Incorrect. All that was required for the update he wanted to do was either (1) updating two locations, not one, or (2) removing one file. The spreadsheet file stores some extra information as an optimization to allow spreadsheets to load faster (basically, information on calculation chains).
Here is a thorough point-by-point response to Rodriguez's points.
BTW, it is interesting to note the comments when this was discussed on Brian Jones' blog. Note that he lets Rodriguez comment, even though Rodriguez is strongly against OOXML to the point of rudeness.
That's a pattern I see on most of the pro-OOXML blogs I've read. They let the other side come in and have its say. They often even link to the opposing arguments. You go read Jones' blog, for example, and the impression you get is that it is written by someone strongly in favor of OOXML, but who wants to be fair and make sure all arguments are heard. He's confident that if that happens, he'll win.
That's sure not how the anti-OOXML blogs operate. They often turn off comments completely. They don't link to opposing blogs. They mostly just attack, and the attacks are often factually incorrect. And when they do actually post a valid technical flaw in OOXML, half the time ODF has the same technical flaw, but that never seems to bother them. I sure don't get the impression from, say, Weir's blog or Sutor's blog that I'm reading the writings of someone who wants to be fair and present a winning technical argument.
Stéphane Rodriguez lost all credibility when he edited an OOXML spreadsheet file by hand, changed the XML so that it was no longer legal according to the schema, and then proclaimed that it was a flaw in OOXML that Excel found an error in the document.
Actually, the main problem with ebooks now that paper-like displays are seeing some progress is the cost. $400 for a Kindle is just nasty
Actually, $400 is basically free, if you are a heavy reader. Kindle books seem to be uniformly, and significantly, cheaper than the non-Kindle editions. A heavy reader will make up that $400 in a year or so, and then start pulling ahead.
From the specs, and the reviews I've read, it was a lousy computer. When you put Linux on a lousy computer, you have...a lousy computer with Linux it.
I don't understand why so many in the Linux community were pleased by this. Having Linux associated with low-end machines that people buy because they can't afford what the really want does not help Linux. We shouldn't be promoting Linux as the OS for those who have to settle for less.
But I still fail to see how Web 2.0 will make an operating system irrelevant. The browser has to run on something
The idea is that if the browser is all you are interacting with, the OS is just some invisible thing to you. It's irrelevant in the same sense that, say, the particular brand of CPU you are using is irrelevant--you still need a CPU, but since you are dealing with it at a higher level, you don't really care which it is.
This isn't the first time someone has said the browser will make the OS irrelevant. There was Marc Andreessen's famous quote that Netscape was going to reduce the importance of Windows to just a poorly debugged set of device drivers for the browser. That statement, and other trash talk against MS by Netscape executives, before MS was into the internet, actually got Microsoft to recognize the threat and move to supporting the internet. In retrospect, I suspect that the Netscape guys wish they had just kept quite!
You flaws 1 and 2 also applied to the ODF ISO standardization. If ODF had been subjected to the same level of scrutiny as OOXML, it would also have hit your flaws 4 and 5 and 6. But it wasn't subject to such scrutiny. so sailed through, with the knowledge that it could be finished in later revisions. (And it needed fixing...see the massive changes in ODF 1.2 and compare to 1.0).
So why is Microsoft being required to operate under different rules? People seem to want theirs to be flawless before allowing it to be an ISO standard--a requirement no one else has been subject to.
Additions to the standard made once Sun was no longer a participant would not be covered. That seems completely reasonable considering anyone could add anything in Sun's absence
Would that be a bad thing?
Letting people add anything they want works fine with software. Why not for specifications, too?
You are right that Sun's doesn't have "necessary claims" language, and I'll agree that it is better for these kind of documents to not have such language. Sun's is better in that regard. Sun's is also better in that it covers future versions of ODF, unlike the MS and IBM covenants. However, they don't get full credit for the later, cause it does have that restriction to future versions that Sun participates in. Note that this means that if Sun pulls out of OASIS, future OASIS development of ODF is under a patent cloud. I think that to be really called open, whether it be software or a specification, development has to be allowed to continue even if the founders pull out, and the community should even be allowed to fork in directions that the founders might not like. But neither IBM now MS allow that either, so perhaps none of these specs should really be called open?
