Wrong fix. It still leaves you open to injection attacks. Best solution is to get rid of all the hooks web pages (whether through Javascript or not) have to open windows without decoration.
In short, Microsoft finally got something right, and those that are in IT security for the sole reason of bashing MS to make a buck, are having a hard time doing so.
Microsoft hasn't fixed the underlying problem, which is that a web browser or a component used by a web browser has no business providing a mechanism by which a web page can even request the execution of a downloaded native-code applet or scripts with local file access. That capability should not even be in the HTML display control.
That way it wouldn't make a difference what "zone" a page is in, there wouldn't be an exploit to take advantage of.
Firefox's XPI installation mechanism and Safari's "open safe files after downloading" are examples of capabilities that are completely unnecessary and have the potential of being used to open similar holes. They're about two or three decimal orders of magnitude less dangerous, but are still things that no browser should implement.
However, for their professions, there are certain programs that only run in Windows and they have to use Windows.
I'm not sure what your point is, here. Have a look at the message I'm responding to. The poster is arguing that switching to Intel eliminates the demand for Macintosh applications because people will be able to run Windows applications under emulation at full native speed, and so will be entirely satisfied with the Windows version. Not only that, but he's assuming that people will be satisfied by rebooting into Windows just to run an application!
So, I'm talking about whether people who buy a Macintosh are satisfied by running a Windows application rather than a native Macintosh version. Of course they aren't... the fact that the Windows application runs faster still doesn't make it the equal of a native Macintosh application, because many of the advantages of Mac OS X over Windows XP are only available within native applications.
You're talking about whether it's to the user's benefit to be able to run Windows applications on their Mac when Mac versions aren't available. Of course it is, but if there's a Mac application or a Windows application that are otherwise comparable they'll still buy the one that's available for the Mac over the Windows one. In fact, I'm pretty confident that they'll choose the Mac version even if the Windows version has a longer feature list so long as the Mac version satisfies their requirements.
Apple on Intel means that Windows applications can run via emulation at full native speed.
If all I wanted to do was run Windows applications at full native speed, I can buy a Windows box to do that right now. It's CHEAPER to buy a Windows box. Even if you already have a Mac it's cheaper. Virtual PC costs more than a refurb Compaq, a KVM, and a copy of Windows XP Home.
If people running OS X were happy running Windows applications instead of OS X applications, they'd already be running Windows applications. If they have to run a Windows app they'll do it, but, sheesh, when I'm using Camino - a native port of the Gecko engine to OS X - and it's obvious when I'm in the Gecko engine rather than the native wrapper because stuff that I use all the time, like contextual menu plugins and many services, don't work. They do in Safari.
I can't imagine Windows apps being anything but steerage-class citizens under OS X.
I've never seen a post from anyone espousing the paid-for distros.
I've bought Suse, FreeBSD retail box, and I've got half a dozen Red Hat Enterprise systems getting ready to go out the door right now.
I've also bought separate copies of Jaguar for each of the Beige Powermacs I've installed it on, even though Apple doesn't support them and I had to use XPostFacto to get it to run. Right now OSX is my favorite commercial distor based on Free UNIX by far.
I'm not going to get into OS X Intel until it's been around for a while (I've got a bad case of dotzerophobia). I may buy a copy of Leopard and try and hack it to run on a Thinkpad just because I hate the Powerbook and iBook physical hardware... but if Apple produces a laptop with a decent keyboard, two mouse buttons, and a high resolution display that might be my first Intel Mac.
Without the snarkiness, this is one of the concerns I have about the Apple switch. Microsoft has already done a lot of work to make Windows Media Player as unsniffable as they can, so the only way to take WMV and de-DRM it is through the analog hole, and they've been pushing to make this something that goes all the way down to the hardware (Palladium, whatever Intel denies they're doing with Pentium D, etc...).
It has been alleged that the reason for this switch is so Apple can show a strong DRM technology that'll get Hollywood online for an iTunes Video Store. I think it's a bit unlikely, but Steve's the master of the knight's move, so I wouldn't rule anything out.
People, remember the issue of drivers is not like we are dealing with a totally new subsystem now. It's just darwin. Drivers that are relased for a BSD (darwin, in our case) are by default out for OS X, or, for that matter, any driver out for mostly any *NIX.
This is why I have been concerned about how long Apple will continue Darwin releases.
how many linux programs (for example) do you know that really give a sh** if you compile them on PPC SPARC x86 or Alpha?
PPC, Sparc, or x86... no sweat.
Alpha? You would be AMAZED how much software is unhappy if you run it in LP64. That's why Microsoft defaults to L32P64 in Win64, and why Tiger doesn't default to 64-bit mode. Unless your processor vendor used the word size change as an excuse to fix problems in the instruction set (say, a really stupidly small register file) you're really a lot safer in 32-bit until you REALLY need the extra address space.
