Re:When did the Communists take over outer space?
on
Lawyers In Space...
·
· Score: 1
The terms I used to initiate the search (Mayflower Compact, Plymouth Colony, William Bradford) would have been easy for most educated USians, but the info on your homepage seems to suggest you are Canadian. My assumption, my bad.
The Mayflower Compact is probably more promininent in my mind than most. One of my middle school teachers was a descendant of John Howland, one of the signers.
Well, to be charitable, that $10 billion may include insurance costs against business failure and the $11 billion may represent value in futures contracts. And if retreiving the asteroid takes 3 months on average, that's 40% annualized, risk-hedged return.
If they had better weapons, and a method for immigrants to purchase land,
You don't have to be hypothetical. Just look at the contrast between European colonies in the far east to those in the Americas.
Re:When did the Communists take over outer space?
on
Lawyers In Space...
·
· Score: 1
John Stossel did a documentary of it, for one. Or read the
Mayflower Compact. Good ol' Rush Limbaugh relates the same story. The primary source for all of them is William Bradford's own On Plymouth Plantation.
Private corporations do make large, long-term investments sometimes. Witness the Trans-Alaska Pipeline. $8 billion invested and the oil didn't start to flow for over three years. It does help, though, to have a well-defined market so the risk can be spread out on the financial markets. Everyone knows that oil is valuable, but no one can be sure what space is good for.
The history of Vermont & Iowa is illustrative here. Vermont was legally part of Massachsetts, and land was supposed to be allocated by an English-style royal charter. But a bunch of settlers with guns made their own rules, saying, "Who's gonna stop us?" When the same thing started happening in Maine, they held a plebescite and broke off from Massachusetts peacefully.
When Iowa was settled, an ad-hoc legal tradition developed defining property rights of increasing strength as a settler's presence on the land became more established. If you simply mark the land by slashing down foliage, you don't own it, but get first dibs to farm it. If you farm it and build a cabin, then you own it in the eyes of the local courts. The Ohio/Michigan model (from the Northwest Ordinance of 1787) of the gov't selling the land to land companies who then surveyed and subdivided never could be enforced. The Homestead Act came later and codified what was already common practice.
It's worse than that. Laws are complex because the people want them that way. Or at least, the politicians offer laws up to solve problem X and then the people fall for it.
Hanging murderers not enough? Let's have gun control. Gun control not working? Let's have restrictions on how guns are portrayed in movies, etc. People getting bored with (content-regulated) movies and playing video games instead? Let's put legal restictions on the video games. Oops, what exactly is a video game? No one cared before a prison sentence hung on the distinction. Enter 5,000-word definition of the term "video game".
The tax code is even worse because it is used as an instrument to so many contradictory purposes. To much income inequality? Progressive rates. Steep rates crushing certain segments of the economy? Put in exemptions (loopholes) for various circumstances. Loopholes being abused? Tighten the loopholes with more complex verbiage. Oops, penalizing marriage? Satisfy conservatives by allowing gays to get married so they pay more tax...uh, wait a minute there, I lost track.
Legalistic smother should be familliar most here. It is esstially the same thing as software bloat. Zillions of fixes piled up with no thought given to architecture or even a clear definintion of the problem at hand.
Though it will have its uses, I suspect that voice-recognition input will be less useful than is generally assumed. There are many problems that don't lend themselves to technical solutions.
Written language is different from spoken language. Sentence structure can be more complex when you have time to edit, and the reader can review at his own pace. Homophones need not be avoided. How do you edit a word on a page by speaking? Literal transcripts of non-trivial conversations are full of spoken corrections, clarifications, and follow-ups to the extent that they are cumbersome to read.
Written language contains symbolic hints not present in spoken language. For example, the parent post contains 64-hyphen-bit, motherf***er, 1982 written as number instead of a word, and capitalization of Star Trek.
Spoken language contains audible hints (eg, tonality, speed) not duplicated in written language. We would have to learn to not rely on them.