As far as the necessary claims language goes, note that IBM has it, too. Open source developers seem to have managed to deal with it there, so I expect they'll manage to deal with it in Microsoft's covenant, too.
The non-assert agreement only guaranteed that the patents would not be asserted against fully conforming implementations. That requirement is also in IBM's non-assertion covenant. (Well, IBM says fully compliant, not conforming, and Microsoft doesn't have the word fully in theirs).
But the specifications of the standard (at that time) were such that nobody, including MS, could actually build a fully conforming implementation. (Including such wonderful statements as "split the text layout in the same way that Word 95 did."(paraphrase. I'm *NOT* going to read that mess of garbage again!)
It never had anything like that in it. What it had was basically a set of flags, that someone importing documents from old versions of Word and WordPerfect (and a few others) could use to record the fact that those documents had formatting settings that OOXML does not handle, so that, for example, if you wanted to convert back to the old document format, you could preserve that. The spec also said that programs producing new documents should not use these flags.
Note that you can do the same thing in ODF, by using application-specific tags or metadata. The only real difference is that the OOXML spec reserves from names for this, so that if two different implementations want to write WordPerfect 5 importers/exporters, say, they have a chance of interoperating. In ODF, unless they talked to each other first, they would probably pick different application-specific tags or metadata, and so not interoperate.
Also the non-assert agreement named a particular version of the specifications to which it applied. Which didn't imply in any way that if some security fix was mandatory, that it would be legal to apply that fix. Just like Sun's non-assertion covenant. You should actually look at the MS, Sun, and IBM licenses. I made it ridiculously easy to do so, and they are all quite reasonably sized, and the language isn't bad for legal documents.
That patent non-assert covenant is almost identical (and the differences are in the parts that aren't important) to Microsoft's patent no- assert covenant for its XML formats. Many have said that the latter is unacceptable for use with free software. It's also interesting to compare those two non-assert covenants against the one IBM provides for their patents that cover OpenOffice, and for Microsoft's OSP. I've made a little page that lists all four of these non-assertion covenants, side-by-side, with corresponding sections highlighted in matching colors.
I've occasionally considered doing a similar thing for media under DRM, such as audiobooks from Audible. My friends and I all have old iPods, and even new iPod shuffles are cheap. So, why not just have a floating pool of old iPods or Shuffles that we pass among us, whenever we want to loan a friend an audiobook?
Note that you only need to distribute through Apple if you want to run on iPhone and iPod Touch. You can run on the iPhone Simulator, which is part of the free SDK, without having to go through the store.
I saw emails from Bill Gates raising similar concerns over Windows 98 performance, in quite heated tones, so I'm not sure that this sort of thing isn't just normal for Microsoft culture.
(A company I'm associated with sued Microsoft, and I attended the trial. Plaintiff entered into evidence all kinds of interesting internal Microsoft emails they obtained through the discovery process. Anyone interested in seeing interesting internal Microsoft tidbits should consider finding one of the many court cases they are involved in, and attending)
If by Airport you mean a wireless connection to an actual Apple Airport router, then skip this reply. If, on the other hand, the wireless router is not an Apple router, check to see if the router has some kind of "turbo" or "speed boost" or similar mode. Those modes do some things that are outside the standard but often work (especially with wireless cards made by the same manufacturer as the router!), but sometimes don't. If the router has such modes, try turning them off.
Always take what any expert witness says with a large grain of salt. They are paid handsomely for their opinions. In a trial I witnessed, for example, in a controversy involving RAM cache, an expert testified that storing data on a hard disk would count as storing it in RAM cache, because hard disks are random access devices. You'd have a hard time finding anyone in the industry who ever used RAM cache to include caches on hard disk before that expert wrote his report, but having that opinion was better for the party that hired him, so he dug deep, and found a way to say that with a straight face.
This guy was definitely an expert. He had a string of well respected papers a mile long. He was an IEEE Fellow. He'd been, I believe, the head of the EE department at one or major engineering colleges. Advisor to numerous top companies, and member of numerous standardization committees. But he's retired now, and was probably getting $50-$100k (plus expenses) to find a way to say that disk cache was RAM cache, so even if that is a ridiculous position to take, it's not going to harm his career, or even his reputation. Even if his ridiculous position at this one trial came to the attention of people currently active in his field, they all know about the expert witness game, and will dismiss this. As long as you don't outright perjure yourself, taking a ridiculous position for money in court won't hurt you.
