I'm totally lost as to why it is necessary in the first place to arrange for non-discriminatory access. When Open Source is left out of the picture anyway, why not leave it to the market place what Intellectual Property is worth?
You really can't blame Microsoft when even you admit that it's 3rd party apps that are causing the problem.
Of course I can. There used to be a time when a virus checker only had to care about accesses to.EXE,.COM and.DLL. If you disable the "scan all file types" feature nowadays, you're vulnerable to macro attacks, and of course to the brilliant feature that allows files with the.CMD and a slew of other extensions to have an MZ magic header and be treated as a binary.
Those are design problems, that a virus checker has no speedy workaround for. It has to treat every file as hostile.
I don't want to know how many of our virus infections have a user who "optimized" his virus checker as the root cause.
Hrm. Fixing a bug in a product as complex as OpenOffice is not particularly easy -- especially if it is not a crashing bug so you don't have a starting point in the debugger. Learning your way around such a huge source tree is a major undertaking.
It took me months to find my first crashing bug in Mozilla (and that bugfix was obsolete by the time I got the patch to the developers).
The coolest thing about having the source is that when you disagree with the developers, you can Just Hack It. This doesn't buy you much if you then rely on your hacked copy (and have to maintain your hack), but it gives a much more level playing field if you want to discuss why making such a change would be a good thing, because you can show them how your proposal would behave.
In the case of MS Office, first thing I would have done years ago if I had the source is instrument the binary just to find out who is using macros and what for. I hate being told by users that they need dangerous feature X, only to learn later that they don't know how to use it if their lives depended on it.
This only goes to show how complex the matter is... Nature has shown to be very resiliant to mans feeling of having conquered disease X. As a kid, when I had a cough, the doctor prescribed antibiotics. The result of this collective overreaction has, over the years, been that (a) a number of dangerous mutations of low-risk bacteria have sprung up (hospital syndrome), and (b) that if I ever get an infection with a life threatening hostile organism, the doctors will have a hard time figuring out an antibiotic that will work.
A very good example of what I think is wrong with the mechanical (i.e., the human body is just a machine) approach is the E.coli issue. A number of people have died from eating hamburgers that weren't fully cooked, and happened to be infected with a dangerous E.coli strain. Everyone points at E.coli as the culprit, but at least some of the deaths can be attributed to the use of Imodium, a drug that stops diarrhea by basically shutting down the bowels, stopping the annoying diarrhea, but allowing E.coli to grow unchecked in the bowels. Just allowing the body to be violently ill but get rid of the disease seems to have disappeared off the list of treatment methods.
I'm not a medical scientist, and in fact I know diddly squat about medicine, but I think there is genuine cause for alarm when I hear scientists claim that cures are around the corner. The most effectives cures I've seen so far are clean drinking water and a certain level of hygiene.
Oh, and has it ever occurred to anyone that happy people seem to be healthier too?
Re:The chipset is the magic...
on
Itanium Update
·
· Score: 1
Huh, I will beat you on that one. My main squid proxy for the longest time was a 386SX with 12MB of RAM, and a really slow IDE disk scavenged from a laptop.
Only reason I upgraded was that I got tired for waiting for the prompt to appear when ssh'ing into the machine.
That's good. I'm happy to be reminded every once in a while that whatever half baked wisdoms I spout, they are usually based on a corporate image of what makes economical sense and what doesn't.
The art of computing wasn't furthered that much by the corporates, I know!
In its heyday, the 6502 was an eight bit RISC processor avant la lettre. It featured a whopping 256 memory locations that could be accessed with near-zero overhead. The famous page zero.
Needless to say, this great concept had gone to the dogs before the first consumer laid his/her hands on the device. Oblivious to the CPU design, a major manufacturer of operating systems (we called them BASIC interpreters at the time, by the way) has decided that most of page zero should be allocated to the OS^WBasic interpreter. I'll leave it to our hidden conscience to name the prepretrator of this gruelsome mistake.
I have long grown over the idea of using assembly as a faster programming language. The number of times I beat an assembly program with something hacked up in Perl, I don't even want to remember. Not because Perl is the best thing since sliced bread, but because humans are so poor at dealing with complexity. Get it working first, and leave optimization to the compiler. Then, if you have a bottleneck, analyze it, and fix the bottleneck in a targeted piece of code (whether C, or assembly, or something else).
Alpha and Sparc architecture had this (almost) from day 1.
