It's been par for the course for Linux support to require some final hurdle, in this case reverse engineering the music database.
Yeh, and it's been par for the course for Linux support of a product to require a completely new effort for each version of some product. I remember when 3COM ethernet support just started getting good, and 3COM kept screwing with the drivers buy shipping multiple different chipsets under the same model name. Sometimes these hurdles just keep coming back.
When that hurdle is ALSO a hurdle that the company's using to keep competitors (like Real) out of their pants, then you gotta expect it's going to get moved on you.
In general, this applies to an organisation that begins by entirely adhering to standards too - they could suddenly decide all updates from today rely on non-standard extensions unless they've made some sort of promise to the contrary at the time of sale *very rare*.
I think we're talking about different things here. I'm not just talking about *using* standards, I'm talking about *supporting* them.
I'm talking about the commitment a company has to make to maintaining compatibility with published APIs. Apple has a reputation for breaking APIs when they think they need to, to move forward, but even they very rarely drop an API (if it's an open systems API) or even make a major change in it (if it's a proprietary one) without months of lead time, and if it's a major API they'll keep it alive for years. When they even look like they're not following this principle, they get roasted.
And they have major customers and developers who REALLY drag their feet on catching up with API changes. They came up with a timeframe for abandoning the classic Mac OS API in 1997 and Adobe kicked and bucked and they ended up dropping all their plans for Rhapsody and going back to the drawing board, gave classic Mac OS a new lease on life with carbon and a whole major release that they hadn't expected to need, which meant that a practical Mac OS X didn't come out until 2002 with Jaguar. Then with the Intel shift Adobe was late to the party again.
And Adobe isn't weird, they're just more aggressive about it.
So, if Apple had *published* an API for loading music onto the iPod, or officially supported some third open standard for doing so, then they would have to maintain that API for a while. they would have to announce when they were going to quit supporting the open API, or when they were going to quit supporting applications using their published API.
Yes, they really would. Because it would have an effect on people who weren't going into the dance by printing up their own tickets.
But since they didn't, it doesn't matter... the only customer of that API is Apple.
So they could decide that the iPod was going to partition the flash into a UMASS file system and a fixed size sqlite database, with the songs stored in that database, and suing HTTPS with a certificate hidden in iTunes to install songs into it. Since they don't support UMASS for loading music into the iPod, they could do that.
On the other hand, you can bet that the iPod won't suddenly require that your MPs3 be branded by Apple to play on the iPod, because that's an open standard the iPod actually supports, and people would scream from Seattle, WA to Washington, DC about it.
I like OS X - specifically, I like the ObjC frameworks that came from NeXT (I'm a sucker for Smalltalk too;-)
Having worked for the goverment, I'm simply sickened by the billions of pounds that are wasted on incompetence, ignorance and stupidity every single year.
I've been a Windows system admin, so I can sympathize.
That was not Apple's decision, and they have continued releasing Darwin updates since then.
macosforge has seen no posts since November 2006
I know, and it's annoying that this seems to have been a PR response to the last time this came up... but on the other hand it may have been a matter of testing the waters. Remember, that means nothing from Apple *and* nothing from the rest of the open source community either. If they were testing to see if the FOSS community would respond in kind, well, I think that's failed.
requests on the darwin mailing list have been responded with "it's ready when it's ready"
If this continues long after Leopard is out I'll be complaining too, but Leopard is a big push-up for them... both because of the iPhone and because of the problems exposed in developer releases. Just like the last time this happened. Frankly, I was honestly surprised to see them put up ANY Intel releases at all, I had expected them to quietly drop the open source effort then given the obvious advantage it gave to people porting OS X to non-apple platforms.
A delay, as around the time of the initial Tiger Intel release, is not the same as a complete dismantling of support structure.
The only part of the support structure that was both under Apple's control and was around back then that you've indicated is missing are the build instructions. Opendarwin was not Apple's, and macosforge came out after the LAST brouhaha.
By all means yell at Apple if things don't improve after the Leopard release. But while they're not keeping up with your expectations they're still doing about as much as they've ever done.
The point is that it's nicer as far as developer productivity and uniform user experience goes to be able to use published APIs than to have to write your own custom hack.
I'm not talking about a new API. I'm talking about maintaining an API that Apple has removed, like Unsanity did with Menu Extra Enabler. Unsanity could release an Input Manager Enabler that maintained the same API the same way. I wold be surprised if they didn't.
iPods play MP3, which while not open is almost as-good-as
MP3 is an open format, with patent encumbrances, yes. But then AAC is an open format as well - it's just MP4 audio. The only non-open music format on the iPod is Fairplay. But that's not what I'm talking about.
