I was being a bit cryptic for ironic effect, but my point was: why buy from an American company with all its management overhead when it can be easily undercut by a company where executive salaries aren't so high?
If the VC and regulatory situation gets straightened out in (for example) India, I have no doubt whatever that an Indian company could be just as well-run and innovative as an American one, even though executives would need far less salary for a given standard of living.
It's only a matter of time before this starts to happen.
Eventually, the most expensive employees -- VPs, CEOs, etc. -- will get outsourced as well. Why pay more to US companies for their exorbitant executive salaries when you can go directly to an offshore company?
I'm serious.
This is the most logical outcome.
And, of course, things will cost less as a result.
But what do you want to bet that by the time this starts
happening, the trade barriers will go back up?
Given that you've probably done more to help their security than your own, they have good reason to follow through.
This is why the "just ignore it" advice a number of folks are giving is wrong.
Even though you might have your network locked down perfectly tight, your attacker is likely to be coming from a compromised host.
Thus notifying the owner of that system is going to give them a lot of incentive to improve their security.
And that helps improve the whole network "ecosystem" in some small way.
In 2000, the Commission also found that prior to the elections, officials in Florida had removed thousands of names on the grounds that they were convicted felons. In fact, only a fraction of those removed actually had felony convictions; most were Floridians who had similar names. After the election, state officials ordered the names replaced on the rolls, but affected Floridians had already been barred from voting.
As I tried to express in my original comment, it's not just the license, though that's probably a factor.
Although there's no conflict here, per se, BSDers tend to prefer licenses with as few restrictions as possible. For example, this means that, all things being equal, they'll choose a BSD-style license over the GPL.
But in my experience pragmatism rules -- if the GNU tool is better than a BSD-licensed one, the superior tool is generally the one chosen. And I suspect that if XFree86 were technically better than X.org, this switch wouldn't be happening (though that's just my opinion based on conversations I've had).
The interesting thing is that the original BSD license was quite similar to the new XFree86 license.
The FSF (and others) objected to the original license since it had an "advertising clause" that read a lot like the new clause in the XFree86 license. The old BSD license read in part:
3. All advertising materials mentioning features or use of this software
must display the following acknowledgement:
This product includes software developed by the University of
California, Berkeley and its contributors.
Although you'll still find BSD software here and there with this clause, the trend has been to eliminate it wherever possible.
Thus XFree86's move to add san advertising clause is seen by many as a step backwards.
I don't think this shift is entirely a license issue.
I was chatting with one of the FreeBSD core team guys
around the time the decision was being made, and he felt that the frustration of getting fixes fed back into the XFree86 code base in a timely manner was a big part of the motivation. And this certainly isn't the first time I've heard complaints of XFree86 foot-dragging by the FreeBSD folks.
I guess you might say it's all of a piece -- the XFree86 user community simply didn't find the developers responsive (whether on license or technology), and when X.org proved a viable alternative, they voted with their feet.
I've bought several cassette decks and CD players/changers over the years, and they all included cables. Granted, they usually weren't very high-quality cables, but I have a pile of about 15 stereo RCA/RCA patch cords stashed away in a box to show for it.
Even the VCR/DVD unit I bought for the bedroom TV last month (we're not talking top-of-the-line here) had sufficent cables.
I'm normally an Apple fan [pats powerbook], but I think they slipped up on this one.
I think an increasing number of business computers are running little more than what comes with MS Windows and
MS Office, and perhaps another MS product or two, with the only third-party software perhaps being an antivirus and/or some remote backup tool. In other words, Microsoft's control of an increasing amount of the software marketplace is squeezing out other software vendors.
They're still available, both at retailers (including Apple's own storefronts) and online (although Apple isn't selling them from their own online "store" anymore). No doubt retailers are limited to their existing allotment. But I'd have to think that Apple's announcement will help stretch out the remaining supply of units for a bit longer than if they said nothing.
Nonetheless, I agree that it's a bad situation for them to be in.
Thinking about it, the announcement makes perfect sense. Usually pre-announcements have the effect of depressing sales as folks decide to wait for the upgraded version. That generally makes them a bad idea, but in this case it's exactly the results desired. It will help eke out the remaining inventory such that fewer people are left unhappy -- those that need the latest and greatest will wait, with the limited inventory going to those who can't wait.
Yeah, gotta love those NATO monikers.
