Except, sadly, that they were responsible for asking the Feds to arrest Dmitry Sklyarov. While out of character for the company, the enormity of this is such that I have to think of them as scum for ever after, at least until they actually confess that they were wayyyyy out of line.
This is awful. A huge unwieldy process bound standards organisation is now unable to choose which huge unwieldy standard to give its support to.
"We were really hoping to get the spellcheck dictionary custom term event model moved from 2nd round consultation to initial pre-ratification, but because of byelaw 12 part c regarding type P participants, it looks like we'll be delayed until the *next* meeting of the syntax objects (editing mode) technical steering group. That could be months away." - said Gerhardt Steiner, an XML expert who is a regular voter and sits on several subcommittees.
"Well, I don't understand the fuss" said Sanjeev Raj, "We joined so that our small company that makes custom MSWord plugins for the pharmaceutical industry doesn't go out of business. We paid our membership fee so we could vote on the one technical aspect we care about. I'd love to vote on lots of other things by I've got a company to run."
It's that kind of attitude, says Gerhardt, that's undermining the whole process. "Everyone knows that these kinds of standards organisation are dominated by bearded loons dredged up from the R&D departments of fortune 500 companies, with a few tenured professors thrown in to keep them out of trouble. We work hard to invent hopelessly impractical standards that are overblown, overspecified and impossible to implement unless you are a fortune 500 company with a large R&D department. Even then, you usually have to get a tenured professor to come in and give you some expensive consultancy. It's a fair and equitable system. I think it's a real shame when people try to warp these standards processes for their own selfish financial ends."
Yes, I am a member of one of these committees. No I never vote.
Oh, I don't know, I think it's pretty close to impossible. People are working on the easy stuff - names, addresses, events, locations. I think there will be real progress there, and we'll get some useful software. But semantic modelling is extremely hard, because people do not work on semantics. Most people don't, in their heads, categorise things in ways that a computer make make head nor tail of. Semantic models that work for computers require deep heirarchies, relatively few relation types, and fixed degrees of reflexivity, transitivity and symmetry for those relations. People just don't care about that stuff, and probably never will.
The problems I have with writing semantically aware software are all to do with the fact that people aren't semantically aware. No two people will ever categorise things in the same way. You might, in the case of something dull like addresses or phone numbers manage to get them to compromise for the sake of sharing, but you'll never do it for something more complicated that they really care about.
At best, certain tasks will make use of semantic webs (plural!). Maybe light plastics manufacturing will start to use certain semantic interchange standards. Maybe consumer travel will agree some standards for hotels and flights and trains.
That's a silly analogy and always has been. Here's why:
Imagine you take the whole of human civilization as a day. For the first 23 hours we were naked. Then for the next 50 minutes we wore one or two pieces of crude clothing. In just the last 10 seconds we have acquired scarves, tights running shorts. Therefore in the future we will all have millions of items of clothing, and perhaps wear up to 10,000 of them at once.
The fact that for a certain slice of time, there is an exponential curve, just isn't very interesting.
Take volcanoes. For decades they do nothing. Then over a few days they start to output some heat and smoke, then over a few hours they output vast quantities of heat, and then... then they stop. They don't amazingly continue to output energy at an exponential rate until the planet melts.
So what if a bunch of little bipeds on a planet somewhere spent ages picking fruit, and then suddenly figured out flatscreen TV and breast implants? Whose to say the normal curve isn't like the volcano? Why not predict that 5,000 years from now, we will be picking fruit again, but this time with a few myths about an ancient pre-cursor race, which in 12,000 years time will be verified when man-made fragments are discovered deep within the unusual mineral deposits we now call cities.
Although I'm not hands on now, I originally moved from Linux to Solaris (with some Irix stuff in the middle). I still prefer Solaris for the following reason:
Simple is better.
This single thought is perhaps the biggest lesson I've learned in my whole career, about almost any aspect of computing. Complexity is the enemy.
caveat: by 'Linux' I mean 'The particular distro your company has standardised on' caveat: I'm only concerned with servers. Solaris may be the worst desktop OS in the world FAIK.