True, but probably unimportant. OpenFirmware has been around (I think even as an IEEE standard) for ages, but apparently the PC world doesn't care.
If I sound frustrated, it's because I am. OpenFirmware is such a small bone to throw the techies that I think it's criminal it never came about in the PC world. After years of haggling the BIOS vendors, we now have BIOSes on some machines that can optionally emulate an ANSI terminal for console access on a serial port. Which means you can use Kermit under DOS to manage the machine remotely. Even "tip" on UNIX is suboptimal here, and the best this BIOS will do is emulate the full-screen config menu. Beeeuuurk.
The chipset is the magic...
on
Itanium Update
·
· Score: 2
I've been extremely reluctant in going the AMD route. My first AMD processor was a 133MHz 486, which was branded in a way as to resemble the Pentium 75 (on the premise that it was as fast). I put it into my firewall, which was not getting heavy duty at the time.
The thing sucked eggs, and I threw the motherboard in the trash and used the CPU as a paperweight.
At some stage, I needed a faster CPU, needed a motherboard to go with it, so I made the jump to a PentiumIII/450. I needed to revive my firewall, so I bought a decent ASUS 486 mobo at a fair. On a hunch, I put my paperweight AMD133 in. I was pleasantly surprised, and I only replaced the thing when I got a real cheap 300MHz Cyrix mobo.
Bottom line, it't the motherboard (or rather the chipset on it) that makes or breaks the CPU. I'm now running an ASUS A7V-E with a 1GHz Athlon, and I've been a happy camper. I'm not an overclocker (matter of fact, I underclock some machines just because I don't need CPU power for other things than video recoding, and some machines are on the other end of the globe, so I don't want to lose sleep over fan failure).
My main gripe with the VIA KTA133 chipset is the fact that I have to sign a $#@#%$#% NDA to get decent specs on it. FreeBSD doesn't seem to grok its I2C based hardware monitoring, and without those docs I'm SOL. Apart from that, it's working great. Even under Carmageddon^WWindows.
Sourceforge is, indeed, full of dead projects. But please note that many of those dead projects either contain no code at all, or code of the
int
rescue_the_world()
{
// TODO: figure out how to do this
}
variety. Code that does *nothing* at all.
It is at the stage where code does at least one task well that the wheat is seperated from the chaff (except that one means wheat is another mans chaff). Before I pour considerable time into something, I test the waters: I do some small enhancement and report it to the authors. If they don't acknowledge that, I consider the thing to be unmaintained. Rejection of the code is not of a great concern, it's being ignored that hurts.
I've had pretty good luck with a number of Open Source projects which involve no paid staff at all. Especially when compared with the support for some closed source offerings.
The commercial open source thing is still very new. People jump up and down about Mozilla's failure, about OpenOffice's failure, etcetera. Let's face it, those are huge projects, and the success rate for their commercial counterparts is equally abysmal. The big hindrance for community contributions is the extremely steep learning curve for the infrastructure surrounding huge projects, and combined with rapid change that curve becomes in unsurmountable mountain. Once a more or less successful 1.0 is out, that ought to change.
Unless I'm seriously mistaken, no huge commercial open source project has reached 1.0. Unless you count Linux, but that sort of underscores the point that contributions on any level have always been possible building on a stable base. You can rewrite, say, the IP stack or the VM system, but you can do that in relative isolation, despite the complexity of the task. API's are pretty well defined.
Most of the grand commercial open source projects are just too big, with too many internal couplings. Writing a spreadsheet app is one thing (and plenty of those abound without commercial support). But keeping the thing alive when someone else is dicking with text editor code that your subproject needs is another. The horrible thing is, they need to be huge, because they need to appeal to the mass market.
If I had my way, I'd had a simple mail client, a simple web browser, a simple spreadsheet, with the minimal glue between them to make it work for me, and allow me to replace a component I don't like with another.
So, I guess you're right. Not because great things don't happen without people who are paid for writing free code, but because the marketplace demands solutions that are just too complex to deal with in part time.
I still have an original MacPlus at home. I still feel it is So Cool to have all those signatures cast inside the casing that I just couldn't get myself to ditching the thing.
It's been years since I opened the thing, but I remember going through them with friends and trying to make out the individual signatures.
Really, anyone who has not seen Waynes World should not be taking part in this discussions.
