Not only that, but you can't really trust the figures Microsoft gives out regarding sales of their current product lineup. Office 2007 for instance. Due to a big increase in hiring, we needed a few dozen more copies of Office last year, that were not covered by our original site license. We currently run Office 2002 (XP). Now, the Microsoft sales rep I spoke to said, oh, sorry, we don't sell Office 2002 any more, BUT, you can purchase a few dozen copies of 2007, and you can then use 2002 on those machines instead. We did the same thing for OneNote. Additionally, pretty much all of the machines we get from Toshiba now come with Vista pre-installed. Well, the first thing we do is remove it and slap on XP, but right there, we've just generated a Vista sale. So if anyone else is like us (and I suspect that they are), Microsoft's figures for new software are at least misleading.
I don't argue your main point, which is how folk tales become mythos which become organized religion, but we currently have no evidence that Jesus really existed. Your explanation is plausible, of course (and my opinion is that it is even likely), but to say that Jesus' existence, life, and death are fact is to entirely miss the concept of fact-hood. Facts are verified information, which means we have evidence to support them. We don't currently have any evidence to support the claim that Jesus existed, thus it is not a fact that Jesus existed.
I'm not sure if you're being witty or just naive, but this really does appear to be a general software engineering strategy that works. I don't know much about how Windows' kernel works, so I can't say whether their implementation is any good-- I suspect that their business imperative to provide backward compatibility and rich APIs have probably hindered their efforts on the security front.
But if you go out and look at software that is written to be secure, the subsystem approach is how it is done. Postfix, for example, is actually a collection of simple applications. One application does queueing, one specializes in spewing SMTP, one specializes in receiving SMTP, and so on. Also, system call policy enforcement mechanisms (ala systrace) and privilege separation (like in Apache or SSH) can be formally verified to work. I think UIUC is on the right track here. Whether their browser becomes THE web browser is somewhat unimportant, since they're researching an area of security that has had a fair amount of attention from good programmers but not computer scientists. In some ways this is the ultimate in enforcing "object-oriented"-ness: code isn't just a collection of modules, the application is a collection of small applications, too.
All the poo-pooing elitist crap about "Acrobat being bloated" is little more than trolling for mod points (since it's obviously just group-think anymore). The xpdf binary for Linux, statically linked to Motif, t1lib, and FreeType, is about 6.3MB. Source code is about 600k. It's snappy and does everything I want. So, speaking of poo, Acrobat is a steaming pile of it.
I think Apple's shitty Performa machines from the mid-90's did a lot to spread Apple-hate among first-time Apple users. Unfortunately, since these machines were cheap, many middle and high schools bought them, and the first-time users in question were young, impressionable men. Having gone through public school prior to this period, and having been exposed to the Macintosh when MSDOS machines were all the rage, I had a very high opinion of the Mac. But my younger brother was comparing Performas against early PIIs running Windows-- so despite the hacked-togetherness of Win98 and the general shoddiness of Windows apps at the time, he came to hate the underperforming "candy-coated" Macs at school.
It's true that those requirements are rather steep, but keep in mind that the D block is an auction for a special piece of the spectrum-- it is going to be used not just for the private sector, but also for public safety. With that in mind, the requirements don't seem as bad. Also, since this will be a public-private partnership, with public safety presumably as a captive customer, there's some reason to think that the D block would be a solid money-maker, and that private services offered on top of the public ones would come "for free", since building infrastructure for the public services would also serve the private. Ten years for 99% penetration does light a fire under the winning bidder's ass, but then again, the taxpayer will probably be paying for this in the long run, so in that context, ten years doesn't sound so bad to me...
If I recall correctly, from reading 1491: New Revelations of the Americas Before Columbus, one hypothesis as to why Native American populations had such low disease resistance to common European diseases such as smallpox, measles, tuberculosis, etc, was that almost all Native Americans shared the same haplogroup. Thus, genetic diversity was fairly low. Additionally, the author said that their immune systems were optimized to resist parasitic infections rather than bacteriological/viral infections (I am not a biologist, so take my interpretation with a grain of salt). The author speculates that the reason Europeans had these defenses were precisely because their populations were routinely devastated by disease. So your comment is spot-on-- most of the people living in the world today have some kind of resistance to these diseases. That said, I'm glad that the WHO essentially eradicated smallpox, so I have much to thank science for as well.