The iPod plays open format music, but you can't put that music on it using an open systems interface, and it doesn't have an open API. THAT is why it's not an open device. For a Linux user, it's not open in any way that matters if you actually want to use it as a music player. Even if you can get in through a backdoor that backdoor depends on undocumented features so it can be closed anytime without warning.
I never implied I was "buying an iPhone";
This is a general question, though. I didn't ask "why are you buying...", I asked why ANY Linux user would buy any iPod or the iPhone. For the Linux user they're clearly a really bad choice of device, and yet many have obviously done so despite Apple repeatedly demonstrating that they consider these devices to be appliances and that they have no intention of supporting any third party music software in any way.
I'm illustrating what Apple would have to do in order to gain this geek's appreciation and custom.
Apple has many geek's appreciation and custom. Apple even has many Linux Geek's custom! My question is why they've got Linux geeks appreciation and custom. It's like people buying Mac desktops and notebooks to run Linux on. Apple's hardware isn't anything to get excited about. It's got lousy ergonomics and its design is all about style... not functionality. If you want a Linux notebook the Thinkpad is a far better choice.
I like what Apple create
Why? The only thing that Apple has created that I particularly like is O
What little OS X malware there has been seems to feature them more often than not.
What OS X malware are you talking about?
There's been one OS X malware release in the wild, and that was a social engineering attack over AIM. The only protection against social engineering exploits is user education. You can't solve them through hardening the OS or applications (though you CAN avoid them by not training people to answer affirmatively to routine security dialogs by having as many of them as Microsoft does. Alas Apple has started down the same slippery slope.
The biggest categories of vulnerabilities exposed in OS X so far are buffer overflows (which everyone is subject to) and attacks through LaunchServices, half of them using Safari's daft 'Open "Safe" files' option. These are also two of the three biggest categories of vulnerabilities in Windows (the third, and probably bigger than either, is ActiveX embedding in the HTML control). InputManager plugins don't even rate.
If Apple was concerned with security, they would:
* Remove 'Open "safe" files after downloading' and improve the Safari download manager to force attackers to use social engineering to get people to launch attacks on secondary applications for them.
* Provide an API for applications to register as "safe applications" for web browsers and other applications that display untrusted documents. Applications opening URIs, downloaded documents, and viewers would check in *that* database instead of the standard LaunchServices database.
Microsoft needs to take these actions, too. I don't think any UNIX desktops provide the APIs that Windows and Mac browsers use to allow these attacks... apps have to explicitly add themselves to the browser's own MIME and extensions registry. If they do, they need to take the same kind of actions.
Mac OS X is going completely closed source - 10.4.9 was the last open sourced release of the base kernel/BSD toolset
That's what people were saying the last time Apple was slow getting an update out, during the Tiger release.
If Apple doesn't support InputManager plugins, Unsanity or someone will hack them back in. Apple can't stop that because OS X is not an embedded platform, it's a general purpose operating system, based on an open systems platform, with powerful debugging tools. Unless they completely redesign it from the kernel on up along the lines of Vista that's not going to change.
But none of that applies to the iPhone/iPod!
Apple's embedded devices have NEVER been either open source or open systems. Why on earth are you buying an iPod or an iPhone if you care about that kind of thing?
iTunes is an *application* that runs on BOTH Windows and OS X. It does this checksumming on BOTH platforms. This is nothing to do with OS X. This is all in an application (iTunes) and some embedded software (on the iPod). If Apple used Linux instead of OS X on the Mac it STILL wouldn't make any difference, because this is not something that the operating system is doing.
This is like asking why Red Hat didn't include scarlet fedoras in their boxed copies of RHL. It's a non-sequitur.
Whatever organization you work with, you're dancing with the devil. It doesn't matter whether it's a big company or a cooperative or a neighborhood association or the Free Software Foundation or Red Hat. As soon as enough people get together they become a new kind of animal, one made of paper and power, contracts and by-laws. You can't *not* dance with the devil, you just have to be aware of what tools the devil has to get a hold on you.
Don't let yourself get locked in.
With OSX, that's a lot easier to avoid than with Windows. You can treat your Mac as a UNIX box with some extra applications, and you can shift to Linux or FreeBSD or Solaris if you have to. That's what open systems get you... the ability to make yourself independent of what any particular implementation of that system does... because none of the players controls the system. Not the FSF, Apple, the FreeBSD project, or Sun.