Page formatting doesn't exist in spoken language, nor does punctuation. Problems like proper quotation marks and poetic stanzas are probably AI-complete.
Typing is silent. A room full of people talking to their machines would be a cacophony.
No, that is perfectly consistent with UK usage. As a mental crutch, just tack on a -people suffix on company names:
"Sun-people Want to Buy Novell-people."
While the idea of a plane "invisible to radar" is an oversimplification, it is just as oversimplified to call stealth useless. Stealth complicates getting target lock-on with air-to-air missiles, and makes traditional jamming methods more effective.
A better analogy is armor. Whatever armor you have, the enemy can make ordnance to penetrate it. But is still better to defend against an armor brigade with $2 million Abrahms tanks, rather than a fleet of $20,000 jeeps.
The Concorde was profitable on an operating basis (or so BA/Air France claimed), but it never recovered its development costs. The economic case for SST's doesn't look promising though. In the 1960's, jets supplanted piston-driven aircraft because, even though they costed much more, they flew much faster and required less maintenence so they returned more passenger miles in a day. Jets burn more fuel, but jet fuel is cheaper than aviation gas. An SST burns more of the same type of fuel and turn-around time on the ground limits how much you gain in utilization rate. To be profitable, an SST must command a Concorde-sized ticket price. Now if somebody figures out how to make a fuel-efficient SST, then the cost goes down and trans-pacific (and trans-asian if you solve the sonic boom) flights become possible, allowing the aircraft to spend a greater fraction of the time in the air making money.
There was more to it than that. One of the design goals for Boeing's SST was trans-pacific endurance. When they found that the required fuel load didn't leave any payload for passengers, the economic case fell apart: A fuel stop in Hawaii killed the speed advantge vs. 747's, and they couldn't fly NY to LA (sonic boom). Transatlantic flights alone couldn't recover the development costs (as evidenced by the Concorde, which never recovered its development costs).
PM was once a serious and useful publication, but that must have been more than 15 years ago.
Way more than that. As early as the late 60's (the oldest issues in my memory) PS was pretty much a gee-whiz magazine not intended for actual practitioners of science or engineering. But if you look at pre-WWII reprints, it has the flavor of a trade rag, with actual practical knowlege for projects (homebuilt radio sets were common, I think this was before a separate Popular Electronics) The premiere ussue was in the 19-teens I think, and has a newsletter appearance to it. I think one of the feature articles was on telegraph sets.
OK, way off topic here, but that phrase is not a correct Yoda-ism. The order is predicate-subject-verb, not German/Latin style subject-object-verb. (I don't think the movies are entirely consistent though.)
If so strong in the Force Yoda is, why words in the right order he cannot put?
They built their uber expensive gear by taking apart consumer systems and repackaging them.
Is that necessarily bad? Limo builders and custom van shops do basically the same thing. I can see cracking open a piece of consumer-grade gear, putting on better connectors & capacitors, a quieter power supply, jeweled bearing inserts on the plastic gears, etc. as a perfectly sensible way to produce a top-notch unit.
Well, if you ask me for Bill Gates' & Microsoft's Great Contribution to the World, I would say it boils down to
a) Recognizing that software publishing differs from book publishing.
b) Knowing that, in the early days of the PC's it was more important for "platform" software (OS, programming languages, etc) to be cheap, universal, & early to market, rather than high-quality & bug-free.
Factor (a) is obvious to everyone now, of course. But it's easy to forget that, in the early 80's, the assumption was that most PC software would be sold at retail stores, stocked on shelves displaying descriptive boxes. Even the box-shipping software discounters were somewhat revolutionary: No one bought books that way. We now know, of course, that retail stores get stuck with outdated versions, complicating inventory mangement. Similarly, it took a long time for the industry to grok the shareware model; conventional wisdom didn't forsee Phil Katz' shareware PKZIP crushing the established SEA's ARC, even after SEA forced Phil Katz to use an incompatible format.