Also, this is somewhat akin to the U.S. Merchant Marine, which is nonmilitary during peacetime, but becomes a branch of the military during wartime. So this is a more recent example. Not that we couldn't just do what we wanted anyway. Laws are what you make them. Our current administration has demonstrated this point, repeatedly. Keep in mind that laws and ethics (e.g., "founding principles") are two different things; preferably they overlap, but if not, hopefully they are justifiable.
Little Snitch for Mac OS X lets you write per-application firewall rules. It's pretty sweet. Not that this will help you if your favorite application is secretly sending your diary to your mom.
Thing is, these patently false and self-serving claims seem to have a big effect on a very important group of people: CIOs. I'm not saying that CIOs are stupid people per se, but in my purely anecdotal experience, there's been a strong correlation. It's like they lack the brain structure responsible for critical thinking or something...
You can arguably pick d) for any branch of knowledge that relies on making generalizations (except the ones that do so mathematically, i.e., through mathematical induction). Being able to make generalizations seems to be a key characteristic of intelligent creatures. Obviously, there's a real balancing act here, and this is the very point of statistical methods, to tell us how confident we can be in our generalization. But I've found among my peers that it is virtually impossible to have a rational discussion about a culture in any way that might be construed as being judgmental, even among scientists. This kind of knee-jerk reaction makes talking about any topic in sociology an exercise in frustration.
That said, the line about fashion magazines is stupid. How many women even read fashion magazines? My girlfriend (who will soon be a medical doctor) hates them. "Men's magazines" are the same kind of crap, but you don't see generalizations being made about them because of them. Maxim is essentially a hundred-page ad, telling you what to wear, how to act, and what to think. It even smells of cologne. It just uses less pink ink and more black than women's magazines.
I stand corrected. I was mistaken-- the person who I thought was a paid peer reviewer was actually a paid editorial freelancer, for a science publication. You're right-- I do not work in the scientific branch of our publishing company (although we have one, and once shared office space with them). I work in the college textbook business, and even then, I am only an IT person. I've simply been around long enough in that capacity to see how certain things are done, since I have contact with all facets of the business. I am currently training as a scientist, in school at night, but I am not one yet.
But my mistake does not invalidate my main point, being that peer review ala Wikipedia, and peer review ala corporate publishing are not all that different. People are who they are-- I don't know if we're ever see a publishing model that is truly honest. The best thing I can come up with at the moment, is a mixture of all of the above, in the hope that in a free exchange of ideas, the truth will win out. I can't put my faith in any one model, and I recognize that Wikipedia neatly trades one set of epistemological problems for another.
My personal opinion is that the more someone defends their 'expertise', the less trustworthy they are. An expert is someone who has not just a great deal of knowledge, but an appreciation for the failings of their knowledge. Wikipedia and its ilk is not trendiness-- it is a rational acknowledgement that human understanding is limited. People who feel threatened by Wikipedia should ask themselves why they feel that way.
You're kidding, right? Apparently you've missed out on Margaret Jones, or James Frey, or the entire bogus memoir industry that produces crap like this with the help of a ghost writer. I work for a publisher, and simply put, they rarely fact-check. Instead, what they do is send prerelease books to reviewers. The hope is that the reviewers will be smart enough to catch glaring errors. How knowledgeable the reviewers are depends somewhat on the audience of the book. College textbooks typically go to professors and grad students. Trade paperbacks can go to pretty much anybody, but usually quotable people or professional book critics.
In any case, this is exactly the same mechanism that Wikipedia uses: throw it out there and see if anyone catches something. As a practical matter, publishers cannot fact-check. They do not have the resources. The only books I would depend on fact-checking for are the ones that claim to do so as a principle of their cognitive authority: dictionaries and encyclopedias. The imprint I work for publishes several hundred textbooks a year, and reprints darn near a thousand. We have a little over 200 employees. See what I'm getting at?
Even scientific articles are "fact-checked" this way: throw it out there. Typically the reviewers are peers, and quite knowledgeable. This works better than with trade publishers because the reviewers have specific knowledge about that particular field. But does the publisher fact-check themselves? No! I should add that the pay scale for reviewers goes up depending on the relative reliability of the reviewers. Reviewers for scientific reviewers are often paid in the several hundreds range. Reviewers for college textbooks in the low hundreds (sometimes in trade for other goodies), and trade paperback reviewers, not much, if anything. Often it's for the privilege of seeing pre-release stuff.