Re:Its interesting that Internal...
on
Breaking Windows
·
· Score: 2
This question is a very good one, in that it is unanswerable. Most people that leave MS do so after their stock options vested, and by that time they may not have lost their wild hairs, but at least adjusted their views to include the 1, Microsoft Way, as a possibly correct one (if with modifications).
One thing that struck me about successful start-ups from the post-MS area is that they invariably hinge on some as of yet unharvested opportunity left by MS, as opposed to the blank-slate approach that herald most innovative enterprises.
Raise of hands here, folks. Outside of Sun's move with OpenOffice, and Netscape's attempts at opening the browser, which new enterprises dared to challenge the MS monopoly^Walleged monopoly?
Ewww, screw that. PPPoE is the worst of two worlds:
Anyone who hoards an address is free to do so, and software to renew leases before they expire is widely available (including software that reflects a random new IP address to DNS services, in the unlikely event the provider implemented address churn).
On the other hand, it does nothing to promote a societally clean usage of the privilege of being able to spew IP packets. My provider gave me a static IP address, and I'll be damned before I get kicked off the service for AUP violations. Try to instill that kind of awareness to end users if they don't have anything to lose.
I've defended our local incumbent operator just once too many... I don't believe in such coincedences any more.
We have real providers doing this, visors up. "Bring us thirty likely subscribers in your ZIPcode area, and we'll connect them". That's not only more friendly, it's also more encouraging for people in an unserved area to work their neighbors, and don't tell me that advertising costs aren't visible to the big guys!
Speaking from the bowels of corporate hell, I can assure anyone that the bandwidth issues are as to nothing compared to the manpower invested.
I've applied the C2 security fixes to out IIS server (they're secret, don't ask me about details or I'd have to bury you). But still, the bleeding thing kept attacking our Apache and Netscrape servers, and you don't want to know the pain and suffering of explaining the risks to the end users...
Re:Frequent Flyer's Principle
on
Books on Demand
·
· Score: 3
No, no, you got it the wrong way around. Airline pax will pay any amount of money to fulfill their real or imaginary needs, so margins can be higher.
Try to buy a roll of tape to hold your suitcase shut. Then buy a cup of coffee. Add the experience of the eternal smile and good humor of the check-in handlers, and you'll be softened up to the point that you will need to buy a nicely bound copy of, say, Taiwan's Agricultural Statistics 1980-1989 and go for the leather binding option!
The OpenDWG effort is laudable, but last I checked, the public won't get source to the library. Apart from the library not being available for the platform I use, it's not very sustainable: what if they fold? What if you upgrade and the libraries are no longer compatible with your new OS?
I am writing this from a 100% pixel-perfect laptop right now
I am writing this from what used to be a 100% pixel perfect desktop TFT screen, until they day I dropped my pen and the back of it grazed the screen. On a black background, I now have some red and green speckles. Glass tubes are more resilient.
I was in line in a computer shop the other day, and heard the sales clerk ask if the prospective buyer had kids. He had, and left with a glass screen.
It's much the same as with television: the Web is next to useless these days without color. Seeing that almost everyone want to run Web stuff, the market is really small for it.
For the longest time, I has a 1280x1024 black and white screen on my desk. It is just easier on the eyes, and it's perfect for coding and e-mail. I eventually replaced it with a color screen because of all the web sites I had to use that abuse color.
What is an interesting market once someone uncovers it, is cheap 640x480 B&W LCD's to use as a console in colo facilities. I hate wasting 6U os rack space for a monitor, and I'm not the only one
Whole bunch of info on the Web site. 96kbps. It's not for the faint of heart to build and won't be speedy, but it's a nice grassroots thing...
If it weren't June, I'd take it for an Aprils Fool, but at a cursory glance it does look like the guy actually built it. It would be cool in areas where copper is available, but ADSL gear is out because of the distances.
Re:While we're at it: car mountable, anyone?
on
Tiny Little Computer
·
· Score: 1
Well, obviously the makers of car mountable GPS units solved that issue.
We're talking about something like 20 watts. I'd hazard that there are manufacturers that sell complete DC-DC convertors for precisely this purpose, without burning the excess 12-5=7 volts in heat like a 7805 would.
But then again, if my electronics skills were any better I'd have built that bit myself:-)
The cappucino site specifically mentions Linux as a supported OS.
That said, the 810 is Intel's chipset for El Cheapo motherboards, and it shows in a number of ways. I eventually got XFree to work on an 810 based mobo, but I never got to like the board.
