A good deal of the character of fine wines and spirits comes precisely from the "impurities," which are actually just the various flavors present in the wine. The goal (my statements here actually apply to Scotch Whisky, with which I'm more familiar, but should generalize to wine) is to produce something with an intricate, multifaceted flavor -- exactly the opposite of what most beverage manufacturers (Coke, Pepsi, etc.) are trying to do. These sorts of "impurities" are the sort of things that set a painting apart from a photograph, or a live musician apart from a MIDI performance.
Of course, a large amount of expensive wine and spirits (likely the majority) end up being purchased by wealthy people more interested in showing off their sophistication than actually drinking the stuff; the true connoisseurs are more likely to be the college student who trades seeing movies for a couple months for a single bottle which he finishes in a couple days or the regular middle-class guy who feels somewhat uncomfortable when he goes to buy a bottle of wine at an unnecessarily metrosexual storefront than the Paris Hiltons of the world.
If you open up the original iPod firmware in a hex editor with the proper number of columns, the first thing you see is an ASCII-Art stop sign and a scary legal message. This is nothing new.
Re:Hard to admit, but that is quite clever
on
Sober Code Cracked
·
· Score: 1
Perhaps "havoc" isn't an entirely appropriate term. According to your link, Zotob only affects computers running unpatched versions of Windows 2000. According to most usage statistics, Windows 2000 only has about 10 - 15% of the operating system market, and that's patched and unpatched systems. Further, while I haven't checked this I'll bet that Microsoft uses Windows 2000 to run their spiders for MSN search, meaning that the actual number is probably even lower. Now, while having 7% of desktops infected by a worm is annoying, the only way that this would count as "havoc" is if you were managing a corporate network full of identical W2K boxes, and in any case that's nowhere near the sort of damage that is routinely unleashed by social-engineering bots.
Re:Hard to admit, but that is quite clever
on
Sober Code Cracked
·
· Score: 3, Insightful
First, I have a hard time believing that a professor took students from being "computer newbies" to being able to print out "hello world" ten times in thirty days, much less write some sort of working virus; trying to teach students anything outside of their major is roughly equivalent to pushing dead whales uphill in terms of efficiency. I've been in a lot of classes and taught a few, and I know that the average student will not do any work if it's at all plausible that a significant number of other students won't do it either -- school these days has become a generalized prisoner's dilemma situation, in which the teacher can only fail so many students before being reassigned.
In the larger scope, I'll just say that it's very tempting to think that one's computer programs just scale automatically, but this is simply not the case. Chances are that you were working on a very homogeneous network at that point, which most machines running rollout-synchronized versions of the same software. I've written "worms" that work under such an environment myself -- unlocking the parental protection on the middle-school computers made lunch-time in the library a lot more interesting. In such a situation, a worm either doesn't spread at all or immediately takes over the entire network, so any success is an impressive one.
On the real internet, on the other hand, we have a very complicated mesh of various systems with different sorts of protections, some explicitly designed as such but most just due to random variations that prevent a given buffer overflow from working on more than one system. Even if someone is running a vulnerable system somewhere out there, there's a good chance that getting at it may involve going past some other system that is simply going to eat it alive. We're not talking just about computers, but also about routers, switches, and all that Cisco equipment that's silently running a good deal of the net without anyone ever thinking about it.
That's why there hasn't been a real worm on the internet in quite a while; essentially every major virus in recent memory has relied on social-engineering to trick the user into manually installing the virus onto his own computer. In fact, I'd seriously doubt that it's even feasible to create a self-distributing worm on the internet at this point, unless Microsoft is dumb enough to build remote-execution capability into their application software again.
Of course, if you were actually working on a diverse, real-world type network, and you managed to devise cross-platform vectors, that's quite different and it'd be interesting to hear about. But if you're like the majority of people who make claims like these, I'm gonna have to say that your eyes are probably a little bigger than your mouth on this one.
A single incident is not an example of natural selection, any more than the collision of two particles is an example of the ideal gas law. Common sense is not genetic, and what we consider common sense isn't necessarily anti-survival. I'll bet that this kid, had he lived, would have quickly begun engaging in unprotected sex with multiple partners, which would be doing a lot more to promote his genetic material than most of us are doing. Evolution is a scientific theory just like Maxwell's equations or quantum mechanics. If you wouldn't feel comfortable talking about the latter, you likely talk about the former at your peril.