On factor (b), it is easy to forget that, 20 years ago, a typical PC user's thirst for featureful software would often cost as much as the hardware. GEM was better than Win3.1 by a mile, but it cost $80, while the OEM paid less than half that for Win. Wordperfect typically cost $250, and 123 amost as much. Microsoft's bundling practices spread the costs out (& reduced the incentive for piracy) and arguably avoided duplication of effort in the industry.
What has changed is that, as the amount of software out there has accumulated, the thirst has been mostly slaked, and most current work on desktop software is either custom, or adapting to new platforms, protocols etc. As the industry matured, Microsoft's bundling and upgrading practices have changed from constructive (by avoiding duplication of effort) to destructive (a profusion of undocumented and incompatible features, formats & protocols).
If, say, Gary Kildall lived and Bill Gates died, the state of the industry would probably be better today, but it would have taken longer and costed more to get here. So reasonable people can disagree whether, in toto, Microsoft's actions were a net positive contribution to the state of the art. But speaking today it is sort of irrelevant, it is plain to any informed person that today they are a net drag on the state of the art. They have been handsomely rewarded for their efforts already; the legal regime that allows them to continue to extract monopoly profits is a weakness in current public policy.
On security, we don't have to imagine what a windows-free world would look like; that was the state of the Arpanet/Internet in 1988 when the Robert Morris Worm broke out. The difference is
(a) Few then took seriously the threat of a worm (who would bother?)
(b) The Unix world quickly changed its practices and Unix-hosted worms have been comparatively rare ever since.
Microsoft has less of an excuse because they introduced vulnerable network features when the threat of worms was common knowledge. Linux is more secure not just because of its minority presence (though that helps) but also:
(a) Open source code reduces the number of security bugs.
(b) Secure design doesn't have to compete with an interest to lock-in customers with a profusion of poorly-tested features.
(c) Security through diversity. The open architecture permits diverse configurations, reducing the impact of any one security bug. A bug in sendmail, for example, won't affect Exim users.
you can go on about the business tactics bla bla, but surely microsoft would not have had the leverage on the big pc makers if they themselves were confident in the ability of other os's (im sure the guys at Dell, HP, etc were clued into what software was available at the time)
The hardware makers were beholden to MS because they needed MS-DOS. Want MS-DOS? Not allowed to pre-install GEM. (This is what everyone complains about.) By the time DR-DOS came along in the 90's Windows had become nearly universal and no one was writing aftermarket software for DOS anymore. (The DOS clones destroyed MS in the embedded PC market, though.) Now all the hardware makers are beholden to MS for Windows. IBM had a less restricted license to MS-DOS and for a while they fought it out with OS/2. But IBM lost because their hardware wasn't competitive, and they didn't market OS/2 effectively (they should have licnesed it for free the first few years). It didn't help that pre-2.2 versions of OS/2 weren't much better than contemporary versions of Windows. This was the golden age of MS astroturfing: MS shills, posing as BIX and Canopus users, promoted the MS API and dissed OS/2 among develpers, and got away with it.
If the PC makers were really smart, they would have formed an industry consortium in the 80's to reverse-engineer DOS, propose improvements, and maybe license ehancements on RAND terms on condition of open source code. But they abrogated this responsibility to MS now we all are paying the price.
As far as struggling with installs, all operating system installs can be difficult. The difference is that, when you buy a new PC, a team of 5 specialists at Dell spent 2 weeks getting Win to install correctly. With Linux you are on your own or have to buy support separately. My experience with installs is that Windows either works or it doesn't; if it doesn't you either need to download something from the hardware maker, or buy new hardware (or more software). You never spend much time struggling with an install but you may end up spending money, or not getting what you want. Linux installs run into problems more often, but when you do, they are easier to fix. Even major roadblocks won't stop you if you invest enough time; you can keep peeling the onion back, troubleshooting and configuring and then it almost always works.
