Doctor: I will defeat you Credulons! Credulon leader: I have faith we will prevail! Doctor: (smugly) Meet my secret weapon - the Professor. Dawkins: Hello. Credulons: No! The skepticism! I'm melting! Dawkins: That was simple. Now, how does this TARDIS thing work, exactly? Doctor: No! The skepticism! I'm melting! Dawkins: Oops. Time for a new title.
Close Credits, including "Next Week on Professor Who..."
Standards are difficult but rewarding...
on
ISO Approves OOXML
·
· Score: 1
I've been involved in development and implementation of international standards (admittedly not ISO, but JCP), and I agree wholeheartedly with the parent poster.
We had a medium-sized Java package to standardize, a couple of dozen classes. It took us about a year and a half, plus an extra half a year of maintenance, clarifying those annoying little omissions that people only spot after release. The total size of the spec was under 200 pages. It was all achieved by consensus, with several competing vendors represented, and other interested parties - including users, developers, and companies who would deploy the implementations.
Although there were some lively debates, at no point was the mood partisan, there were no "irregularities" and since consensus was the norm, issues very rarely came to even an informal vote. According to the rules, the spec could not be ratified until both a test suite and a reference implementation were delivered. In our case, these were developed completely separately from one another.
It was all actually a very pleasant and rewarding process.
If we had seen anything like the irregular behavior we have seen on the OOXML NBs, we would have voted off the spec lead quicker than a very quick thing. (Nothing like this happened, of course - our spec lead was and is well respected.)
I wanted to post this to point out that spec development can be, and usually is, open, honest and friendly, even between competitors.
I also wanted to point out that the technical side takes a long time. It took us two years to get to the point where we were completely happy with our little 200-page spec.
I fail to see how the 6000+ pages of DIS 29500 can even be reviewed properly, let alone be swept through a standardization process this fast. Even a good spec has holes that must be patched up, and are only spotted when someone looks at them from a different angle than you. Rob Weir's excellent blog entry on defect sampling his has been linked elsewhere, and suggests that, statistically, only a small minority of the defects in the spec have actually been found.
This, coupled with a lack of test suite or reference implementation, make it difficult for me to see how the ISO national bodies could possibly have voted anything other than "no", on purely technical grounds.
Unfortunately, this is the argument from personal incredulity, and clearly false, because they did. I would therefore like to know why.
(I have been unable to find a direct email link for this enquiry.)
I would like to know what, if any, action ISO is taking in respect of the widely reported process irregularities in JTC 1/SC 34, specifically with regards to DIS 29500 (Office Open XML).
I would also like to know whether, in the face of any such investigation, the certification of DIS 29500 would be suspended for the duration of the investigation.
In recent days, it has been reported that the BSI is likely to approve DIS29500 (OOXML) for ISO fast-track approval.
As a previous and current participant in international standards processes, I would like to convey my disappointment in this decision. The BSI's raison-d'etre is to promote and protect British interests in the software world, and I fail to see how adopting OOXML does this.
OOXML is simply too large, and has too many technical difficulties, for it to be fast-tracked like this. I could go on about specifics, but I am sure that, by now, you are all too familiar with the chorus of dissatisfaction. I note particularly that even the originator of the specification has not demonstrated a working implementation.
Apart from the technical problems, there are a whole slew of political and legal problems associated with it, none of which work in favour of the British IT industry. The Open Specification Promise which applies to OOXML seems to be very narrowly worded, and effectively excludes whole classes of implementation (ones with extensions, for example). The "ownership" of the specification by ECMA, rather than ISO, can also be used to ensure that the only up-to-date working implementation is one from Microsoft.
Add to this the egregious abuse of the standards process in Finland, Germany, Croatia, Norway and Poland, and I would say that a "no" vote is Britain's only defensible position.
Of course, if the leaks and speculation are false, and the BSI intends to vote "no", then please accept my apologies for wasting your time. However, I feel that this is important enough that not to act would be a failure on my part.