1. Less shovelware. Although a base Solaris install is still annoyingly large, it's not nearly as bad as most Linux distros. It infuriates me that operating systems think its useful to install entire database, programming languages, you name its 'just in case you need them'. 2. Better backward compatability. Upgrades to discreet parts of Solaris don't usually require upgrades to other parts of Solaris. This means that you aren't constantly trying to run the latest versions of everything. 3. Better hardware integration. When you are running a lot of servers, it's very useful to have a nice console, so you can talk to the things properly. I think Linux has improved a bit in this area, but I'm not aware that it has an equivalent to the OK prompt, and the various diagnostic tools therein.
Others have talked about various tools and kernel level stuff, but I wanted to make that point that while the Solaris userland might feel archaic to some, to me it feels pleasantly simple - devoid of hidden complexity, obscure features that badly written apps come to rely on, and all the other 'let's have another feature' attitude prevalent in much OS software.
To me, Solaris feels like HTTP, and Linux feels like SOAP.
Well, until computers become intelligent, they aren't going to ever understand anything, and that's the problem.
Inference isn't understanding. If I tell a computer that "milk" IS A "dairy product" AND "dairy product" COMES FROM "Cow", then, yippee, the computer 'understands' that milk is comes from cows.
But that doesn't help the computer to know what the adjective 'milky' means. It surely does NOT mean 'similar or pertaining to that which comes from cows', or else I could truthfully say 'cow dung is milky', which I can't. The fact is that the linguistic move from noun milk to adjective milky carries with it a vast and subtle payload of meaning that cannot be captured in ontologies. The semantic web fails to address this in any way.
Even ignoring some very basic linguistic problems, people don't agree on the facts. For some people, powdered milk is a dairy product, and is a sub-class of milk. Chef's, for instance, might think that. For others it's not a diary product and is a sub-class of 'dessicated foodstuffs'. Warehouse managers, for instance, might think that.
So, problem number two is that these ontologies (which are basically large clumps of facts and relations about and between classes and instances of things) can only ever be agreed on in narrow domains. And from practical experience, those domains are really very narrow indeed. Not domains like 'medicine' or even 'surgery', but perhaps 'heart surgery'
Now, the semantic web recognises this second problem, of ontologies only working for narrow domains, although I think most proponents fail to realise just how narrow the domains are. The first issue, of deductive reasoning not being very useful, is entirely ignored as far as I can tell.
This is an urban myth. Armour designed for actually fighting in didn't weigh more than 20Kg, and even heavier jousting armour was light enough to walk around in.
Then don't buy sub-standard hardware. And don't blame software for faulty hardware. MS is changing a bunch of defaults that 99% of people won't care about, the 1% that do can set them to whatever they want, and overall we all use less electricity. So what exactly are you complaining about?
"The real test of whether a page is interesting is not whether half a dozen wikipedians think it is, but whether people actually look at it."
Is obviously rubbish. If pages were pulled based on low number of views, it would be catastrophic. It would become a populist publication in the worst sense of the word - plenty of stuff about current celebrities, very little about the etymology of small languages.
The great thing about Wikipedia is that because it has not paper cost, it doesn't have to exclude things based on importance or popularity - only based on accuracy and quality.
Please bear in mind that Wikipedia is not the definitive source of truth in the universe. That fact that person a's opinions differ from person Wikipedia's opinions does not make them de facto incorrect.
Having just returned from a remote part of northern India, I discovered almost everything about the place in Wikipedia is:
a: Plagiarised from the handful of guidebooks that exist b: At least 15 years out of date - just like the guidebooks!
That's not a solution. Are you saying that Vampires exist in Dickensian London? Are you saying that, in the real world, Dracula _isn't_ a Vampire??!!
And that's the tip of the iceberg.
Is powdered milk a dairy product? Can whales sing
I work with ontologies. There are too many contexts, and they are not well defined. You can't reduce human knowledge to an ontology and still have it as being of any use to anyone. Cyc will fail, or, it will succeed and we will have failed.
Gates - creates world's most successful company, becomes world's richest man, leaves day job to spend billions on charity. Us - Made lame borg jokes for 5 years, finally released a browser that's better than IE if you ignore all the unfixed copy/paste bugs. Convinced a few people that Unix sucked less than Windows.
