only being sold by the hardware manufacturer and you say "Microsoft changed that". But CP/M predates MS-DOS by a very long time and it was available on many personal computers from many vendors. You could even get CP/M for the Commodore 128.
While I agree with you that divorcing hardware from the software was important for the growth of the IT industry, Microsoft wasn't the first company to do it. Even UNIX could be properly seen as divorcing hardware from software; you could run UNIX on dozens of different minis (not PCs) well before Microsoft even existed.
These aren't robots to keep them company. They're robots to keep them clean and robots to help them with everyday tasks. Get a grip on the situation and get off your soapbox.
I suppose in your world we shouldn't allow old people to drive cars because then they'd be lonely. Instead, we should have 6 fit young men carrying each old person around on a litter.
Instead of impersonal cooking machines, like microwaves, we can just hire teams of people to breathe heavily on the food until it's cooked. Heavens forbid that old people use a "technical solution" to cook their food.
I wonder what you might say if I explained to you the concept of a phone; a "technical solution" that allows for *greater* human interaction. Probably your head would explode as you tried to reconcile the paradox.
Stop being such a drongo. Robots to keep the elderly clean is a great thing. It means they'll receive better care, at an affordable price, and they can clean themselves when they want to rather than when the overworked nurse is available. A nurse, by the way, who could actually improve the quality of their patients lives if they weren't wasting their valuable time giving sponge baths.
It's often amazing how otherwise intelligent people can be unbelievably stupid.
For example, I have a friend who got top marks in Computer Science, went on to do Honours, and he is now doing his Masters in his spare time. He's one of the brightest people I know. He can code like a natural, can design like a professional, and holds down an important and well paid job as a J2EE designer.
One day he was describing to me how a PIII running Linux is way better than a Sun E10k. I thought he was joking at first. No, he was being serious. Apparently he'd ran the same database software and J2EE application on both systems and the Linux system smoked the E10k in performance.
After a little prodding I find out his domain on the E10k was only 4 CPUs and was memory starved.
After a little more prodding I find out that the database software they were using was single-threaded.
Yet pointing these facts out didn't convince him then of his mistake, and to this day he is convinced that E10ks are gutless wonders (which is true but not for the reasons he thinks).
So I learnt that day that anybody can make a mistake. The important lesson to learn is that the idiot might be yourself. I make an effort these days to be more humble when someone tells me I'm wrong; they might just be right.
The EV1 deal is interesting.
SCO is claiming the deal with EV1 was worth more than a million dollars. EV1 is disputing the 7 figures and the confusion seems to be the weasel word "worth".
In other words, SCO is claiming that $1million+ "worth" of licenses were sold. So that's $1mill/$699 = 1400+ licenses, or $1mill/$1399 = 700+ licenses. SCO's own quarterly says only $20k income from licensing this quarter. It's possible the EV1 payments are in stages, or won't appear on SCO's financials until next quarter, but it's also possible that EV1 only paid $20k for their licenses.
But SCO is spinning this to sound much more impressive. EV1 was the patsy here; they thought they were getting a great deal, but they were just another pawn in SCO's (Microsoft's?) smear campaign against Linux.
Zero. The SCO executives simply wait for a Linux developer to screw in a lightbulb, then SCO claims ownership of the house.
Yes, I am aware of that, but my point was to illustrate that "resource allocation" is an inappropriate way to describe patents. Ideas are not a resource to be allocated.
They already evalute each patent on a case-by-case basis, so I hardly think it would be unworkable.
I've always wondered why Apple (and Microsoft for that matter) haven't partnered with major application vendors to supply application patches and updates through the Software Update feature.
I can imagine some of the reasons - too hard, too many variables, not our problem - but surely if the free software guys can work it out, it's not an insurmountable problem.
I bought an iBook and tried OSX for the heck of it.
After a short period of discomfort I deleted OSX and installed Linux.
Now I'm happy.
My point: not everybody that tries MacOSX is immediately smitten with it.
Utterly wrong.
The downloaders are making a copy. They send the request to the server, initiating the copy and the transfer. The downloaders are responsible for making the copy. If they make a copy of copyrighted material where the license doesn't permit copying, then they have infringed on copyright.
The downloader could perhaps argue that they weren't aware the copyrighted material was licensed in such a way to prevent copies being made from the webserver. Depending on the circumstances, they might even convince a judge of their innocence. But it doesn't seem likely if the filename is "Windows2000SourceCode.zip" in the "warez/0" directory.