MS limits their liability for using their software to $5, as specified in the EULA. However, this would open them up to severe liability concerns. Unless you explicitly opt in, or it's mentioned in the EULA already, you'd have a hard job deflecting liability when a botched patch nukes half the NHS.
With most viruses, you haven't a clue where they come from, so you can't sue. This one will likely be cryptographically signed.
You missed out a word. "Statistically", it's the same everywhere. If it was literally the same everywhere, the whole universe would be a thin gas with a uniform density of a few particles per cubic meter. There are going to be deviations from the norm, and this galaxy is one of them. It may have experienced some unusual event that stripped it of its dark matter after formation.
What I find odd is that the galaxy is otherwise so unremarkable. If it is rotating differently from other galaxies, wouldn't we see a tighter or looser pattern of spiral arms than usual?
To be fair, though, the measurements themselves are still under scrutiny, so this may all be a storm in a galaxy-sized teacup.
I am surprised by the strong reaction to this minor contamination, when many people are quite happy to ingest unregulated herbal "supplements" with no oversight over purity or dosage. In fact, I predict that the strongest reaction to this will be from precisely those people.
Supplementary, Complementary and Alternative Medicine should be subject to the same controls (including proofs of efficacy) as actual medicine.
Pain I'm skeptical of too, and I bow to your superior experience in this regard, but even temporarily dazzling a pilot over a city is serious.
Having had afterimages for several minutes after being exposed to a specular reflection of a 5mW green laser in office-lit conditions (reflected from a whiteboard), I can sympathize.
It doesn't say the power of the beam - you can quite easily pick up green laser pointers on eBay that are advertised as 100mW or more (here in the UK, at least). Also, the copter is likely to be quite low, and you'd only want to try this at night so you could see the dot. The pilot's pupils would be dilated due to the darkness, so I can imagine quite a severe dazzling effect.
While I agree that, absolutely, you can never completely prove or disprove anything, there comes a point where the alternative hypothesis requires such a huge amount of special pleading that it becomes untenable.
Your example is an intelligent designer that happens to make the entirety of evidence look consistent with evolution by natural selection. At this point, it is impossible to distinguish between the two models. The predictions made without the designer match the predictions made with it, so you might as well ignore the role of the designer completely. This is also, emphatically, NOT what ID proponents believe, or want taught.
There is plenty of potential evidence that would cause extreme problems for evolutionary theory, but which would be perfectly consistent with an intelligent designer. "Fossil rabbits in the Precambrian" is the famous example. Copyright messages, in English, encoded in the human genome. We see none of these.
Gravity is consistent with an intelligent pusher, who only just happens to push hard enough to mimic an inverse square law. Do you seriously suggest that we teach "intelligent falling" alongside gravity? At some point, the lack of evidence for a contrived position leads to the practical abandonment of that position, even if, logically, it *could*, just possibly, be true.
Rest assured that the position would be resurrected if credible evidence for it were discovered.
The trouble is that ID is not a "perfectly good theory" - at best it is a hypothesis. A theory is a hypothesis with evidence to back it up. Michael Behe's debunked arguments notwithstanding, all the evidence from the history of life is consistent with an evolutionary past, and ID makes no predictions at all. Whatever evidence you find is consistent with the "Well, God wanted to do it that way, and you can't say otherwise" principle.
The point it, however, that when some John Q Student gets caught doing this, this will be taken as the college not taking "adequate measures" to stop him. If the lawmakers are dumb enough to pass this legislation, what chance do you think a jury will have understanding the minutiae of encrypted vs. unencrypted vs. copyrighted-but-legal content? "He's trying to hide it? Must be illegal!" Then, bang, the college loses a major part of its funding, ups its fees for all, and the neediest kids get it in the shorts (as usual).
Back when New Scientist magazine had the wacky inventions of Daedalus as its back page light-hearted material, textured musical roads was one of these inventions. It is nice to see that someone has actually implemented the invention at last.
(These columns were collected into a book, "The Inventions of Daedalus", by David E H Jones (1982). They are, with very few exceptions, both bizarre and brilliant in equal measure.)