With Windows, for a while it looked like Microsoft were going to let you live the UNIX life on their platform too, with Interix. But no, that's going to be for Vista *server* only. I'd been using Interix on Windows as my UNIX-on-NT solution, but because I wasn't dependent on it I could drop it. No, don't talk to me about Cygwin (or about Wine).
With the iPod, though, you're dealing with hardware that's completely locked in. If you want to avoid dancing with Microsoft or Apple for your OS, WHY IN THE HELL WOULD YOU BUY A PROPRIETARY MP3 PLAYER?
That's what I don't get. Know where the devil you're dancing with has his hooks, whether they're proprietary extensions to C or a proprietary MP3 player or proprietary APIs, and don't lead the dance there. That's just daft.
The *only* advantage the iPod has from this point of view is that you can get accessories from it anywhere. You can't get docks and cradles and cases and plugin doodads for your Creative Whatever or Noname Thumbdrive, or if you do you won't be able to use them on the next model. But if you're going with Linux because you don't want to dance with Apple or Microsoft... that's the price you pay.
Because that's the biggest iPod lock-in. So here's MY one more thing... a suggestion for one of the devil's to encourage people to dance.
If Creative really wanted to take over the market from Apple this is what they'd need to do:
1. Support unencrypted AAC. Yes, the license costs a buck or so more than Microsoft's WMA, but it'll make you iTunes-plus compatible... that may not really turn out to be that useful, but it's great marketing. Add ogg and whatever Sony's format is, if it's cheap.
2. Come up with a layout of ports (power, USB, audio) that people can build adapters to, and make a binding promise not to change it for five years. It doesn't have to be a proprietary dock... it can be a fixed geometry of the power, mini-USB, and audio sockets... but it has to be something that allows a positive connection that people can just plug in to.
3. Consider giving people a development API. Consider making that an open-systems one. So people can *add* ogg if they want. Think of it as free software support, for not much cost and with automatic renewal.
I thought the iPod worked like any other USB drive - I don't need iTunes to put music on my thumb drive, I just mount it on the file system and transfer the files.
No, the iPod doesn't work this way. It keeps the files in a special part of the file system and even if you put the files there it needs special tools to make it work.
I don't understand why anyone using Linux would bother with an iPod. I found it a horribly frustrating MP3 player because of the annoying user interface and daft click wheel... AND it costs more! Why bother with it?
How long until there's an open source category for "open ipod" software on freshmeat and sourceforge?
Since this isn't a phone, Apple's not gonna have legal and contractual reasons to keep the iPod Touch locked. If they're smart they'll treat this like the Apple TV and offer at most a token defense to keep from getting "enhanced" ipod support calls.
"We're talking about a low-power device with very limited programming capacity"
We're talking about a device with (based on the iPhone's specs) 128MB RAM, UNIX, 8-16 gigabytes of mass storage, and a 600 MHz ARM processor. It may be a little less than that, but not much... one of the most CPU intensive apps on the iPhone is Safari, and that's on the iPod Touch as well.
I've run a popular web server with three 1000 member mailing lists, one of which was high volume for the time, plus two MUDs... all on a 486/50 with 5MB of RAM and 110MB hard disk. That's less CPU power, disk, and probably RAM than the original iPod, and not much more than my five year old Clie SJ22. Heck 10 years ago Apple's top product was the Beige G3. That's a G3/233, 64M RAM, and I don't recall how big a hard disk but it wasn't measured in gigabytes... and people were running servers on that.
As for the software, I've written a web server as a 20 line shell script. You don't need Bonjour, or a real FTP server, or a compiler... HTTP right to the IP address of the iPod would work fine.
You don't understand the Everett-Wheeler-Graham multiple-universe model of quantum mechanics.
In the EWG model the collapse of the state vector is an illusion, a mathematical simplification of reality. The state vector never collapses, so in our experience the universe splits as a result of every quantum "collapse", an infinite number of new universes born every instant of planck time. The illusion of a single universe is a result of our existence as an entanglement in the state vector of the universe.
So the whole question of whether our alter-egos are "linked" is an illusion. They are us, as we would be if one of billions of subatomic interactions had gone slightly differently in any of infinitely many ways. We already "prune the universe" by observing only one possible outcome at a time out of the infinite possibilities. Consciously doing so is merely a matter of taking our fate into our own hands.
The user is processing images. The ad, which is enabled in the software, suggests photo development services of several clients.