I always thought names like "flogger" and "foxbat" were
great names for fighter aircraft (miles ahead of the
official MIG designation).
This is why you set up an alternate MX domain. Have a friend set up his SMTP server to relay your mail (and perhaps vice versa in a barter arrangement), and add an MX record to your DNS. Your ISP can disappear from the face of the earth and your mail will just queue up on your friend's server until you get your domain set up again.
Of course, this doesn't work if you share ISP's. Also, it's important that you have the same sort of redundancy with DNS (perhaps via a similar arrangement).
But if you're hardcore enough to set up your own SMTP server it's important to arrange for a backup -- you will need it, eventually.
When Linux came out it needed a MINIMUM of a '386. In fact, that was one of Linus's main motivations -- Tanenbaum had refused to create a '386-based version of Minux, insisting that most students could only afford a '286-based machine. At the same point in time, BSD 2.X ran on PDP-11/70's with less than a megabyte of main memory -- and unlike the '386, the PDP-11 was entirely a 16-bit machine and didn't have demand paging. "Minicomputers" were already on the way out when Linux and 386BSD appeared since the chief performance differences between a mini and the '386's were (1) data-center grade peripherals and (2) a price tag more than an order of magnitude larger.
Of course, the 2.X series of BSD Unix was a dead end (who wants 16-bit Unix based on a platform that was EOL'd nearly twenty years ago?), but it shows that BSD once ran with resources that are probably quite a bit less than the average PDA of today.
To make this a little closer to the article's topic:
the ABI of Bell Labs Unix was so widely known (and not just via the Lions book -- it's all in the man pages) that neither Tanenbaum nor Linus needed any knowledge of Unix internals whatsoever to replicate it. Furthermore, substantially all the details of those internals were published in Bell System Technical Journal articles and elsewhere. This is why the court found in the USL vs. Regents (BSD) suit that the cat was already out of the bag.
Album reviews can be useful if they are written by someone who has similar tastes to yours. Thus I find it useful to go back and look for good reviews of albums I especially like, and then look up other reviews written by the same reviewer. This doesn't always work -- musical taste is multidimensional. Thus I tend to associate a given genre with a favored reviewer and not automatically trust his/her reviews in other genres.
I agree that most reviewers, even ones whose tastes I share,
don't seem to capture very well what makes a given track "good" or "bad" or even "different." But it's still worthwhile seeking reviewers that seem to like what I like even if they can't explain why (and I) like it.
Use of assembly doesn't preclude thinking in terms of functions and algorithms.
Like nearly any form of programming, it pretty much requires such thought -- in abundance.
But given that I have a limited amount of attention to spend on each line of code, focusing on registers, branches, instruction sets and memory layout takes away from time better spent on clarity, modularity, and algorithmic sophistication.
It's much easier to take code written for clarity and correctness and make it fast, than take code written for speed and make it clear and correct.
That's what profilers and coverage tools are for.
Once you've measured, code your inner loops and bit-fiddling in assembler if you must, but only after your program is working and well-tested.
I'd have expected more complaints by now if the article were for-pay. I wonder if the article started out as free but folks at Fortune decided to make it subscriber-only as a way of dealing with the slashdot effect?
FreeBSD/OpenBSD/netBSD? Not at all. If it was impossible for Linux to create Linux and therefore Linux is TheftWare, the *BSDs are next in line for accusations and implications.
Doubtful.
After all, some of the original authors of BSD (e.g. Bill Joy) founded Sun.
They'd have to explain away some of their own history to do
that sort of thing.
The Solaris OS itself contributes relatively little to their income compared to hardware sales and support of both hardware and software.
The threat to them isn't so much Linux, per se, as Linux on cheap but powerful i386 hardware.
Companies that were dropping $1m on huge servers from Sun are now
spending $80,000 on a rack of 1U servers and getting much
better performance in the bargin.
Your comment:
Microsoft? They're not at all in position to capture the Intel *NIX market. Convert to XP? How?
Simply doesn't make sense -- you're acting like Linux somehow isn't a competitor to XP just because (you assert) no one would convert from *NIX to Microsoft. Not only is this false, but it ignores entirely Microsoft's problem with people converting in the other direction -- from Windows to Linux.
They are the most directly threatened.
They've said publicly and repeatedly that Linux is their biggest threat.