My comments were meant humorously. Of course, in an age when you can patent software and/or condemn buildings in order to sell them to private owners, I wouldn't be surprised by much of anything anymore.
What do you want to bet that once the 2.7 kernel does come out that SCO points at the IBM contributions and claims perjury? Actually, that's probably the strategy they're going with here in the first place.
Actually, if you adjust for inflation, it comes in a little lower. Still, $380 million is absurd; that movie didn't deserve to make more than about a buck fifty.
I never understand why people pay so much attention to articles like this. Dan Zambonini runs a small company that nobody has ever heard of which makes a content management system used by a bunch of people you've never heard of. I suppose that's a little better than Joel Spolsky, who makes software that *nobody* uses, but these guys really don't have much of a clue what kinds of jobs are out there in the companies that, you know, you may have heard of before -- they're too busy running their garage operations.
I've talked to recruiters from companies like Google, Microsoft, Symantec, etc. and while they do want someone who can actually sit down and write a program, if you are going to write software you are going to need to understand things like analysis of algorithms or else you are going to end up putting bubblesorts into production code and leaving users wondering why your search feature takes so much longer than your competitors'.
Of course, if you want to work in the database department for a large company and write software that moves tables of numbers around and is never seen by the outside world, then by all means go into MIS. But if you, like most aspiring programmers, want to release software onto the general market and you want to write something other than very basic utilities like "Windows Power Tools" or "Texteditor Extreme 5.3," you are going to have to have a foundation in computer science and a whole bunch of math.
The tone of this article is refreshing, likely because it doesn't deal directly with computers. Every time an independant researcher discovers a buffer overflow exploit, he's branded a criminal by the industry and the media play along. It's time we start to demand that articles about security put the researcher and not the faceless corporation in the protagonist's role.
Yeah... did you happen to notice the little "Search this site with Google" box on the Nutch homepage? I'm a *little* wary of a search spider that doesn't even index its own homepage!
I used Altavista's raging.com for as long as it existed; when they finally took that away I reluctantly moved to this "google" thing that people had been talking about.
Now that I know what's going on, however, I'd say the pagerank algorithm really is so important that it supercedes everything that came before it. Altavista was impressive because of its spiders and the magnitude of its database, but it was often unable to make any sense of the relevancy of various pages. Google, on the other hand, can usually find exactly what you're looking for on the first result.
Search: Google's Pagerank concept radically changed the way that search engines determined which results were relevant. While previous services were based on human rankings or on how many times a particular word was listed on the page, Google put out an automated system which was able to deliver more relevant results when confronted with normal sites and, by its very design, much harder to exploit with SEO techniques. Further, Google continually tweaks the parameters of their search -- if you can go to one of Norvig's talks about the sorts of stuff they do, it's amazing.
Maps: That interface -- scrolling, markers, and all -- is done entirely in javascript. No plugins, no flash, no helpers. Nobody thought that that sort of thing was even possible.
GMail: I don't use it, so I can't comment. But I do have around 1 GB of email on my primary account. When you use email for serious work, it can add up.
Google Groups: It's my group reader. I like it because it shows the discussions in thread format from the top and supresses the quoting that can make USENET discussions turn into pages and pages of greater-than symbols.
As to your assertion that Google hasn't ushered in a new age, I disagree. Ten years ago, when someone wanted information they went to a library, an encyclopedia, or maybe a CD-ROM. Now, any time anyone wants to know anything, they go immediately to Google and chances are that the information will come up on the first page.
Lest you've forgotten, it was Napster and Winamp that 0popularized mp3's, not the iPod, and COBOL, not Oracle, that popularized the database. So I'd respond to you, "Stop the misinformation campaign."
Not so. Producing hydrogen by hydrolosis requires more energy than it produces (as do all processes), so the energy put into the hydrogen would have to be supplied somewhere else -- presumably by more fossil fuel.
The only advantage that hydrogen has over fossil fuels is that fossil fuels can be burned out away from population centers to produce hydrogen; fuel-cell based cars could reduce pollution in large cities at the expense of even more pollution at some factory.
The folks at gizmodo are easily amused...
on
Making Fire From Water
·
· Score: 1, Insightful
Seriously, didn't everyone see this as a demonstration in high-school chemistry? This isn't exactly that new or exciting...