If you need a real world analogy to help you understand this point. If you buy a television from the back of a truck, and the asking price is 1/10th what it costs in the store, but you can see that it's second hand, and the seller wants cash-only, and he won't give you a receipt, then you might have difficulty explaining to the local magistrate that you didn't realise the goods were stolen.
You sound exactly like Karl Marx ;-)
He argued that the master-slave society could only give way to communism once there were sufficient resources to sustain everybody's basic needs.
He also said that the "haves" would fight to prevent the "have-nots" from sharing in the wealth. Looking at the world around us today, it is like Karl Marx had a window into the future.
I disagree with a lot of what Karl Marx wrote - I think he ignores the fundamental problem of human greed, as well as the ever-increasing lower limit of our basic "needs" - but it's always a treat to read The Communist Manifesto and realise how right he was about so many things. Marx has a a better hit-rate for his predictions than Nostradamus!
The problem is that ideas aren't a finite resource to be allocated like wheat or coal or milk. Ideas can be duplicated without effort. I'm reminded of this quote:
I'd also like to express my distate for the "if you don't love the USA then go to Russia" style of argument. It's intellectually bankrupt.
That would be OK if the invention was actually innovative such that the 20 years was justified. Unfortunately a number of patents are granted that are completely without merit - they are either unoriginal, obvious, or trivial - and as such they are undeserving of a 20 year monopoly.
I think the best reform for patents would be variable periods for the government granted monopoly. Say, between 18 months and 20 years, decided by merit on a case-by-case basis.
It was actually only 78 studios, but some of those companies had 8x speed burners.
Uhh, MacOS, GEM, GeOS, AmigaOS. Need I continue?
DOS/3.1 "won" because it had the right apps and came at the right price.
This is the exact same reason why Linux will be 90% of the market in 10 years. Assuming it survives SCO.
In the past 10 years I have spent over $400 on Linux distributions and $1500 on Linux software.
That's just for personal use. I have spent easily 10x that for business purposes.
I only wish he was right. I don't know if I'm the only one, but I DETEST Internet gaming. All the online games are full of cheats and weenies and fu><0r j00 d00d morons. The servers are always full, or down, or slow, or whatever. But it's never a smooth experience. I'm constantly frustrated when trying to connect. Plus with half the online games charging a monthly service, it's simply too expensive for a casual gamer like me.
I'd like to think I could just say "pfeh, I don't play them, but they don't bother me". But it's not true. Internet games are sucking up all the developer resources. Every second game has Internet connectivity. That's wasted effort on a feature I'll never use, but I still have to pay for.
I do like LAN games. LAN parties are great. I also like party games (ie, 4 of you crowd around the same telly). But Internet gaming? No thanks. Been there. Done that. Wasn't interested in the t-shirt.
Don't be silly. The entire EMACS code won't fit into a mere 200 gig of compressed source.
Even worse, the very few games there are have fallen badly into disrepair.
I bought pretty much every game released by Hyperion, Loki, Sirtech and Vicarious. Nearly 2 dozen titles. Half of them don't work on a modern (Debian/unstable) x86 installation. I found 5 of the Loki games were ported to PPC; only 1 of those still works (Heroes 3). This isn't a Linux specific problem. The few DOS games I own (eg, X-Wing) don't work on any modern Microsoft platform.
To contrast, the open-source commercial games (Doom, Quake, Quake2, Starcon2) work flawlessly on all the Linux platforms I've tried. Just goes to show that with enough interest, any game can stay ported to the latest platform, but you need source code!
For Linux gaming I've decided to giveup on closed source. Either the game comes with source - I'll still pay for it - or I'll just stick with what I've got. The open source games aren't very good (with a few notable exceptions like nethack) but I refuse to purchase another closed source game that lasts less than a year before bit-rotting.
PS: I actually get a fair bit of mileage out of emulators (eg, dosbox). But I know that won't appeal to most gamers.
Sounds like you're both winning.
Though if it's hurting that much, maybe you should do it less often.
You don't need to recompile Windows to change its behaviour. You just need to install certain binaries. For example, the GINA authentication DLLs, or the LiteStep explorer replacement.
And viruses don't need to recompile a binary to infect it. They simply modify the PE header then piggyback at the end of the file. This was the case before Windows even existed.
I disagree that it's "easier" to verify that binaries haven't been tampered with. Microsoft releases many versions of their DLLs and they do not provide MD5 sums. You could compare against a known good source but how would you know if the difference was because of deceit or because of an upgrade? You'd need access to all valid versions of Windows, including every intermediate version, including non-Microsoft versions that are bundled with third party software (eg, DirectX betas are often bundled with third party games, but similar bundles occur for non-gaming software).