Only people running w2k3 AND linux were allowed to respond. Hmmmmmn, so how many MS shops with an evaluation linux server (installed by their clueless MSCE) were included in this "survey"
Err, that works both ways, doesn't it. Think of all the Linux shops with one little windows server they had to have because some app they needed didn't run on *nix. And IME *nix admins will happily reboot a windows box claiming "it's the only solution" rather than spend 30 minutes actually learning something about how windows works.
*nix admins seem to consider learning windows to be somehow beneath them, while windows admins usually consider learning *nix to be either worthwhile but hard, or simply irrelevant.
The validity of a critique IS dependant on who says it. This is REALITY.
Obviously not true. The validity of an argument can be determined entirely from the sentences that form the argument. Who wrote them down is entirely irrelevant, as should be clear to anyone.
Are you telling me that the opinion of some high school drop-out living in his parents basement, is just as valid as someone who has a degree is sociology and has spent years travelling the world learning about and trying to help with major social issues?
No, but you are now talking about opinions, which are not at all the same kinds of things are arguments or critiques. The reason _who_ espouses an opinion is important, is due to people's inductive reasonsing. They think "I don't know if this opinion is true, but the person holding it has held many other opinions in the past, which did turn out to be true, so the odds are, this one will also turn out to be true".
There is nothing ad hominem about it. Hell, you're not even using "ad hominem" in the correct context! It seems to me like you're simply using sophisticated terms like "ad hominem" or "cognitive dissonance" to make yourself sound more knowledgable than you really are.
There you are right. Ad hominem is attacking the holder of an argument, with the implication that the argument is therefore also somehow attacked. What you were being accused of was really not ad hominem attacks, but appeals to authority, or even appeals to false authority.
In short: "Reading The Times is good for you - This is true because my doctor said so". The appeal is to an authority in medicine, but of course just because an authority in a field said something, doesn't make it true. "Reading The Times is good for you - This is true because Stephen Hawking said so". The appeal is to an authority in another field - a false authority. "Reading the Times is good for you - This is not true, because Stephen Hawking said so, and he's a nasty man who ditched his family". This is an ad hominem attack on Stephen Hawking.
The appeal to authority is sophistry, in the context of formal argument. However, opinions, especially sweeping political opinions, are not really ever constructed as formal arguments, which is why we tend to listen to the views of some people and not others.
The requirement, therefore, is to decide if the authority in question is really one that's worth anything. Does some 25yr old left-wing guy who's spent five years working for Unicef in Ghana really have a more 'valid' opinion on 3rd world poverty than a 35yr old businessman who's spent 5 years reading extensive, detailed, economic reports and essays about developing economies? I suggest it's almost impossible to say, but we can be sure that people on each side will cite the 'authority' they agree with.
They need to lock down their boxes to make sure that their employees don't discover the utility of free software (like firefox).
Oh _that's_ why they are doing it. That figures. Everyone knows, you always give Linux users root access, so they can install all that great free software. And, equally, we know that if you don't have administrator rights on a Windows box, it's impossible to install Firefox.
Really? Is that right? Well, I think Lucy Davis, age 12, currently in a playground somewhere near Cleveland, would like you to give her smart remark back to her.
The snake was more fun the the cubes. And then they came out with all these silly things in different polygons, cylinders, spheres. It got old pretty fast.
Think of it as a very large database of all the people and institutions your company does, has, or would like to do business with. The idea is that if a tech person goes to conference and meets a tech person from potential customer ACME Ltd, and learns that ACME Ltd have been having real trouble with Competitor Ltd's poorly documented API, then tech person can write that down in a database somewhere, and then sales person rings ACME Ltd, and just happens to emphasize how well documented your product is, and how you are offering an amazing upgrade discount to customers who switch from Competitor Ltd's products.
Rather like content management systems, the idea is better on paper than reality, and they are notoriously hard to implement.
It makes sense for MS to release such a beast, IMO, as it fits well with their 'integrate all our products with each other' philosophy.
Slashdot Mirror
"and as companies go Adobe's quite reasonable"
Except, sadly, that they were responsible for asking the Feds to arrest Dmitry Sklyarov. While out of character for the company, the enormity of this is such that I have to think of them as scum for ever after, at least until they actually confess that they were wayyyyy out of line.
http://en.wikipedia.org/wiki/Dmitry_Sklyarov
A miniature stone sculpture of Natalie Portman with no clothes on. That's by far Slashdot's greatest moment, and I want a memorial to it.