Why dismiss them so cavalierly?
The guy's surname is "Wilson," not "Plame."
If you'd actually read the book you'd know this -- it's on the cover.
His wife, Valerie Plame, "worked" for a CIA front organization.
Her name, the fact that she was married to Joe Wilson, and the name
of her front organization weren't secrets. But the fact that the organization
was a CIA front and Plame was a CIA operative and not just an international business consultant were secret. Revealing this meant that all of her contacts, some of which were involved in illegal arms sales and other activities of potential aid to terrorists, now know that she was a CIA plant. Whether she was presently active or not was immaterial -- The people she had dealt with now knew they had been compromised, and might well shut down ongoing operations and could "out" other operatives based on the chain of contacts. Also, consider that the claim she was no longer active could itself be a cover (think "damage control").
In other words, someone decided that getting revenge on Joe Wilson was worth sacrificing part of the war on terror. IMHO that's politics at its slimiest.
As for Iraq's attempts to obtain nuclear materials, it's well known that as of 1991 Iraq had a sophisticated nuclear program with capabilities for uranium enrichment. They were certainly attempting to obtain uranium from several sources, inclusing Nigeria. But we destroyed Iraq's nuclear program during and after Desert Storm, and there was little evidence that they had made significant progress in reconstructing it. Just because Joe Wilson mentions these earlier, pre-1991 attempts to obtain uranium doesn't contradict his claim that there were no such attempts ten years later.
People aren't going to have a uniform reaction to an ad;
one segment may consider it quite interesting, another
offensive.
A lot of folks might be quite grateful to see an ad for
flower delivery attached to a message saying someone
has died.
The clickthrough rate could be quite high.
Nonetheless, a lot of folks might be offended.
AdSense isn't going to be smart enough to separate the two
(it depends on the recipient and the recipient's relation to the deceased, among other things).
I use death as just the most obvious example of a trouble
area; there are lots of others (just about anything having
to do with relationships and social mores, for example).
This isn't an unsolvable problem by any means.
Google might have to screen ads very carefully
(a lot more carefully than they do for AdSense currently),
and turn down perfectly good ads as a result.
But I suspect there will be a learning period, and a few PR
snafus to overcome in the meantime.
I've learned of the deaths of people close to me via
email, twice.
I also first learned of the cancer that ultimately killed
my father, and my mother's alzheimers, via email.
People use email for the same sort of things they used to use snailmail and even phone calls for, and that includes delivering bad news.
I spent an hour or so yesterday going through news about the Columbine 5th aniversary. (There's a family connection that ties me to the tragedy.) Twice I encountered Google-based ads for shooting schools -- not exactly what I wanted to see.
I hope their ad selection for email is a bit more sensitive.
Another thing: you and I know quite well that keyword-based ads are just the result of some algorithm and not some faceless person perusing the text. But I suspect that a significant fraction of the public is going to find it creepy even if Google manages to avoid the negatives.
Five years from now when direct exposure to AI-based phenomena is more common this won't be as much of an issue. But it might be one now.
That would be funny, yes.
However, I've been signing posts/email/whatever with "-Ed" for longer than many slashdotters have been alive. I even sign handwritten letters that way.
The time to start to worry is if I change it to add a period
at the end...
I know how some people (and yahoo apparently) like to fly off the handle and claim the world is ending without bothering to even RTFA.
Actually, just a cursory examination of TFA shows that it's from Yahoo!'s Associated Press feed; I doubt if Yahoo! has an opinion on the matter.
I'm sure a lot of folks will wind up reading the same article in the business section of their newspaper, or via Google News, or wherever they might read Associated Press articles.
I'm not flaming you specifically (the mis-attribution was in the article summary, after all), but Slashdot in general tends to be incredibly sloppy in getting attributions correct.
I think Charles Steinmetz had at least as much influence.
His development of alternating current theory and the
law of electromagnetic hysteresis were crucial underpinnings
of AC power distribution engnineering.
Tesla may have been an inspired inventor, but as a
theoretician he was a lightweight compared to Steinmetz.
For example, Tesla may have invented the induction motor,
but it was Steinmetz's theories which showed how to make
it efficient.
The first customer for ICs was the military
(for the Minuteman Missile project, IIRC).
Later, NASA was another early adopter of the
technology.
The government is often the only one with deep enough
pockets to buy expensive but unproven technologies.