If the music started out CD quality I'd probably use HYMN, but my ears aren't golden enough to detect the difference between the audio CD version and the protected AAC version, so audio CD backups are good enough for me.
Thank you for reminding us that limited copyright and patent terms force companies and individuals to have to keep moving forwards, to get their profits from new material, not build a company that lives by sitting on a golden goose for decade after decade until long after it's obvious to everyone else that they don't really have anything innovative or new of their own any more.
I think we're a long way from that. What people are demonstrating now is self-repair, using largely complete components, and fabrication of macro-scale objects. Devices that can replicate electronic and especially digital circuits to a fine enough precision that second or third generation machines actually work? I don't see that any time soon.
OK, I've had a few minor slips, I've flirted with easily breakable protection schemes, but my computers have all been free and clear of strong DRM for a quarter of a century now.
The top entry on the darknet site is a link to another blog and that writer's experience with Adobe's digital rights management. That really struck home for me, because it was a similar experience in the early '80s that led me to studiously avoid strong digital rights management for longer, I suspect, than the term has been around.
Almost 25 years ago I found myself huddled in a corner of a computer lab at the University of Houston with one of the local pirate software geeks getting a pirated copy of the game "Wizardry" copied over the legal version on my original floppy. Why? Because the copy protection on the diskette was so aggressive that after saving a game on a slightly mistimed floppy drive, once, it would only ever allow me to play the game using that same drive... and of course a drive whose timing has started to drift is not long for this world.
Wizardry was the last copy-protected game I bought. But, of course, that was no big deal, it was only a game. I don't need to play videogames. If I have to, I can write them myself, so I'm not jonesing for them at all.
Copy protection, of course, is a form of digital rights management, and that one bad experience immunized me to the appeal of DRMed software. Any DRMed software, because while it's no big deal to find yourself missing a videogame it's a much bigger problem if your OS depends on some kind of acid test.
More recently, I've had to email Apple and get them to clear all my authorized computers from their database at the iTunes Music Store because a flakey hard drive and a series of system reinstalls had caused me to exceed my maximum allowed authorizations... and you can't "deauthorize this computer" when "this computer" is now a cloud of quantum states no longer entangled with any physical hardware. This wasn't a big deal, most of my music is stuff I've ripped myself (got the CDs right here) and I had audio-CD backups of most of the tracks I'd bought BECAUSE I didn't trust DRMed content... and this sure reinforced that mistrust nicely.
I've since made ABSOLUTELY sure I've backup copies of my music on audio CDs before I considered any purchase "complete". No big deal, as long as I can do that... and besides it's only music, I can listen to the radio. But I'm going to make sure that any DRMed content I buy in the future has a backdoor... at least through the "analog hole" (so you can keep your watermarked music, Jack, I'll learn to play the piano before I buy any of that stuff).
But getting back to software... I refused to upgrade to Windows XP, because an operating system with DRM in its heart is way outside my comfort zone me, and it was the prospect of having to do so that finally got me to pay the "Mac Tax" and buy a Macintosh instead of dual-booting between FreeBSD and Windows... so I guess I should thank Microsoft for giving me the necessary nudge to make the switch.
The rumors I've heard about Apple using some kind of hardware DRM from Intel to keep people from running Mac OS X for Intel on generic clones really bother me: if the alternatives are a DRMed operating system from Microsoft or a DRMed operating system from Apple or giving up on popular commercial software completely, I guess I'll ride the Power PC bus until the wheels fall off and see if the world's come to its senses five years or so down the road. I don't really believe Steve Jobs is stupid enough to do that, the nudge-nudge-wink-wink protection in iTunes and Cory Doctorow's comments about the differences between his discussions with Microsoft and Apple are encouraging.
So, anyway, avoiding software and media with strong DRM hasn't been any great hardship, so far. But it's early day's yet... with luck I've got at least another quarter century of avoiding DRM ahead of me. I'm confident I can do it, one day at a time.
many Windows-only applications, including Windows-based games, utilities, and business applications, will operate seamlessly and reliably
That would be a great incentive for people to switch, because in my experience many Windows-only applications do not operate seamlessly and reliably under Windows.
If you can point to an actual statement that the patents would be available on a royalty free basis, then there is something called estopel.
Read the pages I linked to.
Microsoft can say "these patents are available on a royalty free basis for any vendor that uses our libraries, we will license these libraries to anyone, at no cost (which is all 'royalty-free' means), to everyone under identical (thus non-discriminatory) terms, using the 'Happy Freeware' license", and then ship the libraries with a license that's incompatible with open source.