CLDC 1.1 does indeed support float (but not double), and the spec was ratified in March 2003.
MIDP is actually quite a capable graphical platform these days, especially with APIs like the Mobile 3D Graphics (JSR184) and Scalable Vector Graphics (JSR226). Many high-end phones already have ARM11-class CPUs with floating point, and the new Cortex A8 and A9 also have FPUs on board. Dedicated GPUs are starting to penetrate into the top end of the mobile space. (Here's a recent link: http://news.zdnet.co.uk/hardware/0,1000000091,39290158,00.htm)
It will be interesting to see the hardware specs of the new phone and extrapolate what it's going to be able to do.
Taht's grate. If i dont hav to spel check my code, their is no way taht my colleags can get up-tite when they hav dificulty undertsanding what itis Im trying to convay. This willl save me litrally minute severy day. And the peopel who are useing my interface com.MyCmopany.MyProdduct.CheckAcount will have no trouble neither.
Unless the thing that they are changing to solves a real problem for them, then they will not change. And having transparent title bars on windows is not a real problem for most people. No amount of begging will convince people that they have a problem when they don't.
Once again, Microsoft proves that its previous versions are its biggest competitor.
Slashdot Mirror
Just to see how many come dressed as the Doppler Effect.
Mmmmeeeoooowwwwwww...
Doctor: I will defeat you Credulons!
Credulon leader: I have faith we will prevail!
Doctor: (smugly) Meet my secret weapon - the Professor.
Dawkins: Hello.
Credulons: No! The skepticism! I'm melting!
Dawkins: That was simple. Now, how does this TARDIS thing work, exactly?
Doctor: No! The skepticism! I'm melting!
Dawkins: Oops. Time for a new title.
Close Credits, including "Next Week on Professor Who..."
I've been involved in development and implementation of international standards (admittedly not ISO, but JCP), and I agree wholeheartedly with the parent poster.
We had a medium-sized Java package to standardize, a couple of dozen classes. It took us about a year and a half, plus an extra half a year of maintenance, clarifying those annoying little omissions that people only spot after release. The total size of the spec was under 200 pages. It was all achieved by consensus, with several competing vendors represented, and other interested parties - including users, developers, and companies who would deploy the implementations.
Although there were some lively debates, at no point was the mood partisan, there were no "irregularities" and since consensus was the norm, issues very rarely came to even an informal vote. According to the rules, the spec could not be ratified until both a test suite and a reference implementation were delivered. In our case, these were developed completely separately from one another.
It was all actually a very pleasant and rewarding process.
If we had seen anything like the irregular behavior we have seen on the OOXML NBs, we would have voted off the spec lead quicker than a very quick thing. (Nothing like this happened, of course - our spec lead was and is well respected.)
I wanted to post this to point out that spec development can be, and usually is, open, honest and friendly, even between competitors.
I also wanted to point out that the technical side takes a long time. It took us two years to get to the point where we were completely happy with our little 200-page spec.
I fail to see how the 6000+ pages of DIS 29500 can even be reviewed properly, let alone be swept through a standardization process this fast. Even a good spec has holes that must be patched up, and are only spotted when someone looks at them from a different angle than you. Rob Weir's excellent blog entry on defect sampling his has been linked elsewhere, and suggests that, statistically, only a small minority of the defects in the spec have actually been found.
This, coupled with a lack of test suite or reference implementation, make it difficult for me to see how the ISO national bodies could possibly have voted anything other than "no", on purely technical grounds.
Unfortunately, this is the argument from personal incredulity, and clearly false, because they did. I would therefore like to know why.
My comment to ISO, as submitted on this web form:
http://www.standardsinfo.net/info/livelink/fetch/2000/148478/6301438/enqsvc_ISO_IEC/enqsvc_ISO/enqsvc_ISO_%20general/enqsvc_ISO_general_contact_form.html
----
Sir/Madam,
(I have been unable to find a direct email link for this enquiry.)