This is awful. A huge unwieldy process bound standards organisation is now unable to choose which huge unwieldy standard to give its support to.
"We were really hoping to get the spellcheck dictionary custom term event model moved from 2nd round consultation to initial pre-ratification, but because of byelaw 12 part c regarding type P participants, it looks like we'll be delayed until the *next* meeting of the syntax objects (editing mode) technical steering group. That could be months away." - said Gerhardt Steiner, an XML expert who is a regular voter and sits on several subcommittees.
"Well, I don't understand the fuss" said Sanjeev Raj, "We joined so that our small company that makes custom MSWord plugins for the pharmaceutical industry doesn't go out of business. We paid our membership fee so we could vote on the one technical aspect we care about. I'd love to vote on lots of other things by I've got a company to run."
It's that kind of attitude, says Gerhardt, that's undermining the whole process. "Everyone knows that these kinds of standards organisation are dominated by bearded loons dredged up from the R&D departments of fortune 500 companies, with a few tenured professors thrown in to keep them out of trouble. We work hard to invent hopelessly impractical standards that are overblown, overspecified and impossible to implement unless you are a fortune 500 company with a large R&D department. Even then, you usually have to get a tenured professor to come in and give you some expensive consultancy. It's a fair and equitable system. I think it's a real shame when people try to warp these standards processes for their own selfish financial ends."
Yes, I am a member of one of these committees. No I never vote.
Wow! That's so clever. I love the way you conflate file names and file formats.
Oh, I don't know, I think it's pretty close to impossible. People are working on the easy stuff - names, addresses, events, locations. I think there will be real progress there, and we'll get some useful software. But semantic modelling is extremely hard, because people do not work on semantics. Most people don't, in their heads, categorise things in ways that a computer make make head nor tail of. Semantic models that work for computers require deep heirarchies, relatively few relation types, and fixed degrees of reflexivity, transitivity and symmetry for those relations. People just don't care about that stuff, and probably never will.
The problems I have with writing semantically aware software are all to do with the fact that people aren't semantically aware. No two people will ever categorise things in the same way. You might, in the case of something dull like addresses or phone numbers manage to get them to compromise for the sake of sharing, but you'll never do it for something more complicated that they really care about.
At best, certain tasks will make use of semantic webs (plural!). Maybe light plastics manufacturing will start to use certain semantic interchange standards. Maybe consumer travel will agree some standards for hotels and flights and trains.
That's a silly analogy and always has been. Here's why:
Imagine you take the whole of human civilization as a day. For the first 23 hours we were naked. Then for the next 50 minutes we wore one or two pieces of crude clothing. In just the last 10 seconds we have acquired scarves, tights running shorts. Therefore in the future we will all have millions of items of clothing, and perhaps wear up to 10,000 of them at once.
The fact that for a certain slice of time, there is an exponential curve, just isn't very interesting.
Take volcanoes. For decades they do nothing. Then over a few days they start to output some heat and smoke, then over a few hours they output vast quantities of heat, and then... then they stop. They don't amazingly continue to output energy at an exponential rate until the planet melts.
So what if a bunch of little bipeds on a planet somewhere spent ages picking fruit, and then suddenly figured out flatscreen TV and breast implants? Whose to say the normal curve isn't like the volcano? Why not predict that 5,000 years from now, we will be picking fruit again, but this time with a few myths about an ancient pre-cursor race, which in 12,000 years time will be verified when man-made fragments are discovered deep within the unusual mineral deposits we now call cities.
Although I'm not hands on now, I originally moved from Linux to Solaris (with some Irix stuff in the middle). I still prefer Solaris for the following reason:
Simple is better.
This single thought is perhaps the biggest lesson I've learned in my whole career, about almost any aspect of computing. Complexity is the enemy.
caveat: by 'Linux' I mean 'The particular distro your company has standardised on'
caveat: I'm only concerned with servers. Solaris may be the worst desktop OS in the world FAIK.
1. Less shovelware. Although a base Solaris install is still annoyingly large, it's not nearly as bad as most Linux distros. It infuriates me that operating systems think its useful to install entire database, programming languages, you name its 'just in case you need them'.
2. Better backward compatability. Upgrades to discreet parts of Solaris don't usually require upgrades to other parts of Solaris. This means that you aren't constantly trying to run the latest versions of everything.