They haven't lied. They've done what they said. But they've still got you by the short and curlies if you don't want to spend time in court or unless you don't want to ship source. AND they've tried to play basically this game with the EU. Several of us predicted this... and we got the same kind of reaction to that THEN as you're having to me NOW. THIS time we seem to have a European regulator who doesn't buy in to the idea that these are "reasonable terms", but you can't depend on that... they've gotten away with worse elsewhere.
Did you notice that just about any software whatsoever, including Free software, is subject to patent claims
Yes, but there's a big difference here. Here, we know ahead of time EXACTLY what the claims are, because those claims are on patents that are explicitly an essential part of what's being implemented. There's no "innocent infringement" here.
Who is going to start the patent nuclear war?
There doesn't need to be a "nuclear war". A patent "cold war" will serve Microsoft just as well: they have effectively fought a clandestine war on many fronts for decades, from "DOS isn't done until Lotus won't run" through the very effective file format battles that lead to Office's dominance on the desktop, and facing down the Department of Justice over IE. Patents are just another tool for them.
Microsoft is already demanding royalties and receiving payments for embedded use of the FAT file system. They're establishing a strong presence in a relatively inconspicuous corner of the market for their patent portfolio. They have strongarmed free software they didn't approve of, like ntcrashme, out of the limelight... and hardly made a ripple. Thinking they won't continue to expand this precedent is trusting to their good will.
With Mono, I could, for example, use GTK#, but not with.NET.
Whatever direction.NET goes, Mono will follow, unless the Mono group are willing to abandon compatibility with Windows... something I find about as likely as Samba abandoning compatibility with Windows.
If a software developer, or any other potential inventor, does realize what they've done is something new, then it likely what they did was pretty obvious.
I assume you mean does not realize.
And, no, that doesn't follow. In writing software, if you do it right, very little of what you do is obvious. Anything that you do that's obvious gets bundled up in a procedure or a template or a class or a library so you don't have to do it again. Then you go on to do something else that's non-obvious so you can turn it into something you don't have to do again. The bits that aren't new are the dull and boring bits that you get over with as quickly as possible, or you figure out a way to bundle a bunch of them in a new invention (a new class, a new procedure, a new library) so you don't have to do them again.
Or you find someone else who's done it and you use their code... in a commercial library, a text book, a posting on the net, or an open source package.
So for any programmer, a significant part of his time writing the software is simply the process of creating one potentially patentable process after another. They're not normally patented because stopping to see if every new subroutine, template, or package is really unique would increase the cost of programming a thousand fold.
This is where the ease of manufacturing comes in. In a typical physical invention, there may be half a dozen patentable ideas, which then take months to turn into something that can be manufactured, and then you spend more months and lots of money making them. Spending a couple of weeks in the middle to check on half a dozen ideas to see if they're patented is a small part of the cost.
A computer program, though, you spend months creating thousands of unique procedures (if they're not unique, you've probably missed an opportunity to make them unique, or create a tool for building them out of templates), debugging them (and creating more of them in the process), testing them (and the testbed is itself more software, more potentially patentable ideas), and documenting them (if the documents weren't being developed in parallel). Then you spend weeks arranging for printing and reproduction.
If you were to do a patent search on every one of those unique pieces of code, it would take years.
But throwing out software patents means companies with large cash reserves can outright kill small companies with truly unique ideas through mimicry.
Those large cash reserves mean large patent portfolios. If I'm a small company who's trying to sell a product, and I sue a big company because they stole my patent on the Wakalix Interface, they'll turn around and say "OK, you'll license that to us for free and we'll license the Bogus Browser you're using on your main window for free, is that a deal?"
Unless the small company isn't actually producing a product, they're in a Mexican Standoff. Because the small company's got a thousand potential patents in their application, they've taken out half a dozen because they can afford to, but the big company's got thousands of patents and they're violating a dozen of them and they didn't even know it.
I mean, I wrote an anti-spam program. A guy just suggested a feature to me, and I see a clever way to implement it that would take me half an hour to code and post on my website. It may be a patentable idea, I don't know, I haven't seen exactly the same idea used anywhere, and certainly not in anti-spam software, and it's certainly as different from anything I *have* seen as many of the physical process patents I've read are different from the stuff they're building on.
If I implement the feature, I've spent half an hour. I've probably spent more time reading and replying to this message. If I were to take a patent out on it, it'd cost me thousands of dollars and would still probably be worthless. But, Microsoft might come out with anti-spam software that uses it and then I could shake them down for millions... OR the
That final post in the page made me pause also. But it still sounds like a problem in the system, specifically in the enforcement during trial.
