Tweet
*sigh*. It would take a very long time to write a thorough, proper response to some of the above, but I'll try a cursory response once, then give up.
What I was hoping for was more intelligent comments, criticism, debate, and opinions on a very serious issue (in re: an environment where I've had a lot of professional experience), and there was some from Dghost, The PIT and Marshmellowman. I didn't draw in one frequent poster to these forums who I thought might have an opinion worth reading. Debate does not mean don't counter what someone says. That isn't "just finding others who agree." (Don't know where that one came from.) Debate means discuss merits. You should note that some didn't agree with everything I said, just like I didn't agree with them, and some brought up new, valid points. But none of those comments needed insults. Name calling, sophomoric attempts at derogatory characterizations, or sarcasms hardly constitute intelligent debate or professional criticism, nor are they "not" attempts "to start a flame war," nor are they very mature.
I may have misjudged these forums. They may be like so many others on the web. I generally don't participate because of exactly this kind of behavior. I quit posting on mensa's forum for the same reason. Computer people seem to be some of the worst offenders. Maybe I just blasphemed someone's technoreligious beliefs. With computer people (or wannabes), it always seems to be a dance of who can piss higher on the wall. I'm tired of that crap.
Yes it's true I don't like Microsoft. So what? I dislike and distrust very large corporations in general. History has shown they frequently abuse their power (goes back to the trusts of the late 19th/early 20th century; read some history). Once corporations become large enough, they almost naturally abuse their markets, buyers, suppliers, employees, and produce inferior, expensive products. Pick up almost any economics book. Specifics in re: M$ behavior would be another topic more approrpiate for The Lounge. Suffice it to say I would not apply for work at Microsoft (nor Apple, nor IBM). I disagree with MS's business and "technical" philosophies. I prefer building blocks, not a batholyth.
Yes, it's also true I frequently write word parodies, but they have some ring of truth, or they are iconoclasms to bring the high and mighty back down to earth. Micro$oft is a very rich, engulf and devour corporation making HUGE economic profits (a technical econ term with a specific meaning). Microsoft Windoze is extremely slow and crash prone on what Micro$ucks claims is the minimum necessary h/w. I haven't had that problem with some other systems. (Realistically, M$ minimums should be doubled or quadrupled as a more proper minimum. This is microsoft marketing's fault, but they mislead (I believe deliberately) because of the lack of judgement and candor. They should be more honest.) M$ isn't my only target. I found it amusing when some friends who worked for Hughes Aircraft had some T-shirts made up that said "Huge Air Crash." Hughes management got very upset about that. There are a few names for IBM. Tried calling IBM to get warrantly repacements on PC h/w? I probably should have been more careful to leave out the iconoclastic word paradies, but they are second nature. A careful reading of what I wrote should show I did no prosletyzing for any particular technoreligious alternative. I don't know why criticizing microsoft warrants a fatwah. You'd think I'd written a techno Satanic Verses in the middle of a haj to Redmond.
My point was that Windows for Warships is not a good idea. Windows is commercial s/w with a hacker origin (that's an applicable chracterization if you know the history) developed for a completely different purpose. Windows has become so large (14M-40M lines of code) that it is extremely difficult to maintain or upgrade for a variety of technoreasons. This, and the references to 14M lines of code, kludges, failures, long test/retest cycles, and bureaucratic paperwork were based on history -- WWMCCS, Star Wars, Ada, 1750A, and a report that 80% of engineering time in aerospace and the military industrial complex is now paperwork, unlike in the days of the moon race. It was all relevant to what I meant.
Since no one posting here yet has probably heard of all of these, WWMCCS was a computerized World Wide Military Command and Control System. In the '60s and 70's it was brought crashing to its knees during simulations of the kinds of processing needed during wartime. It took 25 years to get it mostly right, then it was retired. I think Windows for Warships would suffer similarly. I was too young to work on WMMCCS, but I had collegues who did.
When I worked in aerospace and the military-industrial complex, I spent a lot of my time dealing with bureuacracies, generating and shuffling paper. I was too young to work on Apollo, but some of my teachers had. Many left NASA and aerospace because they could no longer do engineering as they did on the Apollo program. They gradually had come to be required to generate more and more paper.
Star Wars was Reagan's Strategic Defense initative. I worked on that. One part of it was an estimated 50,000,000 lines of code to run the whole thing. That was one of its downfalls. That 50,000,000 lines of code couldn't be tested as a complete system in anything but a war situation, with thousands of inputs and controls, all of which had to be accepted and processed in real time, and the whole system had to work flawlessly (leaking 1 nuke is not a good thing; calling it human nature doesn't work). It never got off the ground. Integrating a Windows for Warships is a similar monumental task, starting with millions of lines of code written for a completely different environment and purpose. That makes worrying about its size, complexity and comprehendability a valid concern.