Microsoft tried using digital signatures as an improvement over MD5 sums, but their vetting processes were flawed. You might remember the ActiveX control that was signed despite its only action being to reboot the computer. Microsoft revoked the signature but only after the author brought their attention to it. Who says that there aren't trojans in existing ActiveX controls? They don't even have to be intentional; the flaws in Microsoft code used by viruses aren't intentional.
Finally, malware already exists despite there being no access to Windows source code. Scripting languages in Windows are powerful enough to cause the same disruption as corrupted binaries. So a perfect set of uncorrupted binaries proves nothing.
I hear what you're saying but I disagree that it makes any practical difference. The example the author provided - a rogue contractor delivering trojaned software - can happen with closed source. The author claimed that open-source makes trojans easier to deploy, but I see no evidence to support such an outlandish claim, and my own understanding of the issues leads me to disagree with such claims.
You need to trust your staff and your vendors. That is the bottom line.
"Let's say Vendor X gets a contract to provide a government agency with 800 desktop computers, with Windows, Office, etc. Meeting a bunch of carefully written specs from that agency's IT department. Vendor X takes Windows XP and customizes it, complete with a "Foo Agency" splash screen, encrypted disk partitions, escrowed bypass for crypto, etc.
"How do we know they didn't plant malware in Windows? What geeks will have access to this binary? Geeks won't even know this mini-distro exists. "
The problem with your example, and with the article that preceded this thread, is that it discusses problems that are common to both open and closed source. The real question is "how can we trust contractors to not screw us". Blaming open-source is disingenuous.
Whenever I use Windows I find it a frustrating experience, having to deal with obscure registry settings and drivers and service packs.
Comparing FreeBSD to Linux is like comparing a Toyota Corolla to a V8 engine. Try comparing FreeBSD to a distribution like Suse, Red Hat or Mandrake.
My experience is that FreeBSD is no better or worse than any of the community driven distros like Gentoo or Debian. Seeing as the majority of userspace is the same (XFree86, OpenOffice and GNOME) that's really no surprise. It's strange to claim FreeBSD is "designed" whereas Linux is not, because most of the software in FreeBSD is accumulated in exactly the same way that it is accumulated in every Linux distribution.
In my opinion... don't buy many peripherals. The problem with buying lots of peripherals is that your ~2kg laptop suddenly needs a bag the size of a small suitcase weighing in at 10kg or more to carry around all the junk you "need".
Take the mouse, for example. You've already decided to get one. Why? The trackpad really is not that bad. The mouse needs a flat surface to work on, which ties you to a desk. I've found I can work the trackpad fairly well, after just a few days practise.
Cables? Why? If you're going to be in a place where you need a cable, there's likely to be one there already. Take for example people who carry a serial cable around so they can "plug-in" to a Cisco router while onsite. Have you ever seen a comms room without a serial cable? I haven't. So why carry the cable? Same for Ethernet. If you simply *must* have a backup cable, put it in the boot of your car.
Extra batteries are only useful if you keep them charged. In my experience, nobody is that organised. They'll have two batteries and one of them is always flat. So when their live battery goes flat... well... they're in the same boat as the rest of us with only one battery.
I managed to avoid getting a computer bag for the laptop. A leather briefcase costs the same and looks heaps more professional. Modern cases have fairly soft interiors and it's not like I will be throwing this thing around, anyway.
One thing I would buy as an additional extra is a second power pack. It's useful to have your home setup permanently "wired" with the second supply in the bag.
And a USB key. Damn, those things are so useful!
I see a lot of people saying this - even PJ@Groklaw - but I don't think it's realistic.
If I read the FreeBSD source code then I'm not obligated to make my contributions to Linux under the BSD license.
If I work for Sun writing Solaris drivers then I'm not "tainted" for life from working on Linux. And yes, all Solaris developers receive a full copy of the source code.
Merely seeing code doesn't mean you can't work on other software. Though it does make it more difficult to prove your innocence, if the issue of code theft ever comes up. But if your code is obviously different from the Microsoft code then what is the problem? I can't imagine that merely viewing Microsoft code would mean that Microsoft owns you for the rest of your programming life. Otherwise noone could ever move between software companies, and that happens all the time!
For example, Digital doesn't own Windows just because Dave Cutler jumped ship. Borland doesn't own Microsoft Visual C++ just because Microsoft stole half the Borland developers. If seeing code was such a big deal then the entire industry is already corrupted. Why does it suddenly matter if Microsoft's code is involved?
Admittedly, as I said earlier, it's harder to prove innocence if you have looked at the code. But surely it's not cut-and-dry "you looked, you forfeit all your own code". Surely not.