I'm not sure I understand what you're getting at. The problem is that pure software patents are inherently like "patenting Boyle's Law". If they are allowed, then when it gets to trial what's going to happen? "All my client did was implement a mathematical operation." "So you're admitting it, then?"
And in any case... for any open source programmer, and most small businesses, if it gets as far as a trial they've already lost.
And because the process of creating software is ALL design... the manufacturing component is negligable... it inherently involves many orders of magnitude more opportunities for violating patents than a physical object or even a business process with the same kind of time and effort invested in it.
I'd love tgo switch to NeoOffice or Pages or another decednt alternative to Word on the OS X platform, but they all ignore a feature of vital importance to professional writers like myself: A halfway decent word count function.
Yeh, that's why I'm sticking with "vi". Being able to go ":,'aw !wc" or "!}fmt" or "!/^From/-1sort" instead of having to write a script or cut-and-paste from a whole other document is a killer feature.
OK, OK, I'm a retro old fogey. I won't tease you any more (but I really do prefer to work in "vi" than GUI apps)... but this isn't entirely irrelevant either... because Mac OS X has a really neat capability that many people don't fully "get", that gives you a lot of this kind of easy integration between applications that made UNIX such a revolutionary environment back in the '70s (and, goldarnit, still does today).
In Pages, you can do a word count on the whole document, but not on a highlighted selection.
Does the Services menu not work in Pages? I'm sure it does. In any Cocoa app and many Carbon ones you can select text and perform operations on it through "Application -> Services", and any application can publish a service in the Services menu. There's a freeware application called "WordService" that provides among other things a "statistics" dialog that gives you characters, spaces, words, and lines in the selected text.
I don't know how well NeoOffice/J supports services, but Pages is Apple's own application so I would be positively astounded if it fell short in this area. This kind of deep integration between applications is what really makes Mac OS X a joy for me. You don't need to provide every feature over and over again in every application, you just need to provide a way for programs to work together.
Did you notice that Mono supports numerious languages?
Did you notice that Mono is primarily an implementation of.NET, and thus subject to patent claims by Microsoft? Microsoft has stated that these patents will be avilable on a "royalty free and otherwise reasonable and non-discriminatory basis", but short of an irrevocable legally-binding release worded in such a way that it's unambiguously clear these patents can not be used against open-source software, I am unwilling to trust to their good will.
Statements like "Furthermore, our release of the Rotor source code base with a specific license on its use gives wide use to our patents for a particular (non-commercial) purpose, and as we explicitly state we are open to additional licenses for other purposes." -- Microsoft applies for.NET patent are less than encouraging.
In Europe, Microsoft are already showing their true colors: "If developers want to build the protocols into their products, they must agree not to distribute that product in source-code form, or to subject it to licenses that require source-code disclosure, a formula that excludes many open source licenses."
And they have other tricks up their sleeve, "At every release the focus of Microsoft's tools that provide a compelling Linux development environment could break or prevent mono-compatibility the same way Microsoft's J++ broke Java compatibility by replacing JavaBeans, RMI,and JNI with COM, DCOM, Direct/J. At that point, you would face the choice of either forking the API's or forking over some royalty payments." --Mono developer meeting
With Mono you can hitch your wagon to Microsoft's oxen, never knowing just where they're going to go.
20 separate flash applets...
on
The Onion in 2056
·
· Score: 4, Insightful
I can literally hear my computer huffing and puffing to keep up with this page, because the temperature-controlled fan turned on as they rendered.
You're right, current usage of these words means that I should have said "typefaces" rather than "fonts". But that's a side issue: the copyright on the font isn't relevant to the patent on Truetype rendering. I could create a true type font and release it into the public domain and it would still be necessary to use the algorithm covered by this patent to properly render the typeface embodied in the font.
That is, this is a patent that restricts the typeface itself. You can only approximate way the characters are supposed to look - the typeface - without using the technique covered by this patent. That's why this particular software patent is more disturbing than most.
fonts on Linux are a bit "blurry"...that is, they are not as clear/crisp as their those on their windows counterparts.
According to this it's a patent issue. I think there's something deeply wrong with patents on operations required to render fonts correctly, above and beyond the already troubling issue of software patents in general. Remember that in the US fonts are explicitly not copyrightable to prevent even the potential of copyright being used to prevent free speech. Shouldn't this easement be extended to any communication or presentation technology.
Slashdot Mirror
Wrong fix. It still leaves you open to injection attacks. Best solution is to get rid of all the hooks web pages (whether through Javascript or not) have to open windows without decoration.
In short, Microsoft finally got something right, and those that are in IT security for the sole reason of bashing MS to make a buck, are having a hard time doing so.
Microsoft hasn't fixed the underlying problem, which is that a web browser or a component used by a web browser has no business providing a mechanism by which a web page can even request the execution of a downloaded native-code applet or scripts with local file access. That capability should not even be in the HTML display control.