1750A was a paper architecture design by committee of a microprocessor for embedded military applications. I worked with that, too. It was a terrible kludge. It looked like a cross between a PDP-11/70 and an IBM 360/370, both way obsolete at that time. It was made a worse kludge by trying to wedge Ada to run on it. 1750A was implemented as several different types of incompatible h/w. I think Windows, originally written for a completely different purpose than military use, would be a similar kludge. That's my professional opinion. Others might have a different professional opinion.
Ada was supposed to be a do-all, be-all of embedded systems programming. I worked with that, too. It eventually turned out that it was OK for large s/w projects on large systems with lots of memory and CPU power, but was terrible for the embedded systems that it was originally intended to benefit. I think Windows is coming to be thought of as the do-all, be-all to everything s/w. T'an't no such thing. That's why both got so many kitchen sinks.
It's laughable to call alternatives to Windows, or criticism of Windows, "counter-culture" as if it was derogatory. Microsoft was the counter-culture s/w of 1980's. It was "hacker" culture. Attempts to do systematic, professional s/w development at M$ didn't happen for many years. Early windows versions were terrible. W9x inherited all of this legacy, and some 9x elements made it into the NT line. The only reason BASIC is still around is Gates can't let it go (he wrote a BASIC interpreter product for an ancient Altair long ago). BASIC was a terrible language for computer science and s/w engineering from the beginning.
My windows TCP/IP comment actually came from a friend when we were trying to communicate over the Internet, and I'd periodically get knocked off line due to a minor tweak by the phone company at ~1:30 AM every night, while he didn't. He said the Windows TCP/IP stack implementation was not very robust. It was a comment about how windows wasn't as robust on a network connection as, in that case, Linux, which he was using. My friend graduated BSEE/CE (early, too) from one of the top 10 engineering schools in the world (it has "technology" and "institute" in its name), worked for 3 years at the major backbone internet equipment developer in the world, worked for several start-ups involved in product development for WANs, and has spent his whole professional career in network and telecommunications development. I'll trust his judgement until someone with equal talent convinces me otherwise. (Oh yeah. I have a BSCS, too. So what.)
I hardly think his background or mine qualifies as "Wal-Mart" background, nor was there ever any foundation to make such a leap. All of my opinions on suitablity were legitmate professional concerns based on professional experience working on similar systems. Does the above give a proper hint?
I never said TCP/IP would be used in a military environment. I thought it was obvious that it would be some special mil-spec protocol. My intent was to say if TCP/IP under Windows has robustness issues, was there some same underlying element that would be involved with some mil-spec protocol? I don't know, but it's a valid concern. I didn't necessarily mean break into it. I did mean render it unreliable. As to cryptographic security, I can say nothing, other than to note some here can't say anything either, but for different reasons.
(cont...)
What I was hoping for was more intelligent comments, criticism, debate, and opinions on a very serious issue (in re: an environment where I've had a lot of professional experience), and there was some from Dghost, The PIT and Marshmellowman. I didn't draw in one frequent poster to these forums who I thought might have an opinion worth reading. Debate does not mean don't counter what someone says. That isn't "just finding others who agree." (Don't know where that one came from.) Debate means discuss merits. You should note that some didn't agree with everything I said, just like I didn't agree with them, and some brought up new, valid points. But none of those comments needed insults. Name calling, sophomoric attempts at derogatory characterizations, or sarcasms hardly constitute intelligent debate or professional criticism, nor are they "not" attempts "to start a flame war," nor are they very mature.
I may have misjudged these forums. They may be like so many others on the web. I generally don't participate because of exactly this kind of behavior. I quit posting on mensa's forum for the same reason. Computer people seem to be some of the worst offenders. Maybe I just blasphemed someone's technoreligious beliefs. With computer people (or wannabes), it always seems to be a dance of who can piss higher on the wall. I'm tired of that crap.
Yes it's true I don't like Microsoft. So what? I dislike and distrust very large corporations in general. History has shown they frequently abuse their power (goes back to the trusts of the late 19th/early 20th century; read some history). Once corporations become large enough, they almost naturally abuse their markets, buyers, suppliers, employees, and produce inferior, expensive products. Pick up almost any economics book. Specifics in re: M$ behavior would be another topic more approrpiate for The Lounge. Suffice it to say I would not apply for work at Microsoft (nor Apple, nor IBM). I disagree with MS's business and "technical" philosophies. I prefer building blocks, not a batholyth.
Yes, it's also true I frequently write word parodies, but they have some ring of truth, or they are iconoclasms to bring the high and mighty back down to earth. Micro$oft is a very rich, engulf and devour corporation making HUGE economic profits (a technical econ term with a specific meaning). Microsoft Windoze is extremely slow and crash prone on what Micro$ucks claims is the minimum necessary h/w. I haven't had that problem with some other systems. (Realistically, M$ minimums should be doubled or quadrupled as a more proper minimum. This is microsoft marketing's fault, but they mislead (I believe deliberately) because of the lack of judgement and candor. They should be more honest.) M$ isn't my only target. I found it amusing when some friends who worked for Hughes Aircraft had some T-shirts made up that said "Huge Air Crash." Hughes management got very upset about that. There are a few names for IBM. Tried calling IBM to get warrantly repacements on PC h/w? I probably should have been more careful to leave out the iconoclastic word paradies, but they are second nature. A careful reading of what I wrote should show I did no prosletyzing for any particular technoreligious alternative. I don't know why criticizing microsoft warrants a fatwah. You'd think I'd written a techno Satanic Verses in the middle of a haj to Redmond.