There's only one kind of publishing where fact-checking (aside from dictionaries, etc.) is done as a rule: journalism. But there have been many scandals there as well. There was a study mentioned in the book Trust Us, We're Experts that said that nearly half of the Wall Street Journal's article's were simply slightly modified press releases. And the Wall Street Journal is regarded as one of the more reliable papers! I think I only need to mention cable TV journalism for you to see where I'm going with this.
The publishing industry is not reliable. They're in it for the money. Books like Frey's sell just as well, if not better, than the real ones. Just look at the demand for O.J. Simpson's book-- a book that never even claimed to tell the truth! People want something juicy, and the publishing industry is happy to give it to them. Sorry, ptrourke, your premise is false.
Yeah, exactly. Maybe Silverlight will light the fire under Adobe's collective ass and make them go, "SHIT! We need market saturation, like, NOW. Let's open-source this biatch! We need UNIX fanatics on our side!"
I definitely look forward to running Flash natively on my OpenBSD machine. I know, wishful thinking. Adobe's solution will probably be to ratchet up the price and require that you hang a dongle out your ass-- as if it wasn't piracy that essentially established their competitor's dominance in the DTP industry. Funny that people fail to see that piracy is only the next best thing compared to open source if you want to see your product become the standard. But it's this whole proprietary mindset that's the problem...
The odd thing is that de Icaza is a Real Hacker (tm) having worked on Gnumeric and GNOME among other things. And he knows that "Web 2.0" currently depends on either a cumbersome but open setup like AJAX, or something simpler but proprietary, like Flash or Silverlight. With the latter, proprietary type, the OS is not irrelevant at all. Since he is currently working on porting Silverlight to Linux, how could he not know this? This, combined with other things he's said in the past (like supposedly having tried to convince MS management to release their IE browser code), makes me think that he's either a liar or that he has some ulterior motive.
Slashdot Mirror
I just figured that maybe my sense of humor was so atrocious, it crashed trying to think of one.
Not only that, but you can't really trust the figures Microsoft gives out regarding sales of their current product lineup. Office 2007 for instance. Due to a big increase in hiring, we needed a few dozen more copies of Office last year, that were not covered by our original site license. We currently run Office 2002 (XP). Now, the Microsoft sales rep I spoke to said, oh, sorry, we don't sell Office 2002 any more, BUT, you can purchase a few dozen copies of 2007, and you can then use 2002 on those machines instead. We did the same thing for OneNote. Additionally, pretty much all of the machines we get from Toshiba now come with Vista pre-installed. Well, the first thing we do is remove it and slap on XP, but right there, we've just generated a Vista sale. So if anyone else is like us (and I suspect that they are), Microsoft's figures for new software are at least misleading.
I don't argue your main point, which is how folk tales become mythos which become organized religion, but we currently have no evidence that Jesus really existed. Your explanation is plausible, of course (and my opinion is that it is even likely), but to say that Jesus' existence, life, and death are fact is to entirely miss the concept of fact-hood. Facts are verified information, which means we have evidence to support them. We don't currently have any evidence to support the claim that Jesus existed, thus it is not a fact that Jesus existed.
Funny, but that's not logical. A simple Venn diagram can show this without having to go into a formal proof.
I'm looking forward to the inevitable Photoshop vs. GIMP flamefest on this one.
Look at how they come up with them.
I'm not sure if you're being witty or just naive, but this really does appear to be a general software engineering strategy that works. I don't know much about how Windows' kernel works, so I can't say whether their implementation is any good-- I suspect that their business imperative to provide backward compatibility and rich APIs have probably hindered their efforts on the security front.
But if you go out and look at software that is written to be secure, the subsystem approach is how it is done. Postfix, for example, is actually a collection of simple applications. One application does queueing, one specializes in spewing SMTP, one specializes in receiving SMTP, and so on. Also, system call policy enforcement mechanisms (ala systrace) and privilege separation (like in Apache or SSH) can be formally verified to work. I think UIUC is on the right track here. Whether their browser becomes THE web browser is somewhat unimportant, since they're researching an area of security that has had a fair amount of attention from good programmers but not computer scientists. In some ways this is the ultimate in enforcing "object-oriented"-ness: code isn't just a collection of modules, the application is a collection of small applications, too.