That way it wouldn't make a difference what "zone" a page is in, there wouldn't be an exploit to take advantage of.
Firefox's XPI installation mechanism and Safari's "open safe files after downloading" are examples of capabilities that are completely unnecessary and have the potential of being used to open similar holes. They're about two or three decimal orders of magnitude less dangerous, but are still things that no browser should implement.
However, for their professions, there are certain programs that only run in Windows and they have to use Windows.
I'm not sure what your point is, here. Have a look at the message I'm responding to. The poster is arguing that switching to Intel eliminates the demand for Macintosh applications because people will be able to run Windows applications under emulation at full native speed, and so will be entirely satisfied with the Windows version. Not only that, but he's assuming that people will be satisfied by rebooting into Windows just to run an application!
So, I'm talking about whether people who buy a Macintosh are satisfied by running a Windows application rather than a native Macintosh version. Of course they aren't... the fact that the Windows application runs faster still doesn't make it the equal of a native Macintosh application, because many of the advantages of Mac OS X over Windows XP are only available within native applications.
You're talking about whether it's to the user's benefit to be able to run Windows applications on their Mac when Mac versions aren't available. Of course it is, but if there's a Mac application or a Windows application that are otherwise comparable they'll still buy the one that's available for the Mac over the Windows one. In fact, I'm pretty confident that they'll choose the Mac version even if the Windows version has a longer feature list so long as the Mac version satisfies their requirements.
Since I'll have to send them to the people using Hotmail I correspond with.
I dumped my own Hotmail account when they started requiring a Passport account to use it.
Apple on Intel means that Windows applications can run via emulation at full native speed.
If all I wanted to do was run Windows applications at full native speed, I can buy a Windows box to do that right now. It's CHEAPER to buy a Windows box. Even if you already have a Mac it's cheaper. Virtual PC costs more than a refurb Compaq, a KVM, and a copy of Windows XP Home.
If people running OS X were happy running Windows applications instead of OS X applications, they'd already be running Windows applications. If they have to run a Windows app they'll do it, but, sheesh, when I'm using Camino - a native port of the Gecko engine to OS X - and it's obvious when I'm in the Gecko engine rather than the native wrapper because stuff that I use all the time, like contextual menu plugins and many services, don't work. They do in Safari.
I can't imagine Windows apps being anything but steerage-class citizens under OS X.
I've never seen a post from anyone espousing the paid-for distros.
I've bought Suse, FreeBSD retail box, and I've got half a dozen Red Hat Enterprise systems getting ready to go out the door right now.
I've also bought separate copies of Jaguar for each of the Beige Powermacs I've installed it on, even though Apple doesn't support them and I had to use XPostFacto to get it to run. Right now OSX is my favorite commercial distor based on Free UNIX by far.
I'm not going to get into OS X Intel until it's been around for a while (I've got a bad case of dotzerophobia). I may buy a copy of Leopard and try and hack it to run on a Thinkpad just because I hate the Powerbook and iBook physical hardware... but if Apple produces a laptop with a decent keyboard, two mouse buttons, and a high resolution display that might be my first Intel Mac.
lower cost items like video cards, controllers, etc. that don't have to have special firmware for PowerPC platform.
Maybe. Maybe not. Apple has all kinds of extensions to OpenGL that don't show up on the non-Apple versions of their cards.
Without the snarkiness, this is one of the concerns I have about the Apple switch. Microsoft has already done a lot of work to make Windows Media Player as unsniffable as they can, so the only way to take WMV and de-DRM it is through the analog hole, and they've been pushing to make this something that goes all the way down to the hardware (Palladium, whatever Intel denies they're doing with Pentium D, etc...).
It has been alleged that the reason for this switch is so Apple can show a strong DRM technology that'll get Hollywood online for an iTunes Video Store. I think it's a bit unlikely, but Steve's the master of the knight's move, so I wouldn't rule anything out.
People, remember the issue of drivers is not like we are dealing with a totally new subsystem now. It's just darwin. Drivers that are relased for a BSD (darwin, in our case) are by default out for OS X, or, for that matter, any driver out for mostly any *NIX.
This is why I have been concerned about how long Apple will continue Darwin releases.
how many linux programs (for example) do you know that really give a sh** if you compile them on PPC SPARC x86 or Alpha?
PPC, Sparc, or x86... no sweat.
Alpha? You would be AMAZED how much software is unhappy if you run it in LP64. That's why Microsoft defaults to L32P64 in Win64, and why Tiger doesn't default to 64-bit mode. Unless your processor vendor used the word size change as an excuse to fix problems in the instruction set (say, a really stupidly small register file) you're really a lot safer in 32-bit until you REALLY need the extra address space.