Then again, Intel is no worse than VIA here. I'm still awaiting their response to my query for the data sheets (you can download the table of contents, but VIA has some mumbo jumbo about NDA's and just don't respond to enquiries). Maybe if I learned Taiwanese:-)
I've been looking for a board or kit that would allow me to do kewl things with a GPS in my car. It doesn't require much CPU power, but requirements include
Single 12V power supply
Two RS232/422 ports
Sound
Bidir parallel port
Compact flash
The whole idea is to have two or three dash mounted buttons to mark the current location for review, enough RAM to store the raw location data, a serial hookup to up/download location and config data, and a hookup to the cellphone input of my car radio, to gently remind me when I enter a reduced speed highway and not so gently remind me when I come close to a known speed trap.
It shouldn't be rocket science, but I've yet found a board that takes just a single 12V. Of course, I could just mount a laptop, but the hassle of unhooking the thing all the time to avoid the car being ripped open doesn't sound too exciting to me...
It is entirely unclear what Darren is trying to protect against.
I'll buy his stated goal of not wanting to deal with patches that do no apply cleanly (and anyone who has dealt with multiple OS kernel code can attest to the royal pain that re-indenting and like changes are). However, wishing to codify this is guaranteed to rub people the wrong way (and of course, rubbing Theo the wrong way is a surefire way of starting a war).
The whole thing smacks of all parties doing "what-if"s, and Darren falling prey to Fear, Uncertainty and Doubt. Not good.
You can only prove insecurity. That's a basic tenet of the security as well as the cryptography community.
That said, there is some difference between securing a development server where a gazillion people have shells and need to be able to run arbitrary code, than it is to secure something that, say, only needs to spit out HTML pages.
Slashdot Mirror
I'm totally lost as to why it is necessary in the first place to arrange for non-discriminatory access. When Open Source is left out of the picture anyway, why not leave it to the market place what Intellectual Property is worth?
Of course I can. There used to be a time when a virus checker only had to care about accesses to .EXE, .COM and .DLL. If you disable the "scan all file types" feature nowadays, you're vulnerable to macro attacks, and of course to the brilliant feature that allows files with the .CMD and a slew of other extensions to have an MZ magic header and be treated as a binary.
Those are design problems, that a virus checker has no speedy workaround for. It has to treat every file as hostile.
I don't want to know how many of our virus infections have a user who "optimized" his virus checker as the root cause.
It took me months to find my first crashing bug in Mozilla (and that bugfix was obsolete by the time I got the patch to the developers).
The coolest thing about having the source is that when you disagree with the developers, you can Just Hack It. This doesn't buy you much if you then rely on your hacked copy (and have to maintain your hack), but it gives a much more level playing field if you want to discuss why making such a change would be a good thing, because you can show them how your proposal would behave.
In the case of MS Office, first thing I would have done years ago if I had the source is instrument the binary just to find out who is using macros and what for. I hate being told by users that they need dangerous feature X, only to learn later that they don't know how to use it if their lives depended on it.
A very good example of what I think is wrong with the mechanical (i.e., the human body is just a machine) approach is the E.coli issue. A number of people have died from eating hamburgers that weren't fully cooked, and happened to be infected with a dangerous E.coli strain. Everyone points at E.coli as the culprit, but at least some of the deaths can be attributed to the use of Imodium, a drug that stops diarrhea by basically shutting down the bowels, stopping the annoying diarrhea, but allowing E.coli to grow unchecked in the bowels. Just allowing the body to be violently ill but get rid of the disease seems to have disappeared off the list of treatment methods.
I'm not a medical scientist, and in fact I know diddly squat about medicine, but I think there is genuine cause for alarm when I hear scientists claim that cures are around the corner. The most effectives cures I've seen so far are clean drinking water and a certain level of hygiene.
Oh, and has it ever occurred to anyone that happy people seem to be healthier too?
Only reason I upgraded was that I got tired for waiting for the prompt to appear when ssh'ing into the machine.
That's good. I'm happy to be reminded every once in a while that whatever half baked wisdoms I spout, they are usually based on a corporate image of what makes economical sense and what doesn't.
The art of computing wasn't furthered that much by the corporates, I know!
Needless to say, this great concept had gone to the dogs before the first consumer laid his/her hands on the device. Oblivious to the CPU design, a major manufacturer of operating systems (we called them BASIC interpreters at the time, by the way) has decided that most of page zero should be allocated to the OS^WBasic interpreter. I'll leave it to our hidden conscience to name the prepretrator of this gruelsome mistake.