I'm not going to say Linux is best for everyone, or even that Windows doesn't have some technical strong points vis-a-vis Linux. You have to choose (as with any tool) based on your needs. Windows is fine if you want to get working quickly, and security and long-term data retention are not major concerns. Linux looks better when you consider the long term; it can be pain to get it just right, but you only have to install once, then the benefits of its versatility, security, etc. pay off year after year. I have seen this called the "Newtonian" property: difficult to get moving, but unstoppable once started.
Plug-n-play is similar to the install experience. When it works, it is bone-head easy. When it doesn't, you have to buy new hardware. Sometimes a device will install but not behave the way you want. Suppose you want wireless 802.11b to seek out a connection only when the ethernet can't connect, for example. With Win you pretty much get what they give you, or have to find some aftermarket software to do what you want. Linux is easier to customize here, but it has to expose the underlying workings in order to make things customizable, so the install will necessarily be more complicated.
Mac sorta falls in between Windows and Linux. You get a group of experts to install it for you (like Windows), and its Unix nature allows you to "peel the onion" when necessary (like Linux). But being a minority in the market gives it some Linucoid problems; many aftermarket accessories & services assume you run Windows, for example. And non-free software gives it some Windoid problems; unwanted upgrades can be forced on you, for example.
This post went on longer than I planned...I'll stop typing now.
I think you actually agree with me. I'm not claiming that EULAs are valid. I offer that as a conceit and show that it is ridiculus. And yes, a EULA term that says, "Don't buy music from Real Networks," is another ridiculus consequence of EULAs' legal theory. The terms can be totally capricious because they never really were agreed to.
How short people's memories are. I would claim that Windows was never the best offering in the OS market. Win 3.1 was bettered by GEM, Win 95 by NexStep, and recent versions by Linux. Windows acheived its dominance, then and now, by MS's control of the distribution channel, something they inherited from the days of DOS and have jealously guarded ever since. It ease of use and compatibility are a result of being pre-installed on most machines, not the result of any "innovation". In other words, the only reason Windows is as good as it is, is because Dell, HP, etc. work so hard to make it that way.
Its pricipal technical merit is not sucking so bad that people dump their pre-installs pay money for something else.
Heh, she was a pretty girl & all (as the guys in the machine shop were always quick to point out), but she worked alongside my then-fiancee, same company but different building, so I wasn't "in the market" so to speak.
They did make a corny little award certificate made with one of those silly award ceritificate generator programs, "For excellent work in flux clean-up...". Come to think of it, I probably intalled that program for them; I would often swing by that building to help out because they weren't getting any technical support over there. That's how I met my wife.
So its true all you slashdot 20-somethings...it really is possible to pick up quality chicks by offering free technical support!
Until 6 or 7 years ago, solder had rosin flux and trichloroethylene was the board-cleaning solvent of choice. It works really well but now it is nearly banned for the sake of the ozone layer. Most shops now use a lighter-weight organic-acid flux and water as a solvent. Water doesn't work as well as trichlor but you can make up for it by using more of it. It is also possible to use alcohol, but its flammability makes it too hazardous for mass production use. No-clean fluxes are hard and flaky so they don't attract contaminants and you don't have to clean the board, but they don't work very well as a flux.
Often assemblies will have a combination of, for example, water-cleaned machine-assembled components and connectors soldered on by hand using, perhaps, a no-clean flux.
In a similar incident, I was handed a keyfob transmitter that, after being dropped into a toilet during late-night partying, no longer worked even after having dried out. Upon cracking the unit open, I saw the el-cheapo assembly process had left solder flux on the circuit board, so I scrubbed it clean with alcohol. It worked. Apparently the flux picked up some contaminants that de-tuned the radio transmitter.
Good thing too...without that transmitter she was stranded 60 miles from home.