If the music started out CD quality I'd probably use HYMN, but my ears aren't golden enough to detect the difference between the audio CD version and the protected AAC version, so audio CD backups are good enough for me.
Thank you for reminding us that limited copyright and patent terms force companies and individuals to have to keep moving forwards, to get their profits from new material, not build a company that lives by sitting on a golden goose for decade after decade until long after it's obvious to everyone else that they don't really have anything innovative or new of their own any more.
open source personal self-replicating fabricators
I think we're a long way from that. What people are demonstrating now is self-repair, using largely complete components, and fabrication of macro-scale objects. Devices that can replicate electronic and especially digital circuits to a fine enough precision that second or third generation machines actually work? I don't see that any time soon.
OK, I've had a few minor slips, I've flirted with easily breakable protection schemes, but my computers have all been free and clear of strong DRM for a quarter of a century now.
The top entry on the darknet site is a link to another blog and that writer's experience with Adobe's digital rights management. That really struck home for me, because it was a similar experience in the early '80s that led me to studiously avoid strong digital rights management for longer, I suspect, than the term has been around.
Almost 25 years ago I found myself huddled in a corner of a computer lab at the University of Houston with one of the local pirate software geeks getting a pirated copy of the game "Wizardry" copied over the legal version on my original floppy. Why? Because the copy protection on the diskette was so aggressive that after saving a game on a slightly mistimed floppy drive, once, it would only ever allow me to play the game using that same drive... and of course a drive whose timing has started to drift is not long for this world.
Wizardry was the last copy-protected game I bought. But, of course, that was no big deal, it was only a game. I don't need to play videogames. If I have to, I can write them myself, so I'm not jonesing for them at all.
Copy protection, of course, is a form of digital rights management, and that one bad experience immunized me to the appeal of DRMed software. Any DRMed software, because while it's no big deal to find yourself missing a videogame it's a much bigger problem if your OS depends on some kind of acid test.
More recently, I've had to email Apple and get them to clear all my authorized computers from their database at the iTunes Music Store because a flakey hard drive and a series of system reinstalls had caused me to exceed my maximum allowed authorizations... and you can't "deauthorize this computer" when "this computer" is now a cloud of quantum states no longer entangled with any physical hardware. This wasn't a big deal, most of my music is stuff I've ripped myself (got the CDs right here) and I had audio-CD backups of most of the tracks I'd bought BECAUSE I didn't trust DRMed content... and this sure reinforced that mistrust nicely.
I've since made ABSOLUTELY sure I've backup copies of my music on audio CDs before I considered any purchase "complete". No big deal, as long as I can do that... and besides it's only music, I can listen to the radio. But I'm going to make sure that any DRMed content I buy in the future has a backdoor... at least through the "analog hole" (so you can keep your watermarked music, Jack, I'll learn to play the piano before I buy any of that stuff).
But getting back to software... I refused to upgrade to Windows XP, because an operating system with DRM in its heart is way outside my comfort zone me, and it was the prospect of having to do so that finally got me to pay the "Mac Tax" and buy a Macintosh instead of dual-booting between FreeBSD and Windows... so I guess I should thank Microsoft for giving me the necessary nudge to make the switch.
The rumors I've heard about Apple using some kind of hardware DRM from Intel to keep people from running Mac OS X for Intel on generic clones really bother me: if the alternatives are a DRMed operating system from Microsoft or a DRMed operating system from Apple or giving up on popular commercial software completely, I guess I'll ride the Power PC bus until the wheels fall off and see if the world's come to its senses five years or so down the road. I don't really believe Steve Jobs is stupid enough to do that, the nudge-nudge-wink-wink protection in iTunes and Cory Doctorow's comments about the differences between his discussions with Microsoft and Apple are encouraging.
So, anyway, avoiding software and media with strong DRM hasn't been any great hardship, so far. But it's early day's yet... with luck I've got at least another quarter century of avoiding DRM ahead of me. I'm confident I can do it, one day at a time.
many Windows-only applications, including Windows-based games, utilities, and business applications, will operate seamlessly and reliably
That would be a great incentive for people to switch, because in my experience many Windows-only applications do not operate seamlessly and reliably under Windows.
This is not a joke.
Microsoft has no good will to trust.
.NET.
.NET goes, Mono will follow, unless the Mono group are willing to abandon compatibility with Windows... something I find about as likely as Samba abandoning compatibility with Windows.
No kidding.
If you can point to an actual statement that the patents would be available on a royalty free basis, then there is something called estopel.
Read the pages I linked to.