How precisely is this NOT adware? The user hasn't asked Google or Microsoft or anyone else for photo processing services. The user may not have any intention of asking for information. The user is simply being interrupted in the middle of working with another program with an ad.
interrupted
Google has never interrupted me. No popups. Not even any interstitial ads.
in another program
I've never had Google pop up when I'm on someone else's web site.
in the middle of working
Never had Google come up unless I've actually asked them to.
All it is, an idea, not all that different from the targeted advertising provided by a certain search engine slashbots seem so quick to defend against all claims.
If Google was using pop-up ads...
If Google was popping up those ads over the web pages you found on Google...
If Google was charging $300 for access to their service...
Then you can bet your last dollar that they'd get attacked. But they're not getting attacked for doing something similar to what this patent is proposing because, get this, they're NOT DOING IT.
Usually, format doesn't matter, and only the words matter. Sometimes, however, it does. If it does, you want to know that as soon as possible.
Now if you make ODF your standard, your sort of screwed as to displaying that word 6 document exactly as word 6 did anyways, so what have you really lost by simply marking it as word 6 style in the Word xml?
If you make ODF your standard, and some aspect of Word 6 layout turns out to matter, you will know that the first time you look at the document after the transition to ODF... even if you're still using Word.
If you make OOXML your standard, and some aspect of Word 6 layout turns out to matter, you will never know unless you ALSO switch from Word, because Word will hide that from you.
So just don't save your stuff with the crazy format tags and you should be safe.
"You" is a secretary working on a legal document and saving a file in Word. He has no idea what XML tags are, even if there's an option in Word for "don't save your stuff with the crazy format tags".
And then a decade later something in some document doesn't line up right because it depended on some aspect of Word 6 table layout, and someone's words are attributed to the wrong person...
Browsers lay out HTML differently from one another.
HTML is a structural language, like Docbook.
OOXML and ODF wouldn't be needed if all we were looking at was document structure. These are more detailed markup languages where maintaining the appearance of the content matters, because sometimes meaning actually does depend on layout... even in ASCII, which is why people used to spend so much time making their terminals bug-compatible with the vt100.
It may be that these tags don't actually specify anything that changes the layout of a document enough to matter, and should actually be treated as a comment by any other program, but until Microsoft specifies them we don't know that.
Does anyone still have mission critical documents in word 6 that depend on the exact formating Word 6 had?
The main reason these standards are a "big deal" right now are incoming legal requirements for maintaining archived documents in a format that can still be read when nobody's using the software they were created under any more. So when they depend on said software to figure out what the document looks like it kind of defeats the purpose.
You're pointing at a timeline that starts in 1998, and references HTML output from Word as part of the history of OOXML. I guess HTML's SGML-subset design gave them some experience with *ML, but HTML operates at a completely different level. HTML is a bit of a mongrel, but its main thrust is structure, not layout.
A comparable ODF timeline might go back as far as 1991 with the release of Docbook 1.0 by O'Reilly.
Whatever Microsoft might have been working on while they were a member of Oasis was just catch-up at best.
I said that ECMA was going to document that for the next batch of issues to resolve in the spec.
See, here's the problem. If elements that are obviously going to need to be defined aren't defined then the spec is a long way from finished. This isn't just a matter of the spec not defining things you think should be in there, like ODF, this is a matter of the spec not defining stuff that's actually in the spec itself.
This is like a book proposal that's got things like "Chapter 11: (Add fight scene between goodguy and badguy) Chapter 12: The next morning my hand still looked like a shredded tire under the bandages, but I didn't have time to lie in bed..." in it. A publisher might accept that as a draft, especially if you've got a good track record, but they won't publish it like that.
A spec that's a long way from being finished was just pushed almost all the way through a fast-track process to jump it straight from this stage to "published", which means that Microsoft wants it to be treated as a finished spec... not a draft. Even if it's the best draft in the world, it's still a draft.
So do you mean that it's a good draft, or that it's actually a good standard? Everyone else is judging it as if it was complete, because that's how Microsoft is presenting it. Is that how you're approaching it?
Slashdot Mirror
It's been par for the course for Linux support to require some final hurdle, in this case reverse engineering the music database.
;-)
Yeh, and it's been par for the course for Linux support of a product to require a completely new effort for each version of some product. I remember when 3COM ethernet support just started getting good, and 3COM kept screwing with the drivers buy shipping multiple different chipsets under the same model name. Sometimes these hurdles just keep coming back.
When that hurdle is ALSO a hurdle that the company's using to keep competitors (like Real) out of their pants, then you gotta expect it's going to get moved on you.
In general, this applies to an organisation that begins by entirely adhering to standards too - they could suddenly decide all updates from today rely on non-standard extensions unless they've made some sort of promise to the contrary at the time of sale *very rare*.