I think the issue is when someone writes an article about abortion in the American legal system and someone attempts to replace it with several copies of the word "murder." Other good targets for stasis would be pages pertaining to evolutionary biology, the holocaust, and similar areas where there is a small, vocal lunatic fringe which is militant in publicizing its incorrect dogma.
Scientists and mathematicians are some of the most creative people around -- math itself is a fine art on par with music, painting, or drama; science is slightly more applied, but the art of designing experiments to give precisely the results needed surely requires more creativity than churning out a summer exploitation movie.
For just one of many, many examples, consider Henry Cavendish's "weighing the earth" experiment. A run-of-the-mill screenwriter just doesn't come up with ideas as good as torsion balances.
I'm always glad to see someone criticizing the Mozart-worship in "classical music" circles. Mozart was nowhere near the equal of his teacher Haydn nor his student Hummel, and yet we hear more about Mozart these days than those two combined. Surely he came out with a few good pieces -- the Rondo alla Turca comes to mind -- but for some reason people feel inclined to perform every single piano work the man ever came out with, most of which are just outright annoying to listen to.
We used Excel's built-in LINEST function on the same data on three different computers and got three different numbers. The function which did not work properly was not written by a third party but by Microsoft. The computers, for reference, were all PowerPC-based Macs with various versions of Mac OS 9, and there were various versions of Excel installed.
I then wrote a little macro which calculated the linear regression which gave identical, correct results on each computer. I didn't go to the trouble of sending an error report to Microsoft since I don't use Excel in my day-to-day business -- I had to use the computers in the lab that day because the alternative would have been walking home and doing the data analysis there.
To recap, our home-rolled, off-the-cuff macro worked; the Excel builtin function didn't.
I have a friend who is currently working as a quant at one of the well-known Wall Street analysts who occasionally asks me for math advice (I'm a mathematician). Recently he told me that his work invoved an Excel spreadsheet which took upwards of forty minutes to calculate. Now, perhaps this was the fault of some atrocious flaw in his code, but knowing him I doubt it. All in all I've been decidedly unimpressed with Excel. I'm sure that it's *useable* for serious work, but it certainly doesn't meet the sort of standards that professional-grade software should.
IMHO, historical notability can be equated with producing something that changed the way things were done in a positive manner. Examples would be Fred Brooks, Minsky, Kernighan, Richie, Pike, Alan Kay, etc.
The author of this article, on the other hand, has done nothing of any such noteworthiness. His main "achievement," if it can be called that, was designing and managing the implementation of a BASIC dialect in the 1990's. From his resume:
"Program Manager, Microsoft Excel. Part of the team that designs the world's best selling spreadsheet application. Drove a new macro language strategy for Microsoft (Visual Basic for Applications)..."
Now, most professionals consider Excel inadequate for any serious work, a point that was illustrated recently in a chemistry lab I was working in when three different computers running three different versions of Excel returned three different results for a linear regression. Upon writing an Excel macro which calculated linear regressions correctly, we found that all three computers were giving incorrect results. As data fitting is one of the most basic functions you could want in a spreadsheet and linear regression is the simplest form, I'm decidedly unimpressed by anyone who worked on Excel.
Unless this Joel fellow has some sort of long-forgotten historical notability in software engineering, I fail to see why his articles continually show up here. His company, which he repeatedly plugs in his articles, has put out two pieces of software -- a bugtracker and a content management system -- which nobody I've talked to has ever heard of. Does Joel have some insight into programming that everyone reading Slashdot does not?
I suppose, of course, that the same could be said for most tech journalists, most of whom would have a hard time compiling a source tarball. On the other hand, usually tech journalists are reporting on companies' press releases, not writing editorials about software design.
With the exception of Aesop's Fables and medieval morality plays, most good pieces of fiction are not generally built around a single "moral." There are themes throughout a work, but usually a serious author does not write a story for the express purpose of advancing a message. Upton Sinclair felt strongly enough about this that he prefaced The Jungle by saying that unlike actual literature, his book was written for the purpose of conveying a message.
I haven't read this book and I likely won't; it sounds too much like a the computer-crime version of a Tom Clancy or Stephen King novel. There are probably several flaws in it, but the reasons you cite would rule out e.g. Thomas Pynchon's Entropy, one of the classics of modern American literature.