Microsoft can say "these patents are available on a royalty free basis for any vendor that uses our libraries, we will license these libraries to anyone, at no cost (which is all 'royalty-free' means), to everyone under identical (thus non-discriminatory) terms, using the 'Happy Freeware' license", and then ship the libraries with a license that's incompatible with open source.
They haven't lied. They've done what they said. But they've still got you by the short and curlies if you don't want to spend time in court or unless you don't want to ship source. AND they've tried to play basically this game with the EU. Several of us predicted this... and we got the same kind of reaction to that THEN as you're having to me NOW. THIS time we seem to have a European regulator who doesn't buy in to the idea that these are "reasonable terms", but you can't depend on that... they've gotten away with worse elsewhere.
Did you notice that just about any software whatsoever, including Free software, is subject to patent claims
Yes, but there's a big difference here. Here, we know ahead of time EXACTLY what the claims are, because those claims are on patents that are explicitly an essential part of what's being implemented. There's no "innocent infringement" here.
Who is going to start the patent nuclear war?
There doesn't need to be a "nuclear war". A patent "cold war" will serve Microsoft just as well: they have effectively fought a clandestine war on many fronts for decades, from "DOS isn't done until Lotus won't run" through the very effective file format battles that lead to Office's dominance on the desktop, and facing down the Department of Justice over IE. Patents are just another tool for them.
Microsoft is already demanding royalties and receiving payments for embedded use of the FAT file system. They're establishing a strong presence in a relatively inconspicuous corner of the market for their patent portfolio. They have strongarmed free software they didn't approve of, like ntcrashme, out of the limelight... and hardly made a ripple. Thinking they won't continue to expand this precedent is trusting to their good will.
With Mono, I could, for example, use GTK#, but not with
Whatever direction
If a software developer, or any other potential inventor, does realize what they've done is something new, then it likely what they did was pretty obvious.
I assume you mean does not realize.
And, no, that doesn't follow. In writing software, if you do it right, very little of what you do is obvious. Anything that you do that's obvious gets bundled up in a procedure or a template or a class or a library so you don't have to do it again. Then you go on to do something else that's non-obvious so you can turn it into something you don't have to do again. The bits that aren't new are the dull and boring bits that you get over with as quickly as possible, or you figure out a way to bundle a bunch of them in a new invention (a new class, a new procedure, a new library) so you don't have to do them again.
Or you find someone else who's done it and you use their code... in a commercial library, a text book, a posting on the net, or an open source package.
So for any programmer, a significant part of his time writing the software is simply the process of creating one potentially patentable process after another. They're not normally patented because stopping to see if every new subroutine, template, or package is really unique would increase the cost of programming a thousand fold.
This is where the ease of manufacturing comes in. In a typical physical invention, there may be half a dozen patentable ideas, which then take months to turn into something that can be manufactured, and then you spend more months and lots of money making them. Spending a couple of weeks in the middle to check on half a dozen ideas to see if they're patented is a small part of the cost.
A computer program, though, you spend months creating thousands of unique procedures (if they're not unique, you've probably missed an opportunity to make them unique, or create a tool for building them out of templates), debugging them (and creating more of them in the process), testing them (and the testbed is itself more software, more potentially patentable ideas), and documenting them (if the documents weren't being developed in parallel). Then you spend weeks arranging for printing and reproduction.
If you were to do a patent search on every one of those unique pieces of code, it would take years.
But throwing out software patents means companies with large cash reserves can outright kill small companies with truly unique ideas through mimicry.
Those large cash reserves mean large patent portfolios. If I'm a small company who's trying to sell a product, and I sue a big company because they stole my patent on the Wakalix Interface, they'll turn around and say "OK, you'll license that to us for free and we'll license the Bogus Browser you're using on your main window for free, is that a deal?"
Unless the small company isn't actually producing a product, they're in a Mexican Standoff. Because the small company's got a thousand potential patents in their application, they've taken out half a dozen because they can afford to, but the big company's got thousands of patents and they're violating a dozen of them and they didn't even know it.
I mean, I wrote an anti-spam program. A guy just suggested a feature to me, and I see a clever way to implement it that would take me half an hour to code and post on my website. It may be a patentable idea, I don't know, I haven't seen exactly the same idea used anywhere, and certainly not in anti-spam software, and it's certainly as different from anything I *have* seen as many of the physical process patents I've read are different from the stuff they're building on.
If I implement the feature, I've spent half an hour. I've probably spent more time reading and replying to this message. If I were to take a patent out on it, it'd cost me thousands of dollars and would still probably be worthless. But, Microsoft might come out with anti-spam software that uses it and then I could shake them down for millions... OR the
That final post in the page made me pause also. But it still sounds like a problem in the system, specifically in the enforcement during trial.