I think we're talking about different things here. I'm not just talking about *using* standards, I'm talking about *supporting* them.
I'm talking about the commitment a company has to make to maintaining compatibility with published APIs. Apple has a reputation for breaking APIs when they think they need to, to move forward, but even they very rarely drop an API (if it's an open systems API) or even make a major change in it (if it's a proprietary one) without months of lead time, and if it's a major API they'll keep it alive for years. When they even look like they're not following this principle, they get roasted.
And they have major customers and developers who REALLY drag their feet on catching up with API changes. They came up with a timeframe for abandoning the classic Mac OS API in 1997 and Adobe kicked and bucked and they ended up dropping all their plans for Rhapsody and going back to the drawing board, gave classic Mac OS a new lease on life with carbon and a whole major release that they hadn't expected to need, which meant that a practical Mac OS X didn't come out until 2002 with Jaguar. Then with the Intel shift Adobe was late to the party again.
And Adobe isn't weird, they're just more aggressive about it.
So, if Apple had *published* an API for loading music onto the iPod, or officially supported some third open standard for doing so, then they would have to maintain that API for a while. they would have to announce when they were going to quit supporting the open API, or when they were going to quit supporting applications using their published API.
Yes, they really would. Because it would have an effect on people who weren't going into the dance by printing up their own tickets.
But since they didn't, it doesn't matter... the only customer of that API is Apple.
So they could decide that the iPod was going to partition the flash into a UMASS file system and a fixed size sqlite database, with the songs stored in that database, and suing HTTPS with a certificate hidden in iTunes to install songs into it. Since they don't support UMASS for loading music into the iPod, they could do that.
On the other hand, you can bet that the iPod won't suddenly require that your MPs3 be branded by Apple to play on the iPod, because that's an open standard the iPod actually supports, and people would scream from Seattle, WA to Washington, DC about it.
I like OS X - specifically, I like the ObjC frameworks that came from NeXT (I'm a sucker for Smalltalk too
How do you feel about GNUstep?
Having worked for the goverment, I'm simply sickened by the billions of pounds that are wasted on incompetence, ignorance and stupidity every single year.
I've been a Windows system admin, so I can sympathize.
the OpenDarwin project has been shut down
... both because of the iPhone and because of the problems exposed in developer releases. Just like the last time this happened. Frankly, I was honestly surprised to see them put up ANY Intel releases at all, I had expected them to quietly drop the open source effort then given the obvious advantage it gave to people porting OS X to non-apple platforms.
That was not Apple's decision, and they have continued releasing Darwin updates since then.
macosforge has seen no posts since November 2006
I know, and it's annoying that this seems to have been a PR response to the last time this came up... but on the other hand it may have been a matter of testing the waters. Remember, that means nothing from Apple *and* nothing from the rest of the open source community either. If they were testing to see if the FOSS community would respond in kind, well, I think that's failed.
requests on the darwin mailing list have been responded with "it's ready when it's ready"
If this continues long after Leopard is out I'll be complaining too, but Leopard is a big push-up for them
A delay, as around the time of the initial Tiger Intel release, is not the same as a complete dismantling of support structure.
The only part of the support structure that was both under Apple's control and was around back then that you've indicated is missing are the build instructions. Opendarwin was not Apple's, and macosforge came out after the LAST brouhaha.
By all means yell at Apple if things don't improve after the Leopard release. But while they're not keeping up with your expectations they're still doing about as much as they've ever done.
The point is that it's nicer as far as developer productivity and uniform user experience goes to be able to use published APIs than to have to write your own custom hack.
I'm not talking about a new API. I'm talking about maintaining an API that Apple has removed, like Unsanity did with Menu Extra Enabler. Unsanity could release an Input Manager Enabler that maintained the same API the same way. I wold be surprised if they didn't.
iPods play MP3, which while not open is almost as-good-as
MP3 is an open format, with patent encumbrances, yes. But then AAC is an open format as well - it's just MP4 audio. The only non-open music format on the iPod is Fairplay. But that's not what I'm talking about.
The iPod plays open format music, but you can't put that music on it using an open systems interface, and it doesn't have an open API. THAT is why it's not an open device. For a Linux user, it's not open in any way that matters if you actually want to use it as a music player. Even if you can get in through a backdoor that backdoor depends on undocumented features so it can be closed anytime without warning.