No Moonlight is planned for Linux, which will always be behind Silverlight, because Microsoft won't give them the specs until AFTER each new release of Silverlight, which could mean months of cathup after every Silverlight release
So? You think sites using Silverlight are going to immediately upgrade to each new version, and immediately start using features that aren't in the old version? That's not what happens in the real world. What happens is that it takes sites months or years to decide to go to the new version of this kind of thing, which will give Moonlight plenty of time to catch up.
Here's a great little power brick for travel. It is slim and light (59g according to my kitchen scale) to fit easily in your bag, and the prongs fold in for storage. It can charge two USB devices at once, and it is cheap. I had a long Amtrak trip last year (Seattle to Texas, and back a month later), and two of those kept my phone, two iPods, Nintendo DS, and Tomtom all happy on the train, and in my hotel room. And actually, one would have been enough.
For the curious, the reason I had two iPods is that I took my 40 gigabyte hard-drive based model that has my complete library, for listening to music in the hotel, and my Nano for listening to audio books and podcasts in the hotel and on the go. The flash-based players are better for the latter, because they are more response if you miss something and need to skip back a few seconds.
If I were doing that trip again now, I'd probably buy a Kindle. That would be perfect for a train trip.
Amazon claims 2 million songs on their website. iTunes has a lot more than that.
Songs don't just magically appear when an online store signs a deal with a label. Rather, they start making songs available over time, and it can take quite a while before they have them all. The last album I bought was the second album from the Urban Verbs (a sadly overlooked band--their lead singer was the brother of the Talking Heads drummer, and so people just pigeonholed them as a Heads wannabe). When I bought it a few weeks ago, it was on iTunes and not Amazon. Now it is on Amazon. At the same time, I bought a Julia Ecklar album that was on both, so I bought it from Amazon.
You aren't understanding the market for the Air. It's a niche laptop. A perfect example was given on a recent episode of TWiT, where most of the panelists were not impressed with the Air, but one of them loved it. He is a prolific writer, and when he wants to spend an hour sitting on the couch, say, watching TV, he's still writing. With a regular laptop, that is awkward, as they are hot and heavy when actually used as a laptop computer.
And with a small machine, like his Vaio (or with a machine like an Asus EEE), you've got a small screen and a small keyboard. He can go for 10 minutes or so, but it is just too painful to write for hours on those. Those also have horrible battery life.
The Air, he said, is perfect here. It is light enough and cool enough that he can use it on his lap on the couch for as long as he wants, but he has a decent sizes, beautiful screen, and a good keyboard, and good battery life.
For the niche market of people who write incessantly and don't want to deal with a tiny pain-inducing keyboard and small screen in order to write everywhere they go, it is a winner. And there will be other niche markets like that, where everything comes together with it and it is a 5 star laptop for those people. For people who don't fit into one of those niches, it won't be a good choice.
Slashdot Mirror
Incorrect. All that was required for the update he wanted to do was either (1) updating two locations, not one, or (2) removing one file. The spreadsheet file stores some extra information as an optimization to allow spreadsheets to load faster (basically, information on calculation chains).
Here is a thorough point-by-point response to Rodriguez's points.
BTW, it is interesting to note the comments when this was discussed on Brian Jones' blog. Note that he lets Rodriguez comment, even though Rodriguez is strongly against OOXML to the point of rudeness.
That's a pattern I see on most of the pro-OOXML blogs I've read. They let the other side come in and have its say. They often even link to the opposing arguments. You go read Jones' blog, for example, and the impression you get is that it is written by someone strongly in favor of OOXML, but who wants to be fair and make sure all arguments are heard. He's confident that if that happens, he'll win.
That's sure not how the anti-OOXML blogs operate. They often turn off comments completely. They don't link to opposing blogs. They mostly just attack, and the attacks are often factually incorrect. And when they do actually post a valid technical flaw in OOXML, half the time ODF has the same technical flaw, but that never seems to bother them. I sure don't get the impression from, say, Weir's blog or Sutor's blog that I'm reading the writings of someone who wants to be fair and present a winning technical argument.