I would like to know what, if any, action ISO is taking in respect of the widely reported process irregularities in JTC 1/SC 34, specifically with regards to DIS 29500 (Office Open XML).
I would also like to know whether, in the face of any such investigation, the certification of DIS 29500 would be suspended for the duration of the investigation.
Wasn't this what actually happened, in Norway at least?
And when we complain to MS that Opposite Day wasn't declared, they will just say that they declared it oppositely by not declaring it.
Here's my shot at fame and stardom...
---
Mr. Low, Mrs. Stride, Mr. Stokes,
In recent days, it has been reported that the BSI is likely to approve DIS29500 (OOXML) for ISO fast-track approval.
As a previous and current participant in international standards processes, I would like to convey my disappointment in this decision. The BSI's raison-d'etre is to promote and protect British interests in the software world, and I fail to see how adopting OOXML does this.
OOXML is simply too large, and has too many technical difficulties, for it to be fast-tracked like this. I could go on about specifics, but I am sure that, by now, you are all too familiar with the chorus of dissatisfaction. I note particularly that even the originator of the specification has not demonstrated a working implementation.
Apart from the technical problems, there are a whole slew of political and legal problems associated with it, none of which work in favour of the British IT industry. The Open Specification Promise which applies to OOXML seems to be very narrowly worded, and effectively excludes whole classes of implementation (ones with extensions, for example). The "ownership" of the specification by ECMA, rather than ISO, can also be used to ensure that the only up-to-date working implementation is one from Microsoft.
Add to this the egregious abuse of the standards process in Finland, Germany, Croatia, Norway and Poland, and I would say that a "no" vote is Britain's only defensible position.
Of course, if the leaks and speculation are false, and the BSI intends to vote "no", then please accept my apologies for wasting your time. However, I feel that this is important enough that not to act would be a failure on my part.
Yours,
Sean Ellis
[address elided]
MS limits their liability for using their software to $5, as specified in the EULA. However, this would open them up to severe liability concerns. Unless you explicitly opt in, or it's mentioned in the EULA already, you'd have a hard job deflecting liability when a botched patch nukes half the NHS.
With most viruses, you haven't a clue where they come from, so you can't sue. This one will likely be cryptographically signed.
You missed out a word. "Statistically", it's the same everywhere. If it was literally the same everywhere, the whole universe would be a thin gas with a uniform density of a few particles per cubic meter. There are going to be deviations from the norm, and this galaxy is one of them. It may have experienced some unusual event that stripped it of its dark matter after formation.
What I find odd is that the galaxy is otherwise so unremarkable. If it is rotating differently from other galaxies, wouldn't we see a tighter or looser pattern of spiral arms than usual?
To be fair, though, the measurements themselves are still under scrutiny, so this may all be a storm in a galaxy-sized teacup.
I am surprised by the strong reaction to this minor contamination, when many people are quite happy to ingest unregulated herbal "supplements" with no oversight over purity or dosage. In fact, I predict that the strongest reaction to this will be from precisely those people.
Supplementary, Complementary and Alternative Medicine should be subject to the same controls (including proofs of efficacy) as actual medicine.
Pain I'm skeptical of too, and I bow to your superior experience in this regard, but even temporarily dazzling a pilot over a city is serious.
Having had afterimages for several minutes after being exposed to a specular reflection of a 5mW green laser in office-lit conditions (reflected from a whiteboard), I can sympathize.
It doesn't say the power of the beam - you can quite easily pick up green laser pointers on eBay that are advertised as 100mW or more (here in the UK, at least). Also, the copter is likely to be quite low, and you'd only want to try this at night so you could see the dot. The pilot's pupils would be dilated due to the darkness, so I can imagine quite a severe dazzling effect.
While I agree that, absolutely, you can never completely prove or disprove anything, there comes a point where the alternative hypothesis requires such a huge amount of special pleading that it becomes untenable.
Your example is an intelligent designer that happens to make the entirety of evidence look consistent with evolution by natural selection. At this point, it is impossible to distinguish between the two models. The predictions made without the designer match the predictions made with it, so you might as well ignore the role of the designer completely. This is also, emphatically, NOT what ID proponents believe, or want taught.