3. Better hardware integration. When you are running a lot of servers, it's very useful to have a nice console, so you can talk to the things properly. I think Linux has improved a bit in this area, but I'm not aware that it has an equivalent to the OK prompt, and the various diagnostic tools therein.
Others have talked about various tools and kernel level stuff, but I wanted to make that point that while the Solaris userland might feel archaic to some, to me it feels pleasantly simple - devoid of hidden complexity, obscure features that badly written apps come to rely on, and all the other 'let's have another feature' attitude prevalent in much OS software.
To me, Solaris feels like HTTP, and Linux feels like SOAP.
Five digits? I've got fewer than that, but haven't had mod points in a year. Maybe that's because I never post anything though. Wait, damn....
Well, until computers become intelligent, they aren't going to ever understand anything, and that's the problem.
Inference isn't understanding. If I tell a computer that "milk" IS A "dairy product" AND "dairy product" COMES FROM "Cow", then, yippee, the computer 'understands' that milk is comes from cows.
But that doesn't help the computer to know what the adjective 'milky' means. It surely does NOT mean 'similar or pertaining to that which comes from cows', or else I could truthfully say 'cow dung is milky', which I can't. The fact is that the linguistic move from noun milk to adjective milky carries with it a vast and subtle payload of meaning that cannot be captured in ontologies. The semantic web fails to address this in any way.
Even ignoring some very basic linguistic problems, people don't agree on the facts. For some people, powdered milk is a dairy product, and is a sub-class of milk. Chef's, for instance, might think that. For others it's not a diary product and is a sub-class of 'dessicated foodstuffs'. Warehouse managers, for instance, might think that.
So, problem number two is that these ontologies (which are basically large clumps of facts and relations about and between classes and instances of things) can only ever be agreed on in narrow domains. And from practical experience, those domains are really very narrow indeed. Not domains like 'medicine' or even 'surgery', but perhaps 'heart surgery'
Now, the semantic web recognises this second problem, of ontologies only working for narrow domains, although I think most proponents fail to realise just how narrow the domains are. The first issue, of deductive reasoning not being very useful, is entirely ignored as far as I can tell.
Well, look on the bright side. Neither the states nor the Feds are enforcing capital letters, and you for one must be very glad of that.
This is an urban myth. Armour designed for actually fighting in didn't weigh more than 20Kg, and even heavier jousting armour was light enough to walk around in.
Then don't buy sub-standard hardware. And don't blame software for faulty hardware. MS is changing a bunch of defaults that 99% of people won't care about, the 1% that do can set them to whatever they want, and overall we all use less electricity. So what exactly are you complaining about?
I agree with your post in general, but this:
"The real test of whether a page is interesting is not whether half a dozen wikipedians think it is, but whether people actually look at it."
Is obviously rubbish. If pages were pulled based on low number of views, it would be catastrophic. It would become a populist publication in the worst sense of the word - plenty of stuff about current celebrities, very little about the etymology of small languages.
The great thing about Wikipedia is that because it has not paper cost, it doesn't have to exclude things based on importance or popularity - only based on accuracy and quality.
Please bear in mind that Wikipedia is not the definitive source of truth in the universe. That fact that person a's opinions differ from person Wikipedia's opinions does not make them de facto incorrect.
Having just returned from a remote part of northern India, I discovered almost everything about the place in Wikipedia is:
a: Plagiarised from the handful of guidebooks that exist
b: At least 15 years out of date - just like the guidebooks!
Wheee...
Hey, if it makes them fix the copy/paste bug, it's all good by me.
That's not a solution. Are you saying that Vampires exist in Dickensian London? Are you saying that, in the real world, Dracula _isn't_ a Vampire??!!
And that's the tip of the iceberg.
Is powdered milk a dairy product? Can whales sing
I work with ontologies. There are too many contexts, and they are not well defined. You can't reduce human knowledge to an ontology and still have it as being of any use to anyone. Cyc will fail, or, it will succeed and we will have failed.
Let's see:
Gates - creates world's most successful company, becomes world's richest man, leaves day job to spend billions on charity.
Us - Made lame borg jokes for 5 years, finally released a browser that's better than IE if you ignore all the unfixed copy/paste bugs. Convinced a few people that Unix sucked less than Windows.