I'm not sure I understand what you're getting at. The problem is that pure software patents are inherently like "patenting Boyle's Law". If they are allowed, then when it gets to trial what's going to happen? "All my client did was implement a mathematical operation." "So you're admitting it, then?"
And in any case... for any open source programmer, and most small businesses, if it gets as far as a trial they've already lost.
And because the process of creating software is ALL design... the manufacturing component is negligable... it inherently involves many orders of magnitude more opportunities for violating patents than a physical object or even a business process with the same kind of time and effort invested in it.
I'd love tgo switch to NeoOffice or Pages or another decednt alternative to Word on the OS X platform, but they all ignore a feature of vital importance to professional writers like myself: A halfway decent word count function.
Yeh, that's why I'm sticking with "vi". Being able to go ":,'aw !wc" or "!}fmt" or "!/^From/-1sort" instead of having to write a script or cut-and-paste from a whole other document is a killer feature.
OK, OK, I'm a retro old fogey. I won't tease you any more (but I really do prefer to work in "vi" than GUI apps)... but this isn't entirely irrelevant either... because Mac OS X has a really neat capability that many people don't fully "get", that gives you a lot of this kind of easy integration between applications that made UNIX such a revolutionary environment back in the '70s (and, goldarnit, still does today).
In Pages, you can do a word count on the whole document, but not on a highlighted selection.
Does the Services menu not work in Pages? I'm sure it does. In any Cocoa app and many Carbon ones you can select text and perform operations on it through "Application -> Services", and any application can publish a service in the Services menu. There's a freeware application called "WordService" that provides among other things a "statistics" dialog that gives you characters, spaces, words, and lines in the selected text.
I don't know how well NeoOffice/J supports services, but Pages is Apple's own application so I would be positively astounded if it fell short in this area. This kind of deep integration between applications is what really makes Mac OS X a joy for me. You don't need to provide every feature over and over again in every application, you just need to provide a way for programs to work together.
NeoOffice/J IS WRITTEN IN JAVA with some Carbon for native Mac goodies. What the heck do you think the "/J" stands for?
It stands for "we're using Java for the user interface". NeoOffice is based on the OpenOffice.org source code, which is written in C++.
Did you notice that Mono supports numerious languages?
.NET, and thus subject to patent claims by Microsoft? Microsoft has stated that these patents will be avilable on a "royalty free and otherwise reasonable and non-discriminatory basis", but short of an irrevocable legally-binding release worded in such a way that it's unambiguously clear these patents can not be used against open-source software, I am unwilling to trust to their good will.
.NET patent are less than encouraging.
Did you notice that Mono is primarily an implementation of
Statements like "Furthermore, our release of the Rotor source code base with a specific license on its use gives wide use to our patents for a particular (non-commercial) purpose, and as we explicitly state we are open to additional licenses for other purposes." -- Microsoft applies for
In Europe, Microsoft are already showing their true colors: "If developers want to build the protocols into their products, they must agree not to distribute that product in source-code form, or to subject it to licenses that require source-code disclosure, a formula that excludes many open source licenses."
And they have other tricks up their sleeve, "At every release the focus of Microsoft's tools that provide a compelling Linux development environment could break or prevent mono-compatibility the same way Microsoft's J++ broke Java compatibility by replacing JavaBeans, RMI,and JNI with COM, DCOM, Direct/J. At that point, you would face the choice of either forking the API's or forking over some royalty payments." --Mono developer meeting
With Mono you can hitch your wagon to Microsoft's oxen, never knowing just where they're going to go.
I can literally hear my computer huffing and puffing to keep up with this page, because the temperature-controlled fan turned on as they rendered.
You're right, current usage of these words means that I should have said "typefaces" rather than "fonts". But that's a side issue: the copyright on the font isn't relevant to the patent on Truetype rendering. I could create a true type font and release it into the public domain and it would still be necessary to use the algorithm covered by this patent to properly render the typeface embodied in the font.
That is, this is a patent that restricts the typeface itself. You can only approximate way the characters are supposed to look - the typeface - without using the technique covered by this patent. That's why this particular software patent is more disturbing than most.
Ideally, a common language and runtime that all free software could target would be available that would allow immediate porting to take place.
Having it be a language that doesn't suck dirty swamp water through used oil filters would be a pleasant bonus, but that doesn't seem to be an option.
fonts on Linux are a bit "blurry"...that is, they are not as clear/crisp as their those on their windows counterparts.
According to this it's a patent issue. I think there's something deeply wrong with patents on operations required to render fonts correctly, above and beyond the already troubling issue of software patents in general. Remember that in the US fonts are explicitly not copyrightable to prevent even the potential of copyright being used to prevent free speech. Shouldn't this easement be extended to any communication or presentation technology.