I never implied I was "buying an iPhone";
This is a general question, though. I didn't ask "why are you buying...", I asked why ANY Linux user would buy any iPod or the iPhone. For the Linux user they're clearly a really bad choice of device, and yet many have obviously done so despite Apple repeatedly demonstrating that they consider these devices to be appliances and that they have no intention of supporting any third party music software in any way.
I'm illustrating what Apple would have to do in order to gain this geek's appreciation and custom.
Apple has many geek's appreciation and custom. Apple even has many Linux Geek's custom! My question is why they've got Linux geeks appreciation and custom. It's like people buying Mac desktops and notebooks to run Linux on. Apple's hardware isn't anything to get excited about. It's got lousy ergonomics and its design is all about style... not functionality. If you want a Linux notebook the Thinkpad is a far better choice.
I like what Apple create
Why? The only thing that Apple has created that I particularly like is O
Obviously I agree with you. Why not put your name on such a well thought out post?
What little OS X malware there has been seems to feature them more often than not.
What OS X malware are you talking about?
There's been one OS X malware release in the wild, and that was a social engineering attack over AIM. The only protection against social engineering exploits is user education. You can't solve them through hardening the OS or applications (though you CAN avoid them by not training people to answer affirmatively to routine security dialogs by having as many of them as Microsoft does. Alas Apple has started down the same slippery slope.
The biggest categories of vulnerabilities exposed in OS X so far are buffer overflows (which everyone is subject to) and attacks through LaunchServices, half of them using Safari's daft 'Open "Safe" files' option. These are also two of the three biggest categories of vulnerabilities in Windows (the third, and probably bigger than either, is ActiveX embedding in the HTML control). InputManager plugins don't even rate.
If Apple was concerned with security, they would:
* Remove 'Open "safe" files after downloading' and improve the Safari download manager to force attackers to use social engineering to get people to launch attacks on secondary applications for them.
* Provide an API for applications to register as "safe applications" for web browsers and other applications that display untrusted documents. Applications opening URIs, downloaded documents, and viewers would check in *that* database instead of the standard LaunchServices database.
Microsoft needs to take these actions, too. I don't think any UNIX desktops provide the APIs that Windows and Mac browsers use to allow these attacks... apps have to explicitly add themselves to the browser's own MIME and extensions registry. If they do, they need to take the same kind of actions.
iPod US price does not include sales tax.
iPod UK price surely includes 17.5% VAT.
£150 + 17.5% = £176.25 (13% difference)
£200 + 17.5% = £235.00 (11% difference)
Mac OS X is going completely closed source - 10.4.9 was the last open sourced release of the base kernel/BSD toolset
That's what people were saying the last time Apple was slow getting an update out, during the Tiger release.
If Apple doesn't support InputManager plugins, Unsanity or someone will hack them back in. Apple can't stop that because OS X is not an embedded platform, it's a general purpose operating system, based on an open systems platform, with powerful debugging tools. Unless they completely redesign it from the kernel on up along the lines of Vista that's not going to change.
But none of that applies to the iPhone/iPod!
Apple's embedded devices have NEVER been either open source or open systems. Why on earth are you buying an iPod or an iPhone if you care about that kind of thing?
Google doesn't need windows, they just need browsers, html and http, and their linux servers.
none of which was developed on windows.
iTunes is an *application* that runs on BOTH Windows and OS X. It does this checksumming on BOTH platforms. This is nothing to do with OS X. This is all in an application (iTunes) and some embedded software (on the iPod). If Apple used Linux instead of OS X on the Mac it STILL wouldn't make any difference, because this is not something that the operating system is doing.
This is like asking why Red Hat didn't include scarlet fedoras in their boxed copies of RHL. It's a non-sequitur.
Whatever organization you work with, you're dancing with the devil. It doesn't matter whether it's a big company or a cooperative or a neighborhood association or the Free Software Foundation or Red Hat. As soon as enough people get together they become a new kind of animal, one made of paper and power, contracts and by-laws. You can't *not* dance with the devil, you just have to be aware of what tools the devil has to get a hold on you.
Don't let yourself get locked in.
With OSX, that's a lot easier to avoid than with Windows. You can treat your Mac as a UNIX box with some extra applications, and you can shift to Linux or FreeBSD or Solaris if you have to. That's what open systems get you... the ability to make yourself independent of what any particular implementation of that system does... because none of the players controls the system. Not the FSF, Apple, the FreeBSD project, or Sun.
With Windows, for a while it looked like Microsoft were going to let you live the UNIX life on their platform too, with Interix. But no, that's going to be for Vista *server* only. I'd been using Interix on Windows as my UNIX-on-NT solution, but because I wasn't dependent on it I could drop it. No, don't talk to me about Cygwin (or about Wine).