I think Apple's shitty Performa machines from the mid-90's did a lot to spread Apple-hate among first-time Apple users. Unfortunately, since these machines were cheap, many middle and high schools bought them, and the first-time users in question were young, impressionable men. Having gone through public school prior to this period, and having been exposed to the Macintosh when MSDOS machines were all the rage, I had a very high opinion of the Mac. But my younger brother was comparing Performas against early PIIs running Windows-- so despite the hacked-togetherness of Win98 and the general shoddiness of Windows apps at the time, he came to hate the underperforming "candy-coated" Macs at school.
There are a lot of people on the planet.
It's true that those requirements are rather steep, but keep in mind that the D block is an auction for a special piece of the spectrum-- it is going to be used not just for the private sector, but also for public safety. With that in mind, the requirements don't seem as bad. Also, since this will be a public-private partnership, with public safety presumably as a captive customer, there's some reason to think that the D block would be a solid money-maker, and that private services offered on top of the public ones would come "for free", since building infrastructure for the public services would also serve the private. Ten years for 99% penetration does light a fire under the winning bidder's ass, but then again, the taxpayer will probably be paying for this in the long run, so in that context, ten years doesn't sound so bad to me...
Fortunately for us, making a thing simpler than "as simple as possible" is actually impossible ;^)
If I recall correctly, from reading 1491: New Revelations of the Americas Before Columbus, one hypothesis as to why Native American populations had such low disease resistance to common European diseases such as smallpox, measles, tuberculosis, etc, was that almost all Native Americans shared the same haplogroup. Thus, genetic diversity was fairly low. Additionally, the author said that their immune systems were optimized to resist parasitic infections rather than bacteriological/viral infections (I am not a biologist, so take my interpretation with a grain of salt). The author speculates that the reason Europeans had these defenses were precisely because their populations were routinely devastated by disease. So your comment is spot-on-- most of the people living in the world today have some kind of resistance to these diseases. That said, I'm glad that the WHO essentially eradicated smallpox, so I have much to thank science for as well.
Also, this is somewhat akin to the U.S. Merchant Marine, which is nonmilitary during peacetime, but becomes a branch of the military during wartime. So this is a more recent example. Not that we couldn't just do what we wanted anyway. Laws are what you make them. Our current administration has demonstrated this point, repeatedly. Keep in mind that laws and ethics (e.g., "founding principles") are two different things; preferably they overlap, but if not, hopefully they are justifiable.
Little Snitch for Mac OS X lets you write per-application firewall rules. It's pretty sweet. Not that this will help you if your favorite application is secretly sending your diary to your mom.
Thing is, these patently false and self-serving claims seem to have a big effect on a very important group of people: CIOs. I'm not saying that CIOs are stupid people per se, but in my purely anecdotal experience, there's been a strong correlation. It's like they lack the brain structure responsible for critical thinking or something...
Exactly!
Yeah, purring did wonders for my asthma. It cured me of this affliction called 'breathing'.
You can arguably pick d) for any branch of knowledge that relies on making generalizations (except the ones that do so mathematically, i.e., through mathematical induction). Being able to make generalizations seems to be a key characteristic of intelligent creatures. Obviously, there's a real balancing act here, and this is the very point of statistical methods, to tell us how confident we can be in our generalization. But I've found among my peers that it is virtually impossible to have a rational discussion about a culture in any way that might be construed as being judgmental, even among scientists. This kind of knee-jerk reaction makes talking about any topic in sociology an exercise in frustration.
That said, the line about fashion magazines is stupid. How many women even read fashion magazines? My girlfriend (who will soon be a medical doctor) hates them. "Men's magazines" are the same kind of crap, but you don't see generalizations being made about them because of them. Maxim is essentially a hundred-page ad, telling you what to wear, how to act, and what to think. It even smells of cologne. It just uses less pink ink and more black than women's magazines.
I stand corrected. I was mistaken-- the person who I thought was a paid peer reviewer was actually a paid editorial freelancer, for a science publication. You're right-- I do not work in the scientific branch of our publishing company (although we have one, and once shared office space with them). I work in the college textbook business, and even then, I am only an IT person. I've simply been around long enough in that capacity to see how certain things are done, since I have contact with all facets of the business. I am currently training as a scientist, in school at night, but I am not one yet.