There is plenty of potential evidence that would cause extreme problems for evolutionary theory, but which would be perfectly consistent with an intelligent designer. "Fossil rabbits in the Precambrian" is the famous example. Copyright messages, in English, encoded in the human genome. We see none of these.
Gravity is consistent with an intelligent pusher, who only just happens to push hard enough to mimic an inverse square law. Do you seriously suggest that we teach "intelligent falling" alongside gravity? At some point, the lack of evidence for a contrived position leads to the practical abandonment of that position, even if, logically, it *could*, just possibly, be true.
Rest assured that the position would be resurrected if credible evidence for it were discovered.
The trouble is that ID is not a "perfectly good theory" - at best it is a hypothesis. A theory is a hypothesis with evidence to back it up. Michael Behe's debunked arguments notwithstanding, all the evidence from the history of life is consistent with an evolutionary past, and ID makes no predictions at all. Whatever evidence you find is consistent with the "Well, God wanted to do it that way, and you can't say otherwise" principle.
How can there be 200+ replies and no-one has mentioned "Superiority" by Arthur C Clarke?
The point it, however, that when some John Q Student gets caught doing this, this will be taken as the college not taking "adequate measures" to stop him. If the lawmakers are dumb enough to pass this legislation, what chance do you think a jury will have understanding the minutiae of encrypted vs. unencrypted vs. copyrighted-but-legal content? "He's trying to hide it? Must be illegal!" Then, bang, the college loses a major part of its funding, ups its fees for all, and the neediest kids get it in the shorts (as usual).
Back when New Scientist magazine had the wacky inventions of Daedalus as its back page light-hearted material, textured musical roads was one of these inventions. It is nice to see that someone has actually implemented the invention at last.
(These columns were collected into a book, "The Inventions of Daedalus", by David E H Jones (1982). They are, with very few exceptions, both bizarre and brilliant in equal measure.)
Pamela Jones!
CLDC 1.1 does indeed support float (but not double), and the spec was ratified in March 2003.
MIDP is actually quite a capable graphical platform these days, especially with APIs like the Mobile 3D Graphics (JSR184) and Scalable Vector Graphics (JSR226). Many high-end phones already have ARM11-class CPUs with floating point, and the new Cortex A8 and A9 also have FPUs on board. Dedicated GPUs are starting to penetrate into the top end of the mobile space. (Here's a recent link: http://news.zdnet.co.uk/hardware/0,1000000091,39290158,00.htm)
It will be interesting to see the hardware specs of the new phone and extrapolate what it's going to be able to do.
Oh, mod parent up, up, up, please.
Is this just a big storm over nothing?
I save ODF locally, PDF if someone else needs to print it, RTF if I need to send it to someone to edit, DOC if I need hell to freeze over.
(OT: Has everyone seen the new Open Rights Group T-shirts?)
I think you mean GNU/FreeISO.
Taht's grate. If i dont hav to spel check my code, their is no way taht my colleags can get up-tite when they hav dificulty undertsanding what itis Im trying to convay. This willl save me litrally minute severy day. And the peopel who are useing my interface com.MyCmopany.MyProdduct.CheckAcount will have no trouble neither.
Here's the official list from ISO:
t /TechnicalCommitteeParticipationListPage.Technical CommitteeParticipationList?COMMID=4767
http://www.iso.ch/iso/en/stdsdevelopment/tc/tclis
There are 32 participating countries, and 15 observers, a little short of the 123 claimed.
Spoiled? No way. I used it to enrol them in the Ayn Rand school for tots.
(And yes, the parent's parent post was a joke too.)
I, on the other hand, upgraded to Kubuntu 7.04 and I have $300 to spend on my kids.
(Smug mode off.)
Unless the thing that they are changing to solves a real problem for them, then they will not change. And having transparent title bars on windows is not a real problem for most people. No amount of begging will convince people that they have a problem when they don't.
Once again, Microsoft proves that its previous versions are its biggest competitor.