They've explicitly committed to supporting it in whatever form it is in if/when ISO approves it.
Stéphane Rodriguez lost all credibility when he edited an OOXML spreadsheet file by hand, changed the XML so that it was no longer legal according to the schema, and then proclaimed that it was a flaw in OOXML that Excel found an error in the document.
Patent trivia: patents in the US are always granted on Tuesday
No, I have no idea why that is, but check the dates. They are always Tuesday.
Actually, $400 is basically free, if you are a heavy reader. Kindle books seem to be uniformly, and significantly, cheaper than the non-Kindle editions. A heavy reader will make up that $400 in a year or so, and then start pulling ahead.
The article is about audio books, not ebooks.
From the specs, and the reviews I've read, it was a lousy computer. When you put Linux on a lousy computer, you have...a lousy computer with Linux it.
I don't understand why so many in the Linux community were pleased by this. Having Linux associated with low-end machines that people buy because they can't afford what the really want does not help Linux. We shouldn't be promoting Linux as the OS for those who have to settle for less.
The idea is that if the browser is all you are interacting with, the OS is just some invisible thing to you. It's irrelevant in the same sense that, say, the particular brand of CPU you are using is irrelevant--you still need a CPU, but since you are dealing with it at a higher level, you don't really care which it is.
This isn't the first time someone has said the browser will make the OS irrelevant. There was Marc Andreessen's famous quote that Netscape was going to reduce the importance of Windows to just a poorly debugged set of device drivers for the browser. That statement, and other trash talk against MS by Netscape executives, before MS was into the internet, actually got Microsoft to recognize the threat and move to supporting the internet. In retrospect, I suspect that the Netscape guys wish they had just kept quite!
You flaws 1 and 2 also applied to the ODF ISO standardization. If ODF had been subjected to the same level of scrutiny as OOXML, it would also have hit your flaws 4 and 5 and 6. But it wasn't subject to such scrutiny. so sailed through, with the knowledge that it could be finished in later revisions. (And it needed fixing...see the massive changes in ODF 1.2 and compare to 1.0).
So why is Microsoft being required to operate under different rules? People seem to want theirs to be flawless before allowing it to be an ISO standard--a requirement no one else has been subject to.
Would that be a bad thing?
Letting people add anything they want works fine with software. Why not for specifications, too?
You are right that Sun's doesn't have "necessary claims" language, and I'll agree that it is better for these kind of documents to not have such language. Sun's is better in that regard. Sun's is also better in that it covers future versions of ODF, unlike the MS and IBM covenants. However, they don't get full credit for the later, cause it does have that restriction to future versions that Sun participates in. Note that this means that if Sun pulls out of OASIS, future OASIS development of ODF is under a patent cloud. I think that to be really called open, whether it be software or a specification, development has to be allowed to continue even if the founders pull out, and the community should even be allowed to fork in directions that the founders might not like. But neither IBM now MS allow that either, so perhaps none of these specs should really be called open?
As far as the necessary claims language goes, note that IBM has it, too. Open source developers seem to have managed to deal with it there, so I expect they'll manage to deal with it in Microsoft's covenant, too.
It never had anything like that in it. What it had was basically a set of flags, that someone importing documents from old versions of Word and WordPerfect (and a few others) could use to record the fact that those documents had formatting settings that OOXML does not handle, so that, for example, if you wanted to convert back to the old document format, you could preserve that. The spec also said that programs producing new documents should not use these flags.
Note that you can do the same thing in ODF, by using application-specific tags or metadata. The only real difference is that the OOXML spec reserves from names for this, so that if two different implementations want to write WordPerfect 5 importers/exporters, say, they have a chance of interoperating. In ODF, unless they talked to each other first, they would probably pick different application-specific tags or metadata, and so not interoperate.