I have long grown over the idea of using assembly as a faster programming language. The number of times I beat an assembly program with something hacked up in Perl, I don't even want to remember. Not because Perl is the best thing since sliced bread, but because humans are so poor at dealing with complexity. Get it working first, and leave optimization to the compiler. Then, if you have a bottleneck, analyze it, and fix the bottleneck in a targeted piece of code (whether C, or assembly, or something else).
True, but probably unimportant. OpenFirmware has been around (I think even as an IEEE standard) for ages, but apparently the PC world doesn't care.
If I sound frustrated, it's because I am. OpenFirmware is such a small bone to throw the techies that I think it's criminal it never came about in the PC world. After years of haggling the BIOS vendors, we now have BIOSes on some machines that can optionally emulate an ANSI terminal for console access on a serial port. Which means you can use Kermit under DOS to manage the machine remotely. Even "tip" on UNIX is suboptimal here, and the best this BIOS will do is emulate the full-screen config menu. Beeeuuurk.
The thing sucked eggs, and I threw the motherboard in the trash and used the CPU as a paperweight.
At some stage, I needed a faster CPU, needed a motherboard to go with it, so I made the jump to a PentiumIII/450. I needed to revive my firewall, so I bought a decent ASUS 486 mobo at a fair. On a hunch, I put my paperweight AMD133 in. I was pleasantly surprised, and I only replaced the thing when I got a real cheap 300MHz Cyrix mobo.
Bottom line, it't the motherboard (or rather the chipset on it) that makes or breaks the CPU. I'm now running an ASUS A7V-E with a 1GHz Athlon, and I've been a happy camper. I'm not an overclocker (matter of fact, I underclock some machines just because I don't need CPU power for other things than video recoding, and some machines are on the other end of the globe, so I don't want to lose sleep over fan failure).
My main gripe with the VIA KTA133 chipset is the fact that I have to sign a $#@#%$#% NDA to get decent specs on it. FreeBSD doesn't seem to grok its I2C based hardware monitoring, and without those docs I'm SOL. Apart from that, it's working great. Even under Carmageddon^WWindows.
int
rescue_the_world()
{
}
variety. Code that does *nothing* at all.
It is at the stage where code does at least one task well that the wheat is seperated from the chaff (except that one means wheat is another mans chaff). Before I pour considerable time into something, I test the waters: I do some small enhancement and report it to the authors. If they don't acknowledge that, I consider the thing to be unmaintained. Rejection of the code is not of a great concern, it's being ignored that hurts.
I've had pretty good luck with a number of Open Source projects which involve no paid staff at all. Especially when compared with the support for some closed source offerings.
The commercial open source thing is still very new. People jump up and down about Mozilla's failure, about OpenOffice's failure, etcetera. Let's face it, those are huge projects, and the success rate for their commercial counterparts is equally abysmal. The big hindrance for community contributions is the extremely steep learning curve for the infrastructure surrounding huge projects, and combined with rapid change that curve becomes in unsurmountable mountain. Once a more or less successful 1.0 is out, that ought to change.
Unless I'm seriously mistaken, no huge commercial open source project has reached 1.0. Unless you count Linux, but that sort of underscores the point that contributions on any level have always been possible building on a stable base. You can rewrite, say, the IP stack or the VM system, but you can do that in relative isolation, despite the complexity of the task. API's are pretty well defined.
Most of the grand commercial open source projects are just too big, with too many internal couplings. Writing a spreadsheet app is one thing (and plenty of those abound without commercial support). But keeping the thing alive when someone else is dicking with text editor code that your subproject needs is another. The horrible thing is, they need to be huge, because they need to appeal to the mass market.
If I had my way, I'd had a simple mail client, a simple web browser, a simple spreadsheet, with the minimal glue between them to make it work for me, and allow me to replace a component I don't like with another.
So, I guess you're right. Not because great things don't happen without people who are paid for writing free code, but because the marketplace demands solutions that are just too complex to deal with in part time.
It's been years since I opened the thing, but I remember going through them with friends and trying to make out the individual signatures.
Really, anyone who has not seen Waynes World should not be taking part in this discussions.
One thing that struck me about successful start-ups from the post-MS area is that they invariably hinge on some as of yet unharvested opportunity left by MS, as opposed to the blank-slate approach that herald most innovative enterprises.