Slashdot Mirror
A good deal of the character of fine wines and spirits comes precisely from the "impurities," which are actually just the various flavors present in the wine. The goal (my statements here actually apply to Scotch Whisky, with which I'm more familiar, but should generalize to wine) is to produce something with an intricate, multifaceted flavor -- exactly the opposite of what most beverage manufacturers (Coke, Pepsi, etc.) are trying to do. These sorts of "impurities" are the sort of things that set a painting apart from a photograph, or a live musician apart from a MIDI performance.
Of course, a large amount of expensive wine and spirits (likely the majority) end up being purchased by wealthy people more interested in showing off their sophistication than actually drinking the stuff; the true connoisseurs are more likely to be the college student who trades seeing movies for a couple months for a single bottle which he finishes in a couple days or the regular middle-class guy who feels somewhat uncomfortable when he goes to buy a bottle of wine at an unnecessarily metrosexual storefront than the Paris Hiltons of the world.
If you open up the original iPod firmware in a hex editor with the proper number of columns, the first thing you see is an ASCII-Art stop sign and a scary legal message. This is nothing new.
Perhaps "havoc" isn't an entirely appropriate term. According to your link, Zotob only affects computers running unpatched versions of Windows 2000. According to most usage statistics, Windows 2000 only has about 10 - 15% of the operating system market, and that's patched and unpatched systems. Further, while I haven't checked this I'll bet that Microsoft uses Windows 2000 to run their spiders for MSN search, meaning that the actual number is probably even lower. Now, while having 7% of desktops infected by a worm is annoying, the only way that this would count as "havoc" is if you were managing a corporate network full of identical W2K boxes, and in any case that's nowhere near the sort of damage that is routinely unleashed by social-engineering bots.
First, I have a hard time believing that a professor took students from being "computer newbies" to being able to print out "hello world" ten times in thirty days, much less write some sort of working virus; trying to teach students anything outside of their major is roughly equivalent to pushing dead whales uphill in terms of efficiency. I've been in a lot of classes and taught a few, and I know that the average student will not do any work if it's at all plausible that a significant number of other students won't do it either -- school these days has become a generalized prisoner's dilemma situation, in which the teacher can only fail so many students before being reassigned.
In the larger scope, I'll just say that it's very tempting to think that one's computer programs just scale automatically, but this is simply not the case. Chances are that you were working on a very homogeneous network at that point, which most machines running rollout-synchronized versions of the same software. I've written "worms" that work under such an environment myself -- unlocking the parental protection on the middle-school computers made lunch-time in the library a lot more interesting. In such a situation, a worm either doesn't spread at all or immediately takes over the entire network, so any success is an impressive one.
On the real internet, on the other hand, we have a very complicated mesh of various systems with different sorts of protections, some explicitly designed as such but most just due to random variations that prevent a given buffer overflow from working on more than one system. Even if someone is running a vulnerable system somewhere out there, there's a good chance that getting at it may involve going past some other system that is simply going to eat it alive. We're not talking just about computers, but also about routers, switches, and all that Cisco equipment that's silently running a good deal of the net without anyone ever thinking about it.
That's why there hasn't been a real worm on the internet in quite a while; essentially every major virus in recent memory has relied on social-engineering to trick the user into manually installing the virus onto his own computer. In fact, I'd seriously doubt that it's even feasible to create a self-distributing worm on the internet at this point, unless Microsoft is dumb enough to build remote-execution capability into their application software again.
Of course, if you were actually working on a diverse, real-world type network, and you managed to devise cross-platform vectors, that's quite different and it'd be interesting to hear about. But if you're like the majority of people who make claims like these, I'm gonna have to say that your eyes are probably a little bigger than your mouth on this one.
A single incident is not an example of natural selection, any more than the collision of two particles is an example of the ideal gas law. Common sense is not genetic, and what we consider common sense isn't necessarily anti-survival. I'll bet that this kid, had he lived, would have quickly begun engaging in unprotected sex with multiple partners, which would be doing a lot more to promote his genetic material than most of us are doing. Evolution is a scientific theory just like Maxwell's equations or quantum mechanics. If you wouldn't feel comfortable talking about the latter, you likely talk about the former at your peril.
My comments were meant humorously. Of course, in an age when you can patent software and/or condemn buildings in order to sell them to private owners, I wouldn't be surprised by much of anything anymore.