Dude, I think *he* won.
Only people running w2k3 AND linux were allowed to respond. Hmmmmmn, so how many MS shops with an evaluation linux server (installed by their clueless MSCE) were included in this "survey"
Err, that works both ways, doesn't it. Think of all the Linux shops with one little windows server they had to have because some app they needed didn't run on *nix. And IME *nix admins will happily reboot a windows box claiming "it's the only solution" rather than spend 30 minutes actually learning something about how windows works.
*nix admins seem to consider learning windows to be somehow beneath them, while windows admins usually consider learning *nix to be either worthwhile but hard, or simply irrelevant.
The validity of a critique IS dependant on who says it. This is REALITY.
Obviously not true. The validity of an argument can be determined entirely from the sentences that form the argument. Who wrote them down is entirely irrelevant, as should be clear to anyone.
Are you telling me that the opinion of some high school drop-out living in his parents basement, is just as valid as someone who has a degree is sociology and has spent years travelling the world learning about and trying to help with major social issues?
No, but you are now talking about opinions, which are not at all the same kinds of things are arguments or critiques. The reason _who_ espouses an opinion is important, is due to people's inductive reasonsing. They think "I don't know if this opinion is true, but the person holding it has held many other opinions in the past, which did turn out to be true, so the odds are, this one will also turn out to be true".
There is nothing ad hominem about it. Hell, you're not even using "ad hominem" in the correct context! It seems to me like you're simply using sophisticated terms like "ad hominem" or "cognitive dissonance" to make yourself sound more knowledgable than you really are.
There you are right. Ad hominem is attacking the holder of an argument, with the implication that the argument is therefore also somehow attacked. What you were being accused of was really not ad hominem attacks, but appeals to authority, or even appeals to false authority.
In short:
"Reading The Times is good for you - This is true because my doctor said so". The appeal is to an authority in medicine, but of course just because an authority in a field said something, doesn't make it true.
"Reading The Times is good for you - This is true because Stephen Hawking said so". The appeal is to an authority in another field - a false authority.
"Reading the Times is good for you - This is not true, because Stephen Hawking said so, and he's a nasty man who ditched his family". This is an ad hominem attack on Stephen Hawking.
The appeal to authority is sophistry, in the context of formal argument. However, opinions, especially sweeping political opinions, are not really ever constructed as formal arguments, which is why we tend to listen to the views of some people and not others.
The requirement, therefore, is to decide if the authority in question is really one that's worth anything. Does some 25yr old left-wing guy who's spent five years working for Unicef in Ghana really have a more 'valid' opinion on 3rd world poverty than a 35yr old businessman who's spent 5 years reading extensive, detailed, economic reports and essays about developing economies? I suggest it's almost impossible to say, but we can be sure that people on each side will cite the 'authority' they agree with.
They need to lock down their boxes to make sure that their employees don't discover the utility of free software (like firefox).
Oh _that's_ why they are doing it. That figures. Everyone knows, you always give Linux users root access, so they can install all that great free software. And, equally, we know that if you don't have administrator rights on a Windows box, it's impossible to install Firefox.
And someone gave you an 'insightful'. Geez.
Really? Is that right? Well, I think Lucy Davis, age 12, currently in a playground somewhere near Cleveland, would like you to give her smart remark back to her.
The snake was more fun the the cubes. And then they came out with all these silly things in different polygons, cylinders, spheres. It got old pretty fast.
:-)
Pocketeers were a much better toy craze
Think of it as a very large database of all the people and institutions your company does, has, or would like to do business with. The idea is that if a tech person goes to conference and meets a tech person from potential customer ACME Ltd, and learns that ACME Ltd have been having real trouble with Competitor Ltd's poorly documented API, then tech person can write that down in a database somewhere, and then sales person rings ACME Ltd, and just happens to emphasize how well documented your product is, and how you are offering an amazing upgrade discount to customers who switch from Competitor Ltd's products.
Rather like content management systems, the idea is better on paper than reality, and they are notoriously hard to implement.
It makes sense for MS to release such a beast, IMO, as it fits well with their 'integrate all our products with each other' philosophy.
Well, it sounds to me like someone wants to retain control of their website that they paid for. Big deal.
Imagine if Fred did satirical send ups of well known gay people. Now would you be happy for him to use your logo all over it?