Raise of hands here, folks. Outside of Sun's move with OpenOffice, and Netscape's attempts at opening the browser, which new enterprises dared to challenge the MS monopoly^Walleged monopoly?
Anyone who hoards an address is free to do so, and software to renew leases before they expire is widely available (including software that reflects a random new IP address to DNS services, in the unlikely event the provider implemented address churn).
On the other hand, it does nothing to promote a societally clean usage of the privilege of being able to spew IP packets. My provider gave me a static IP address, and I'll be damned before I get kicked off the service for AUP violations. Try to instill that kind of awareness to end users if they don't have anything to lose.
We have real providers doing this, visors up. "Bring us thirty likely subscribers in your ZIPcode area, and we'll connect them". That's not only more friendly, it's also more encouraging for people in an unserved area to work their neighbors, and don't tell me that advertising costs aren't visible to the big guys!
Speaking from the bowels of corporate hell, I can assure anyone that the bandwidth issues are as to nothing compared to the manpower invested.
I've applied the C2 security fixes to out IIS server (they're secret, don't ask me about details or I'd have to bury you). But still, the bleeding thing kept attacking our Apache and Netscrape servers, and you don't want to know the pain and suffering of explaining the risks to the end users...
Try to buy a roll of tape to hold your suitcase shut. Then buy a cup of coffee. Add the experience of the eternal smile and good humor of the check-in handlers, and you'll be softened up to the point that you will need to buy a nicely bound copy of, say, Taiwan's Agricultural Statistics 1980-1989 and go for the leather binding option!
The OpenDWG effort is laudable, but last I checked, the public won't get source to the library. Apart from the library not being available for the platform I use, it's not very sustainable: what if they fold? What if you upgrade and the libraries are no longer compatible with your new OS?
I am writing this from what used to be a 100% pixel perfect desktop TFT screen, until they day I dropped my pen and the back of it grazed the screen. On a black background, I now have some red and green speckles. Glass tubes are more resilient.
I was in line in a computer shop the other day, and heard the sales clerk ask if the prospective buyer had kids. He had, and left with a glass screen.
For the longest time, I has a 1280x1024 black and white screen on my desk. It is just easier on the eyes, and it's perfect for coding and e-mail. I eventually replaced it with a color screen because of all the web sites I had to use that abuse color.
What is an interesting market once someone uncovers it, is cheap 640x480 B&W LCD's to use as a console in colo facilities. I hate wasting 6U os rack space for a monitor, and I'm not the only one
If it weren't June, I'd take it for an Aprils Fool, but at a cursory glance it does look like the guy actually built it. It would be cool in areas where copper is available, but ADSL gear is out because of the distances.
We're talking about something like 20 watts. I'd hazard that there are manufacturers that sell complete DC-DC convertors for precisely this purpose, without burning the excess 12-5=7 volts in heat like a 7805 would.
But then again, if my electronics skills were any better I'd have built that bit myself :-)
That said, the 810 is Intel's chipset for El Cheapo motherboards, and it shows in a number of ways. I eventually got XFree to work on an 810 based mobo, but I never got to like the board.
Then again, Intel is no worse than VIA here. I'm still awaiting their response to my query for the data sheets (you can download the table of contents, but VIA has some mumbo jumbo about NDA's and just don't respond to enquiries). Maybe if I learned Taiwanese :-)
The whole idea is to have two or three dash mounted buttons to mark the current location for review, enough RAM to store the raw location data, a serial hookup to up/download location and config data, and a hookup to the cellphone input of my car radio, to gently remind me when I enter a reduced speed highway and not so gently remind me when I come close to a known speed trap.
It shouldn't be rocket science, but I've yet found a board that takes just a single 12V. Of course, I could just mount a laptop, but the hassle of unhooking the thing all the time to avoid the car being ripped open doesn't sound too exciting to me...
I'll buy his stated goal of not wanting to deal with patches that do no apply cleanly (and anyone who has dealt with multiple OS kernel code can attest to the royal pain that re-indenting and like changes are). However, wishing to codify this is guaranteed to rub people the wrong way (and of course, rubbing Theo the wrong way is a surefire way of starting a war).
The whole thing smacks of all parties doing "what-if"s, and Darren falling prey to Fear, Uncertainty and Doubt. Not good.
That said, there is some difference between securing a development server where a gazillion people have shells and need to be able to run arbitrary code, than it is to secure something that, say, only needs to spit out HTML pages.