What do you want to bet that once the 2.7 kernel does come out that SCO points at the IBM contributions and claims perjury? Actually, that's probably the strategy they're going with here in the first place.
A quick whois confirms that news.com and com.com are both owned by C|NET. Nothing to see here, folks.
Actually, if you adjust for inflation, it comes in a little lower. Still, $380 million is absurd; that movie didn't deserve to make more than about a buck fifty.
I never understand why people pay so much attention to articles like this. Dan Zambonini runs a small company that nobody has ever heard of which makes a content management system used by a bunch of people you've never heard of. I suppose that's a little better than Joel Spolsky, who makes software that *nobody* uses, but these guys really don't have much of a clue what kinds of jobs are out there in the companies that, you know, you may have heard of before -- they're too busy running their garage operations.
I've talked to recruiters from companies like Google, Microsoft, Symantec, etc. and while they do want someone who can actually sit down and write a program, if you are going to write software you are going to need to understand things like analysis of algorithms or else you are going to end up putting bubblesorts into production code and leaving users wondering why your search feature takes so much longer than your competitors'.
Of course, if you want to work in the database department for a large company and write software that moves tables of numbers around and is never seen by the outside world, then by all means go into MIS. But if you, like most aspiring programmers, want to release software onto the general market and you want to write something other than very basic utilities like "Windows Power Tools" or "Texteditor Extreme 5.3," you are going to have to have a foundation in computer science and a whole bunch of math.
(1) Write a book of haiku that consist of the serial numbers on each piece of paper money you own. Publish it, set up public performances, etc.
(2) Spend money
(3) Sue companies for copyright infringement
(4) Profit!
The tone of this article is refreshing, likely because it doesn't deal directly with computers. Every time an independant researcher discovers a buffer overflow exploit, he's branded a criminal by the industry and the media play along. It's time we start to demand that articles about security put the researcher and not the faceless corporation in the protagonist's role.
Yeah... did you happen to notice the little "Search this site with Google" box on the Nutch homepage? I'm a *little* wary of a search spider that doesn't even index its own homepage!
The very first comment is moderated redundant? Sounds like someone needs to buy a dictionary...
I used Altavista's raging.com for as long as it existed; when they finally took that away I reluctantly moved to this "google" thing that people had been talking about.
Now that I know what's going on, however, I'd say the pagerank algorithm really is so important that it supercedes everything that came before it. Altavista was impressive because of its spiders and the magnitude of its database, but it was often unable to make any sense of the relevancy of various pages. Google, on the other hand, can usually find exactly what you're looking for on the first result.
Nonsense.
Search: Google's Pagerank concept radically changed the way that search engines determined which results were relevant. While previous services were based on human rankings or on how many times a particular word was listed on the page, Google put out an automated system which was able to deliver more relevant results when confronted with normal sites and, by its very design, much harder to exploit with SEO techniques. Further, Google continually tweaks the parameters of their search -- if you can go to one of Norvig's talks about the sorts of stuff they do, it's amazing.
Maps: That interface -- scrolling, markers, and all -- is done entirely in javascript. No plugins, no flash, no helpers. Nobody thought that that sort of thing was even possible.
GMail: I don't use it, so I can't comment. But I do have around 1 GB of email on my primary account. When you use email for serious work, it can add up.
Google Groups: It's my group reader. I like it because it shows the discussions in thread format from the top and supresses the quoting that can make USENET discussions turn into pages and pages of greater-than symbols.
As to your assertion that Google hasn't ushered in a new age, I disagree. Ten years ago, when someone wanted information they went to a library, an encyclopedia, or maybe a CD-ROM. Now, any time anyone wants to know anything, they go immediately to Google and chances are that the information will come up on the first page.
Lest you've forgotten, it was Napster and Winamp that 0popularized mp3's, not the iPod, and COBOL, not Oracle, that popularized the database. So I'd respond to you, "Stop the misinformation campaign."
Not so. Producing hydrogen by hydrolosis requires more energy than it produces (as do all processes), so the energy put into the hydrogen would have to be supplied somewhere else -- presumably by more fossil fuel.
The only advantage that hydrogen has over fossil fuels is that fossil fuels can be burned out away from population centers to produce hydrogen; fuel-cell based cars could reduce pollution in large cities at the expense of even more pollution at some factory.
Seriously, didn't everyone see this as a demonstration in high-school chemistry? This isn't exactly that new or exciting...