My point was that Windows for Warships is not a good idea. Windows is commercial s/w with a hacker origin (that's an applicable chracterization if you know the history) developed for a completely different purpose. Windows has become so large (14M-40M lines of code) that it is extremely difficult to maintain or upgrade for a variety of technoreasons. This, and the references to 14M lines of code, kludges, failures, long test/retest cycles, and bureaucratic paperwork were based on history -- WWMCCS, Star Wars, Ada, 1750A, and a report that 80% of engineering time in aerospace and the military industrial complex is now paperwork, unlike in the days of the moon race. It was all relevant to what I meant.
Since no one posting here yet has probably heard of all of these, WWMCCS was a computerized World Wide Military Command and Control System. In the '60s and 70's it was brought crashing to its knees during simulations of the kinds of processing needed during wartime. It took 25 years to get it mostly right, then it was retired. I think Windows for Warships would suffer similarly. I was too young to work on WMMCCS, but I had collegues who did.
When I worked in aerospace and the military-industrial complex, I spent a lot of my time dealing with bureuacracies, generating and shuffling paper. I was too young to work on Apollo, but some of my teachers had. Many left NASA and aerospace because they could no longer do engineering as they did on the Apollo program. They gradually had come to be required to generate more and more paper.
Star Wars was Reagan's Strategic Defense initative. I worked on that. One part of it was an estimated 50,000,000 lines of code to run the whole thing. That was one of its downfalls. That 50,000,000 lines of code couldn't be tested as a complete system in anything but a war situation, with thousands of inputs and controls, all of which had to be accepted and processed in real time, and the whole system had to work flawlessly (leaking 1 nuke is not a good thing; calling it human nature doesn't work). It never got off the ground. Integrating a Windows for Warships is a similar monumental task, starting with millions of lines of code written for a completely different environment and purpose. That makes worrying about its size, complexity and comprehendability a valid concern.
1750A was a paper architecture design by committee of a microprocessor for embedded military applications. I worked with that, too. It was a terrible kludge. It looked like a cross between a PDP-11/70 and an IBM 360/370, both way obsolete at that time. It was made a worse kludge by trying to wedge Ada to run on it. 1750A was implemented as several different types of incompatible h/w. I think Windows, originally written for a completely different purpose than military use, would be a similar kludge. That's my professional opinion. Others might have a different professional opinion.
Ada was supposed to be a do-all, be-all of embedded systems programming. I worked with that, too. It eventually turned out that it was OK for large s/w projects on large systems with lots of memory and CPU power, but was terrible for the embedded systems that it was originally intended to benefit. I think Windows is coming to be thought of as the do-all, be-all to everything s/w. T'an't no such thing. That's why both got so many kitchen sinks.
It's laughable to call alternatives to Windows, or criticism of Windows, "counter-culture" as if it was derogatory. Microsoft was the counter-culture s/w of 1980's. It was "hacker" culture. Attempts to do systematic, professional s/w development at M$ didn't happen for many years. Early windows versions were terrible. W9x inherited all of this legacy, and some 9x elements made it into the NT line. The only reason BASIC is still around is Gates can't let it go (he wrote a BASIC interpreter product for an ancient Altair long ago). BASIC was a terrible language for computer science and s/w engineering from the beginning.
My windows TCP/IP comment actually came from a friend when we were trying to communicate over the Internet, and I'd periodically get knocked off line due to a minor tweak by the phone company at ~1:30 AM every night, while he didn't. He said the Windows TCP/IP stack implementation was not very robust. It was a comment about how windows wasn't as robust on a network connection as, in that case, Linux, which he was using. My friend graduated BSEE/CE (early, too) from one of the top 10 engineering schools in the world (it has "technology" and "institute" in its name), worked for 3 years at the major backbone internet equipment developer in the world, worked for several start-ups involved in product development for WANs, and has spent his whole professional career in network and telecommunications development. I'll trust his judgement until someone with equal talent convinces me otherwise. (Oh yeah. I have a BSCS, too. So what.)
I hardly think his background or mine qualifies as "Wal-Mart" background, nor was there ever any foundation to make such a leap. All of my opinions on suitablity were legitmate professional concerns based on professional experience working on similar systems. Does the above give a proper hint?
I never said TCP/IP would be used in a military environment. I thought it was obvious that it would be some special mil-spec protocol. My intent was to say if TCP/IP under Windows has robustness issues, was there some same underlying element that would be involved with some mil-spec protocol? I don't know, but it's a valid concern. I didn't necessarily mean break into it. I did mean render it unreliable. As to cryptographic security, I can say nothing, other than to note some here can't say anything either, but for different reasons.
(cont...)
Comment