The high pitched squealing combined with acrid smoke is suggestive that the culprit was an epoxy-coated Tantalum capacitor (a particular type of electrolytic). Aluminum type electrolyics are always vented so they explode, shower sparks, or just smolder, depending on how fast the failure is.
For the more common SRAM and EEPROM types, yes. The anti-fuse type, however, are not too bad on price-performance: 500 MHz clock and 2 million equivalent gates on-chip. Still only about a third of 90 nm CMOS (not counting Intel processors that are specifically architechted for high clock frequency) , but if you have a supercomputer budget you can use more of them and architect for more instructions per clock. They never became popular because you can't erase them, so prototyping is a pain. Imagine trashing a $15 chip every time you recompiled software.
Slashdot Mirror
The terms I used to initiate the search (Mayflower Compact, Plymouth Colony, William Bradford) would have been easy for most educated USians, but the info on your homepage seems to suggest you are Canadian. My assumption, my bad.
The Mayflower Compact is probably more promininent in my mind than most. One of my middle school teachers was a descendant of John Howland, one of the signers.
Try saying it in Klingon, in the orginal Klingon.
Well, to be charitable, that $10 billion may include insurance costs against business failure and the $11 billion may represent value in futures contracts. And if retreiving the asteroid takes 3 months on average, that's 40% annualized, risk-hedged return.
If they had better weapons, and a method for immigrants to purchase land,
You don't have to be hypothetical. Just look at the contrast between European colonies in the far east to those in the Americas.
John Stossel did a documentary of it, for one. Or read the Mayflower Compact. Good ol' Rush Limbaugh relates the same story. The primary source for all of them is William Bradford's own On Plymouth Plantation.
Or, you could simly have googled it yourself.
Private corporations do make large, long-term investments sometimes. Witness the Trans-Alaska Pipeline. $8 billion invested and the oil didn't start to flow for over three years. It does help, though, to have a well-defined market so the risk can be spread out on the financial markets. Everyone knows that oil is valuable, but no one can be sure what space is good for.
The history of Vermont & Iowa is illustrative here. Vermont was legally part of Massachsetts, and land was supposed to be allocated by an English-style royal charter. But a bunch of settlers with guns made their own rules, saying, "Who's gonna stop us?" When the same thing started happening in Maine, they held a plebescite and broke off from Massachusetts peacefully.
When Iowa was settled, an ad-hoc legal tradition developed defining property rights of increasing strength as a settler's presence on the land became more established. If you simply mark the land by slashing down foliage, you don't own it, but get first dibs to farm it. If you farm it and build a cabin, then you own it in the eyes of the local courts. The Ohio/Michigan model (from the Northwest Ordinance of 1787) of the gov't selling the land to land companies who then surveyed and subdivided never could be enforced. The Homestead Act came later and codified what was already common practice.
It's worse than that. Laws are complex because the people want them that way. Or at least, the politicians offer laws up to solve problem X and then the people fall for it.
Hanging murderers not enough? Let's have gun control. Gun control not working? Let's have restrictions on how guns are portrayed in movies, etc. People getting bored with (content-regulated) movies and playing video games instead? Let's put legal restictions on the video games. Oops, what exactly is a video game? No one cared before a prison sentence hung on the distinction. Enter 5,000-word definition of the term "video game".
The tax code is even worse because it is used as an instrument to so many contradictory purposes. To much income inequality? Progressive rates. Steep rates crushing certain segments of the economy? Put in exemptions (loopholes) for various circumstances. Loopholes being abused? Tighten the loopholes with more complex verbiage. Oops, penalizing marriage? Satisfy conservatives by allowing gays to get married so they pay more tax...uh, wait a minute there, I lost track.
Legalistic smother should be familliar most here. It is esstially the same thing as software bloat. Zillions of fixes piled up with no thought given to architecture or even a clear definintion of the problem at hand.