Also the non-assert agreement named a particular version of the specifications to which it applied. Which didn't imply in any way that if some security fix was mandatory, that it would be legal to apply that fix. Just like Sun's non-assertion covenant. You should actually look at the MS, Sun, and IBM licenses. I made it ridiculously easy to do so, and they are all quite reasonably sized, and the language isn't bad for legal documents.That patent non-assert covenant is almost identical (and the differences are in the parts that aren't important) to Microsoft's patent no- assert covenant for its XML formats. Many have said that the latter is unacceptable for use with free software. It's also interesting to compare those two non-assert covenants against the one IBM provides for their patents that cover OpenOffice, and for Microsoft's OSP. I've made a little page that lists all four of these non-assertion covenants, side-by-side, with corresponding sections highlighted in matching colors.
I've occasionally considered doing a similar thing for media under DRM, such as audiobooks from Audible. My friends and I all have old iPods, and even new iPod shuffles are cheap. So, why not just have a floating pool of old iPods or Shuffles that we pass among us, whenever we want to loan a friend an audiobook?
Note that you only need to distribute through Apple if you want to run on iPhone and iPod Touch. You can run on the iPhone Simulator, which is part of the free SDK, without having to go through the store.
(A company I'm associated with sued Microsoft, and I attended the trial. Plaintiff entered into evidence all kinds of interesting internal Microsoft emails they obtained through the discovery process. Anyone interested in seeing interesting internal Microsoft tidbits should consider finding one of the many court cases they are involved in, and attending)
If by Airport you mean a wireless connection to an actual Apple Airport router, then skip this reply. If, on the other hand, the wireless router is not an Apple router, check to see if the router has some kind of "turbo" or "speed boost" or similar mode. Those modes do some things that are outside the standard but often work (especially with wireless cards made by the same manufacturer as the router!), but sometimes don't. If the router has such modes, try turning them off.
This guy was definitely an expert. He had a string of well respected papers a mile long. He was an IEEE Fellow. He'd been, I believe, the head of the EE department at one or major engineering colleges. Advisor to numerous top companies, and member of numerous standardization committees. But he's retired now, and was probably getting $50-$100k (plus expenses) to find a way to say that disk cache was RAM cache, so even if that is a ridiculous position to take, it's not going to harm his career, or even his reputation. Even if his ridiculous position at this one trial came to the attention of people currently active in his field, they all know about the expert witness game, and will dismiss this. As long as you don't outright perjure yourself, taking a ridiculous position for money in court won't hurt you.
This sounds like it could be a process that would not smell pleasant, if you are walking around with decaying organic matter on your clothes.
So? You think sites using Silverlight are going to immediately upgrade to each new version, and immediately start using features that aren't in the old version? That's not what happens in the real world. What happens is that it takes sites months or years to decide to go to the new version of this kind of thing, which will give Moonlight plenty of time to catch up.
For the curious, the reason I had two iPods is that I took my 40 gigabyte hard-drive based model that has my complete library, for listening to music in the hotel, and my Nano for listening to audio books and podcasts in the hotel and on the go. The flash-based players are better for the latter, because they are more response if you miss something and need to skip back a few seconds.
If I were doing that trip again now, I'd probably buy a Kindle. That would be perfect for a train trip.
Songs don't just magically appear when an online store signs a deal with a label. Rather, they start making songs available over time, and it can take quite a while before they have them all. The last album I bought was the second album from the Urban Verbs (a sadly overlooked band--their lead singer was the brother of the Talking Heads drummer, and so people just pigeonholed them as a Heads wannabe). When I bought it a few weeks ago, it was on iTunes and not Amazon. Now it is on Amazon. At the same time, I bought a Julia Ecklar album that was on both, so I bought it from Amazon.
Some of us want particular songs. iTMS has many more songs than Amazon at this point.
The ODF TC Editor at OASIS disagrees with you.
And with a small machine, like his Vaio (or with a machine like an Asus EEE), you've got a small screen and a small keyboard. He can go for 10 minutes or so, but it is just too painful to write for hours on those. Those also have horrible battery life.
The Air, he said, is perfect here. It is light enough and cool enough that he can use it on his lap on the couch for as long as he wants, but he has a decent sizes, beautiful screen, and a good keyboard, and good battery life.
For the niche market of people who write incessantly and don't want to deal with a tiny pain-inducing keyboard and small screen in order to write everywhere they go, it is a winner. And there will be other niche markets like that, where everything comes together with it and it is a 5 star laptop for those people. For people who don't fit into one of those niches, it won't be a good choice.