With the iPod, though, you're dealing with hardware that's completely locked in. If you want to avoid dancing with Microsoft or Apple for your OS, WHY IN THE HELL WOULD YOU BUY A PROPRIETARY MP3 PLAYER?
That's what I don't get. Know where the devil you're dancing with has his hooks, whether they're proprietary extensions to C or a proprietary MP3 player or proprietary APIs, and don't lead the dance there. That's just daft.
The *only* advantage the iPod has from this point of view is that you can get accessories from it anywhere. You can't get docks and cradles and cases and plugin doodads for your Creative Whatever or Noname Thumbdrive, or if you do you won't be able to use them on the next model. But if you're going with Linux because you don't want to dance with Apple or Microsoft... that's the price you pay.
Because that's the biggest iPod lock-in. So here's MY one more thing... a suggestion for one of the devil's to encourage people to dance.
If Creative really wanted to take over the market from Apple this is what they'd need to do:
1. Support unencrypted AAC. Yes, the license costs a buck or so more than Microsoft's WMA, but it'll make you iTunes-plus compatible... that may not really turn out to be that useful, but it's great marketing. Add ogg and whatever Sony's format is, if it's cheap.
2. Come up with a layout of ports (power, USB, audio) that people can build adapters to, and make a binding promise not to change it for five years. It doesn't have to be a proprietary dock... it can be a fixed geometry of the power, mini-USB, and audio sockets... but it has to be something that allows a positive connection that people can just plug in to.
3. Consider giving people a development API. Consider making that an open-systems one. So people can *add* ogg if they want. Think of it as free software support, for not much cost and with automatic renewal.
I thought the iPod worked like any other USB drive - I don't need iTunes to put music on my thumb drive, I just mount it on the file system and transfer the files.
No, the iPod doesn't work this way. It keeps the files in a special part of the file system and even if you put the files there it needs special tools to make it work.
I don't understand why anyone using Linux would bother with an iPod. I found it a horribly frustrating MP3 player because of the annoying user interface and daft click wheel... AND it costs more! Why bother with it?
How long until there's an open source category for "open ipod" software on freshmeat and sourceforge?
Since this isn't a phone, Apple's not gonna have legal and contractual reasons to keep the iPod Touch locked. If they're smart they'll treat this like the Apple TV and offer at most a token defense to keep from getting "enhanced" ipod support calls.
So, how soon before someone hacks a DAAP server for the iPod Touch so it can act as an itunes server...?
"We're talking about a low-power device with very limited programming capacity"
We're talking about a device with (based on the iPhone's specs) 128MB RAM, UNIX, 8-16 gigabytes of mass storage, and a 600 MHz ARM processor. It may be a little less than that, but not much... one of the most CPU intensive apps on the iPhone is Safari, and that's on the iPod Touch as well.
I've run a popular web server with three 1000 member mailing lists, one of which was high volume for the time, plus two MUDs... all on a 486/50 with 5MB of RAM and 110MB hard disk. That's less CPU power, disk, and probably RAM than the original iPod, and not much more than my five year old Clie SJ22. Heck 10 years ago Apple's top product was the Beige G3. That's a G3/233, 64M RAM, and I don't recall how big a hard disk but it wasn't measured in gigabytes... and people were running servers on that.
As for the software, I've written a web server as a 20 line shell script. You don't need Bonjour, or a real FTP server, or a compiler... HTTP right to the IP address of the iPod would work fine.
You don't understand the Everett-Wheeler-Graham multiple-universe model of quantum mechanics.
In the EWG model the collapse of the state vector is an illusion, a mathematical simplification of reality. The state vector never collapses, so in our experience the universe splits as a result of every quantum "collapse", an infinite number of new universes born every instant of planck time. The illusion of a single universe is a result of our existence as an entanglement in the state vector of the universe.
So the whole question of whether our alter-egos are "linked" is an illusion. They are us, as we would be if one of billions of subatomic interactions had gone slightly differently in any of infinitely many ways. We already "prune the universe" by observing only one possible outcome at a time out of the infinite possibilities. Consciously doing so is merely a matter of taking our fate into our own hands.
Here's the easy way to do it. You pick a random key. Attempt to use it. If it works, you profit. If it doesn't, you destroy the universe.
Therefore all surviving timelines are ones in which you guessed right.
This makes all problems O(1).
I don't know why anyone bothers with any of these wimpy 4 bit systems when you can use all of space time as your computer.