No, that is perfectly consistent with UK usage. As a mental crutch, just tack on a -people suffix on company names:
"Sun-people Want to Buy Novell-people."
A better analogy is armor. Whatever armor you have, the enemy can make ordnance to penetrate it. But is still better to defend against an armor brigade with $2 million Abrahms tanks, rather than a fleet of $20,000 jeeps.
The Concorde was profitable on an operating basis (or so BA/Air France claimed), but it never recovered its development costs. The economic case for SST's doesn't look promising though. In the 1960's, jets supplanted piston-driven aircraft because, even though they costed much more, they flew much faster and required less maintenence so they returned more passenger miles in a day. Jets burn more fuel, but jet fuel is cheaper than aviation gas. An SST burns more of the same type of fuel and turn-around time on the ground limits how much you gain in utilization rate. To be profitable, an SST must command a Concorde-sized ticket price. Now if somebody figures out how to make a fuel-efficient SST, then the cost goes down and trans-pacific (and trans-asian if you solve the sonic boom) flights become possible, allowing the aircraft to spend a greater fraction of the time in the air making money.
There was more to it than that. One of the design goals for Boeing's SST was trans-pacific endurance. When they found that the required fuel load didn't leave any payload for passengers, the economic case fell apart: A fuel stop in Hawaii killed the speed advantge vs. 747's, and they couldn't fly NY to LA (sonic boom). Transatlantic flights alone couldn't recover the development costs (as evidenced by the Concorde, which never recovered its development costs).
PM was once a serious and useful publication, but that must have been more than 15 years ago.
Way more than that. As early as the late 60's (the oldest issues in my memory) PS was pretty much a gee-whiz magazine not intended for actual practitioners of science or engineering. But if you look at pre-WWII reprints, it has the flavor of a trade rag, with actual practical knowlege for projects (homebuilt radio sets were common, I think this was before a separate Popular Electronics) The premiere ussue was in the 19-teens I think, and has a newsletter appearance to it. I think one of the feature articles was on telegraph sets.
If so strong in the Force Yoda is, why words in the right order he cannot put?
They built their uber expensive gear by taking apart consumer systems and repackaging them.
Is that necessarily bad? Limo builders and custom van shops do basically the same thing. I can see cracking open a piece of consumer-grade gear, putting on better connectors & capacitors, a quieter power supply, jeweled bearing inserts on the plastic gears, etc. as a perfectly sensible way to produce a top-notch unit.
Well, if you ask me for Bill Gates' & Microsoft's Great Contribution to the World, I would say it boils down to
a) Recognizing that software publishing differs from book publishing.
b) Knowing that, in the early days of the PC's it was more important for "platform" software (OS, programming languages, etc) to be cheap, universal, & early to market, rather than high-quality & bug-free.
Factor (a) is obvious to everyone now, of course. But it's easy to forget that, in the early 80's, the assumption was that most PC software would be sold at retail stores, stocked on shelves displaying descriptive boxes. Even the box-shipping software discounters were somewhat revolutionary: No one bought books that way. We now know, of course, that retail stores get stuck with outdated versions, complicating inventory mangement. Similarly, it took a long time for the industry to grok the shareware model; conventional wisdom didn't forsee Phil Katz' shareware PKZIP crushing the established SEA's ARC, even after SEA forced Phil Katz to use an incompatible format.
On factor (b), it is easy to forget that, 20 years ago, a typical PC user's thirst for featureful software would often cost as much as the hardware. GEM was better than Win3.1 by a mile, but it cost $80, while the OEM paid less than half that for Win. Wordperfect typically cost $250, and 123 amost as much. Microsoft's bundling practices spread the costs out (& reduced the incentive for piracy) and arguably avoided duplication of effort in the industry.
What has changed is that, as the amount of software out there has accumulated, the thirst has been mostly slaked, and most current work on desktop software is either custom, or adapting to new platforms, protocols etc. As the industry matured, Microsoft's bundling and upgrading practices have changed from constructive (by avoiding duplication of effort) to destructive (a profusion of undocumented and incompatible features, formats & protocols).