I think the issue is when someone writes an article about abortion in the American legal system and someone attempts to replace it with several copies of the word "murder." Other good targets for stasis would be pages pertaining to evolutionary biology, the holocaust, and similar areas where there is a small, vocal lunatic fringe which is militant in publicizing its incorrect dogma.
Scientists and mathematicians are some of the most creative people around -- math itself is a fine art on par with music, painting, or drama; science is slightly more applied, but the art of designing experiments to give precisely the results needed surely requires more creativity than churning out a summer exploitation movie.
For just one of many, many examples, consider Henry Cavendish's "weighing the earth" experiment. A run-of-the-mill screenwriter just doesn't come up with ideas as good as torsion balances.
I'm always glad to see someone criticizing the Mozart-worship in "classical music" circles. Mozart was nowhere near the equal of his teacher Haydn nor his student Hummel, and yet we hear more about Mozart these days than those two combined. Surely he came out with a few good pieces -- the Rondo alla Turca comes to mind -- but for some reason people feel inclined to perform every single piano work the man ever came out with, most of which are just outright annoying to listen to.
We used Excel's built-in LINEST function on the same data on three different computers and got three different numbers. The function which did not work properly was not written by a third party but by Microsoft. The computers, for reference, were all PowerPC-based Macs with various versions of Mac OS 9, and there were various versions of Excel installed.
I then wrote a little macro which calculated the linear regression which gave identical, correct results on each computer. I didn't go to the trouble of sending an error report to Microsoft since I don't use Excel in my day-to-day business -- I had to use the computers in the lab that day because the alternative would have been walking home and doing the data analysis there.
To recap, our home-rolled, off-the-cuff macro worked; the Excel builtin function didn't.
I have a friend who is currently working as a quant at one of the well-known Wall Street analysts who occasionally asks me for math advice (I'm a mathematician). Recently he told me that his work invoved an Excel spreadsheet which took upwards of forty minutes to calculate. Now, perhaps this was the fault of some atrocious flaw in his code, but knowing him I doubt it. All in all I've been decidedly unimpressed with Excel. I'm sure that it's *useable* for serious work, but it certainly doesn't meet the sort of standards that professional-grade software should.
IMHO, historical notability can be equated with producing something that changed the way things were done in a positive manner. Examples would be Fred Brooks, Minsky, Kernighan, Richie, Pike, Alan Kay, etc.
The author of this article, on the other hand, has done nothing of any such noteworthiness. His main "achievement," if it can be called that, was designing and managing the implementation of a BASIC dialect in the 1990's. From his resume:
"Program Manager, Microsoft Excel. Part of the team that designs the world's best selling spreadsheet application. Drove a new macro language strategy for Microsoft (Visual Basic for Applications)..."
Now, most professionals consider Excel inadequate for any serious work, a point that was illustrated recently in a chemistry lab I was working in when three different computers running three different versions of Excel returned three different results for a linear regression. Upon writing an Excel macro which calculated linear regressions correctly, we found that all three computers were giving incorrect results. As data fitting is one of the most basic functions you could want in a spreadsheet and linear regression is the simplest form, I'm decidedly unimpressed by anyone who worked on Excel.
Unless this Joel fellow has some sort of long-forgotten historical notability in software engineering, I fail to see why his articles continually show up here. His company, which he repeatedly plugs in his articles, has put out two pieces of software -- a bugtracker and a content management system -- which nobody I've talked to has ever heard of. Does Joel have some insight into programming that everyone reading Slashdot does not?
I suppose, of course, that the same could be said for most tech journalists, most of whom would have a hard time compiling a source tarball. On the other hand, usually tech journalists are reporting on companies' press releases, not writing editorials about software design.
With the exception of Aesop's Fables and medieval morality plays, most good pieces of fiction are not generally built around a single "moral." There are themes throughout a work, but usually a serious author does not write a story for the express purpose of advancing a message. Upton Sinclair felt strongly enough about this that he prefaced The Jungle by saying that unlike actual literature, his book was written for the purpose of conveying a message.
I haven't read this book and I likely won't; it sounds too much like a the computer-crime version of a Tom Clancy or Stephen King novel. There are probably several flaws in it, but the reasons you cite would rule out e.g. Thomas Pynchon's Entropy, one of the classics of modern American literature.