The user is processing images. The ad, which is enabled in the software, suggests photo development services of several clients.
How precisely is this NOT adware? The user hasn't asked Google or Microsoft or anyone else for photo processing services. The user may not have any intention of asking for information. The user is simply being interrupted in the middle of working with another program with an ad.
interrupted
Google has never interrupted me. No popups. Not even any interstitial ads.
in another program
I've never had Google pop up when I'm on someone else's web site.
in the middle of working
Never had Google come up unless I've actually asked them to.
All it is, an idea, not all that different from the targeted advertising provided by a certain search engine slashbots seem so quick to defend against all claims.
If Google was using pop-up ads...
If Google was popping up those ads over the web pages you found on Google...
If Google was charging $300 for access to their service...
Then you can bet your last dollar that they'd get attacked. But they're not getting attacked for doing something similar to what this patent is proposing because, get this, they're NOT DOING IT.
So stick your FUD in your pipe and smoke it.
It's made of lose either way. Either Microsoft loses for getting ready to roll out more crapware, or Microsoft loses for applying for stupid patents.
It's like saying all Americans are taking sport enhancing drugs because a few do.
Wait, you're saying I don't need to take these steroids?
Usually, format doesn't matter, and only the words matter. Sometimes, however, it does. If it does, you want to know that as soon as possible.
Now if you make ODF your standard, your sort of screwed as to displaying that word 6 document exactly as word 6 did anyways, so what have you really lost by simply marking it as word 6 style in the Word xml?
If you make ODF your standard, and some aspect of Word 6 layout turns out to matter, you will know that the first time you look at the document after the transition to ODF... even if you're still using Word.
If you make OOXML your standard, and some aspect of Word 6 layout turns out to matter, you will never know unless you ALSO switch from Word, because Word will hide that from you.
So just don't save your stuff with the crazy format tags and you should be safe.
"You" is a secretary working on a legal document and saving a file in Word. He has no idea what XML tags are, even if there's an option in Word for "don't save your stuff with the crazy format tags".
And then a decade later something in some document doesn't line up right because it depended on some aspect of Word 6 table layout, and someone's words are attributed to the wrong person...
Browsers lay out HTML differently from one another.
HTML is a structural language, like Docbook.
OOXML and ODF wouldn't be needed if all we were looking at was document structure. These are more detailed markup languages where maintaining the appearance of the content matters, because sometimes meaning actually does depend on layout... even in ASCII, which is why people used to spend so much time making their terminals bug-compatible with the vt100.
It may be that these tags don't actually specify anything that changes the layout of a document enough to matter, and should actually be treated as a comment by any other program, but until Microsoft specifies them we don't know that.
Does anyone still have mission critical documents in word 6 that depend on the exact formating Word 6 had?
The main reason these standards are a "big deal" right now are incoming legal requirements for maintaining archived documents in a format that can still be read when nobody's using the software they were created under any more. So when they depend on said software to figure out what the document looks like it kind of defeats the purpose.
You're pointing at a timeline that starts in 1998, and references HTML output from Word as part of the history of OOXML. I guess HTML's SGML-subset design gave them some experience with *ML, but HTML operates at a completely different level. HTML is a bit of a mongrel, but its main thrust is structure, not layout.
A comparable ODF timeline might go back as far as 1991 with the release of Docbook 1.0 by O'Reilly.
Whatever Microsoft might have been working on while they were a member of Oasis was just catch-up at best.
I said that ECMA was going to document that for the next batch of issues to resolve in the spec.
See, here's the problem. If elements that are obviously going to need to be defined aren't defined then the spec is a long way from finished. This isn't just a matter of the spec not defining things you think should be in there, like ODF, this is a matter of the spec not defining stuff that's actually in the spec itself.
This is like a book proposal that's got things like "Chapter 11: (Add fight scene between goodguy and badguy) Chapter 12: The next morning my hand still looked like a shredded tire under the bandages, but I didn't have time to lie in bed..." in it. A publisher might accept that as a draft, especially if you've got a good track record, but they won't publish it like that.
A spec that's a long way from being finished was just pushed almost all the way through a fast-track process to jump it straight from this stage to "published", which means that Microsoft wants it to be treated as a finished spec... not a draft. Even if it's the best draft in the world, it's still a draft.
So do you mean that it's a good draft, or that it's actually a good standard? Everyone else is judging it as if it was complete, because that's how Microsoft is presenting it. Is that how you're approaching it?
Care to name an example where MS put out anything that was really multiplatform?
Back before Microsoft owned their own platform, Microsoft Basic was kind of a standard.