If, say, Gary Kildall lived and Bill Gates died, the state of the industry would probably be better today, but it would have taken longer and costed more to get here. So reasonable people can disagree whether, in toto, Microsoft's actions were a net positive contribution to the state of the art. But speaking today it is sort of irrelevant, it is plain to any informed person that today they are a net drag on the state of the art. They have been handsomely rewarded for their efforts already; the legal regime that allows them to continue to extract monopoly profits is a weakness in current public policy.
On security, we don't have to imagine what a windows-free world would look like; that was the state of the Arpanet/Internet in 1988 when the Robert Morris Worm broke out. The difference is
(a) Few then took seriously the threat of a worm (who would bother?)
(b) The Unix world quickly changed its practices and Unix-hosted worms have been comparatively rare ever since.
Microsoft has less of an excuse because they introduced vulnerable network features when the threat of worms was common knowledge. Linux is more secure not just because of its minority presence (though that helps) but also:
(a) Open source code reduces the number of security bugs.
(b) Secure design doesn't have to compete with an interest to lock-in customers with a profusion of poorly-tested features.
(c) Security through diversity. The open architecture permits diverse configurations, reducing the impact of any one security bug. A bug in sendmail, for example, won't affect Exim users.
There I went with another long post again...
you can go on about the business tactics bla bla, but surely microsoft would not have had the leverage on the big pc makers if they themselves were confident in the ability of other os's (im sure the guys at Dell, HP, etc were clued into what software was available at the time)
The hardware makers were beholden to MS because they needed MS-DOS. Want MS-DOS? Not allowed to pre-install GEM. (This is what everyone complains about.) By the time DR-DOS came along in the 90's Windows had become nearly universal and no one was writing aftermarket software for DOS anymore. (The DOS clones destroyed MS in the embedded PC market, though.) Now all the hardware makers are beholden to MS for Windows. IBM had a less restricted license to MS-DOS and for a while they fought it out with OS/2. But IBM lost because their hardware wasn't competitive, and they didn't market OS/2 effectively (they should have licnesed it for free the first few years). It didn't help that pre-2.2 versions of OS/2 weren't much better than contemporary versions of Windows. This was the golden age of MS astroturfing: MS shills, posing as BIX and Canopus users, promoted the MS API and dissed OS/2 among develpers, and got away with it.
If the PC makers were really smart, they would have formed an industry consortium in the 80's to reverse-engineer DOS, propose improvements, and maybe license ehancements on RAND terms on condition of open source code. But they abrogated this responsibility to MS now we all are paying the price.
As far as struggling with installs, all operating system installs can be difficult. The difference is that, when you buy a new PC, a team of 5 specialists at Dell spent 2 weeks getting Win to install correctly. With Linux you are on your own or have to buy support separately. My experience with installs is that Windows either works or it doesn't; if it doesn't you either need to download something from the hardware maker, or buy new hardware (or more software). You never spend much time struggling with an install but you may end up spending money, or not getting what you want. Linux installs run into problems more often, but when you do, they are easier to fix. Even major roadblocks won't stop you if you invest enough time; you can keep peeling the onion back, troubleshooting and configuring and then it almost always works.
I'm not going to say Linux is best for everyone, or even that Windows doesn't have some technical strong points vis-a-vis Linux. You have to choose (as with any tool) based on your needs. Windows is fine if you want to get working quickly, and security and long-term data retention are not major concerns. Linux looks better when you consider the long term; it can be pain to get it just right, but you only have to install once, then the benefits of its versatility, security, etc. pay off year after year. I have seen this called the "Newtonian" property: difficult to get moving, but unstoppable once started.