But my mistake does not invalidate my main point, being that peer review ala Wikipedia, and peer review ala corporate publishing are not all that different. People are who they are-- I don't know if we're ever see a publishing model that is truly honest. The best thing I can come up with at the moment, is a mixture of all of the above, in the hope that in a free exchange of ideas, the truth will win out. I can't put my faith in any one model, and I recognize that Wikipedia neatly trades one set of epistemological problems for another.
My personal opinion is that the more someone defends their 'expertise', the less trustworthy they are. An expert is someone who has not just a great deal of knowledge, but an appreciation for the failings of their knowledge. Wikipedia and its ilk is not trendiness-- it is a rational acknowledgement that human understanding is limited. People who feel threatened by Wikipedia should ask themselves why they feel that way.
Well, to dumb statisticians, or ones with agendas, anyway.
You're kidding, right? Apparently you've missed out on Margaret Jones, or James Frey, or the entire bogus memoir industry that produces crap like this with the help of a ghost writer. I work for a publisher, and simply put, they rarely fact-check. Instead, what they do is send prerelease books to reviewers. The hope is that the reviewers will be smart enough to catch glaring errors. How knowledgeable the reviewers are depends somewhat on the audience of the book. College textbooks typically go to professors and grad students. Trade paperbacks can go to pretty much anybody, but usually quotable people or professional book critics.
In any case, this is exactly the same mechanism that Wikipedia uses: throw it out there and see if anyone catches something. As a practical matter, publishers cannot fact-check. They do not have the resources. The only books I would depend on fact-checking for are the ones that claim to do so as a principle of their cognitive authority: dictionaries and encyclopedias. The imprint I work for publishes several hundred textbooks a year, and reprints darn near a thousand. We have a little over 200 employees. See what I'm getting at?
Even scientific articles are "fact-checked" this way: throw it out there. Typically the reviewers are peers, and quite knowledgeable. This works better than with trade publishers because the reviewers have specific knowledge about that particular field. But does the publisher fact-check themselves? No! I should add that the pay scale for reviewers goes up depending on the relative reliability of the reviewers. Reviewers for scientific reviewers are often paid in the several hundreds range. Reviewers for college textbooks in the low hundreds (sometimes in trade for other goodies), and trade paperback reviewers, not much, if anything. Often it's for the privilege of seeing pre-release stuff.
There's only one kind of publishing where fact-checking (aside from dictionaries, etc.) is done as a rule: journalism. But there have been many scandals there as well. There was a study mentioned in the book Trust Us, We're Experts that said that nearly half of the Wall Street Journal's article's were simply slightly modified press releases. And the Wall Street Journal is regarded as one of the more reliable papers! I think I only need to mention cable TV journalism for you to see where I'm going with this.
The publishing industry is not reliable. They're in it for the money. Books like Frey's sell just as well, if not better, than the real ones. Just look at the demand for O.J. Simpson's book-- a book that never even claimed to tell the truth! People want something juicy, and the publishing industry is happy to give it to them. Sorry, ptrourke, your premise is false.
Yeah, exactly. Maybe Silverlight will light the fire under Adobe's collective ass and make them go, "SHIT! We need market saturation, like, NOW. Let's open-source this biatch! We need UNIX fanatics on our side!"
I definitely look forward to running Flash natively on my OpenBSD machine. I know, wishful thinking. Adobe's solution will probably be to ratchet up the price and require that you hang a dongle out your ass-- as if it wasn't piracy that essentially established their competitor's dominance in the DTP industry. Funny that people fail to see that piracy is only the next best thing compared to open source if you want to see your product become the standard. But it's this whole proprietary mindset that's the problem...
Wow! How did I fall for that?
The odd thing is that de Icaza is a Real Hacker (tm) having worked on Gnumeric and GNOME among other things. And he knows that "Web 2.0" currently depends on either a cumbersome but open setup like AJAX, or something simpler but proprietary, like Flash or Silverlight. With the latter, proprietary type, the OS is not irrelevant at all. Since he is currently working on porting Silverlight to Linux, how could he not know this? This, combined with other things he's said in the past (like supposedly having tried to convince MS management to release their IE browser code), makes me think that he's either a liar or that he has some ulterior motive.