Plug-n-play is similar to the install experience. When it works, it is bone-head easy. When it doesn't, you have to buy new hardware. Sometimes a device will install but not behave the way you want. Suppose you want wireless 802.11b to seek out a connection only when the ethernet can't connect, for example. With Win you pretty much get what they give you, or have to find some aftermarket software to do what you want. Linux is easier to customize here, but it has to expose the underlying workings in order to make things customizable, so the install will necessarily be more complicated.
Mac sorta falls in between Windows and Linux. You get a group of experts to install it for you (like Windows), and its Unix nature allows you to "peel the onion" when necessary (like Linux). But being a minority in the market gives it some Linucoid problems; many aftermarket accessories & services assume you run Windows, for example. And non-free software gives it some Windoid problems; unwanted upgrades can be forced on you, for example.
This post went on longer than I planned...I'll stop typing now.
I think you actually agree with me. I'm not claiming that EULAs are valid. I offer that as a conceit and show that it is ridiculus. And yes, a EULA term that says, "Don't buy music from Real Networks," is another ridiculus consequence of EULAs' legal theory. The terms can be totally capricious because they never really were agreed to.
How short people's memories are. I would claim that Windows was never the best offering in the OS market. Win 3.1 was bettered by GEM, Win 95 by NexStep, and recent versions by Linux. Windows acheived its dominance, then and now, by MS's control of the distribution channel, something they inherited from the days of DOS and have jealously guarded ever since. It ease of use and compatibility are a result of being pre-installed on most machines, not the result of any "innovation". In other words, the only reason Windows is as good as it is, is because Dell, HP, etc. work so hard to make it that way.
Its pricipal technical merit is not sucking so bad that people dump their pre-installs pay money for something else.
Heh, she was a pretty girl & all (as the guys in the machine shop were always quick to point out), but she worked alongside my then-fiancee, same company but different building, so I wasn't "in the market" so to speak.
They did make a corny little award certificate made with one of those silly award ceritificate generator programs, "For excellent work in flux clean-up...". Come to think of it, I probably intalled that program for them; I would often swing by that building to help out because they weren't getting any technical support over there. That's how I met my wife.
So its true all you slashdot 20-somethings...it really is possible to pick up quality chicks by offering free technical support!
Until 6 or 7 years ago, solder had rosin flux and trichloroethylene was the board-cleaning solvent of choice. It works really well but now it is nearly banned for the sake of the ozone layer. Most shops now use a lighter-weight organic-acid flux and water as a solvent. Water doesn't work as well as trichlor but you can make up for it by using more of it. It is also possible to use alcohol, but its flammability makes it too hazardous for mass production use. No-clean fluxes are hard and flaky so they don't attract contaminants and you don't have to clean the board, but they don't work very well as a flux.
Often assemblies will have a combination of, for example, water-cleaned machine-assembled components and connectors soldered on by hand using, perhaps, a no-clean flux.
In a similar incident, I was handed a keyfob transmitter that, after being dropped into a toilet during late-night partying, no longer worked even after having dried out. Upon cracking the unit open, I saw the el-cheapo assembly process had left solder flux on the circuit board, so I scrubbed it clean with alcohol. It worked. Apparently the flux picked up some contaminants that de-tuned the radio transmitter.
Good thing too...without that transmitter she was stranded 60 miles from home.
The high pitched squealing combined with acrid smoke is suggestive that the culprit was an epoxy-coated Tantalum capacitor (a particular type of electrolytic). Aluminum type electrolyics are always vented so they explode, shower sparks, or just smolder, depending on how fast the failure is.
For the more common SRAM and EEPROM types, yes. The anti-fuse type, however, are not too bad on price-performance: 500 MHz clock and 2 million equivalent gates on-chip. Still only about a third of 90 nm CMOS (not counting Intel processors that are specifically architechted for high clock frequency) , but if you have a supercomputer budget you can use more of them and architect for more instructions per clock. They never became popular because you can't erase them, so prototyping is a pain. Imagine trashing a $15 chip every time you recompiled software.