And it almost always contracts with private industry
to develop them.
Your "computer chips" might not even have been developed
without the Air Force and NASA, since who else would have
paid Fairchild, etc. to make them?
A simple logic gate once cost over a hundred pre-inflation
dollars...
That said, the bureaucratic monstrosity known as "NASA"
is a pale, bumbling and bloated organization with little
resemblance to the group that ran the Apollo
project.
Even away from the coasts old-growth trees thrived because
the larger, healthy trees typically survived the fires.
One issue with fire-suppression is that it allows so much
fuel accumulation that even the big trees die when fire
eventually happens.
Slashdot Mirror
I was being a bit cryptic for ironic effect, but my point was: why buy from an American company with all its management overhead when it can be easily undercut by a company where executive salaries aren't so high? If the VC and regulatory situation gets straightened out in (for example) India, I have no doubt whatever that an Indian company could be just as well-run and innovative as an American one, even though executives would need far less salary for a given standard of living.
It's only a matter of time before this starts to happen.
Eventually, the most expensive employees -- VPs, CEOs, etc. -- will get outsourced as well. Why pay more to US companies for their exorbitant executive salaries when you can go directly to an offshore company?
I'm serious. This is the most logical outcome. And, of course, things will cost less as a result. But what do you want to bet that by the time this starts happening, the trade barriers will go back up?
Given that you've probably done more to help their security than your own, they have good reason to follow through.
This is why the "just ignore it" advice a number of folks are giving is wrong. Even though you might have your network locked down perfectly tight, your attacker is likely to be coming from a compromised host. Thus notifying the owner of that system is going to give them a lot of incentive to improve their security. And that helps improve the whole network "ecosystem" in some small way.
Let's see; from the US Commission on Civil Right's Is America Ready to Vote? Election Readiness Briefing Paper, page 50, last paragraph:
(Emphasis added.)As I tried to express in my original comment, it's not just the license, though that's probably a factor.
Although there's no conflict here, per se, BSDers tend to prefer licenses with as few restrictions as possible. For example, this means that, all things being equal, they'll choose a BSD-style license over the GPL. But in my experience pragmatism rules -- if the GNU tool is better than a BSD-licensed one, the superior tool is generally the one chosen. And I suspect that if XFree86 were technically better than X.org, this switch wouldn't be happening (though that's just my opinion based on conversations I've had).
The interesting thing is that the original BSD license was quite similar to the new XFree86 license. The FSF (and others) objected to the original license since it had an "advertising clause" that read a lot like the new clause in the XFree86 license. The old BSD license read in part:
Although you'll still find BSD software here and there with this clause, the trend has been to eliminate it wherever possible. Thus XFree86's move to add san advertising clause is seen by many as a step backwards.I don't think this shift is entirely a license issue. I was chatting with one of the FreeBSD core team guys around the time the decision was being made, and he felt that the frustration of getting fixes fed back into the XFree86 code base in a timely manner was a big part of the motivation. And this certainly isn't the first time I've heard complaints of XFree86 foot-dragging by the FreeBSD folks.
I guess you might say it's all of a piece -- the XFree86 user community simply didn't find the developers responsive (whether on license or technology), and when X.org proved a viable alternative, they voted with their feet.
I've bought several cassette decks and CD players/changers over the years, and they all included cables. Granted, they usually weren't very high-quality cables, but I have a pile of about 15 stereo RCA/RCA patch cords stashed away in a box to show for it.
Even the VCR/DVD unit I bought for the bedroom TV last month (we're not talking top-of-the-line here) had sufficent cables.
I'm normally an Apple fan [pats powerbook], but I think they slipped up on this one.
I think an increasing number of business computers are running little more than what comes with MS Windows and MS Office, and perhaps another MS product or two, with the only third-party software perhaps being an antivirus and/or some remote backup tool. In other words, Microsoft's control of an increasing amount of the software marketplace is squeezing out other software vendors.
They're still available, both at retailers (including Apple's own storefronts) and online (although Apple isn't selling them from their own online "store" anymore). No doubt retailers are limited to their existing allotment. But I'd have to think that Apple's announcement will help stretch out the remaining supply of units for a bit longer than if they said nothing.
Nonetheless, I agree that it's a bad situation for them to be in.
Thinking about it, the announcement makes perfect sense. Usually pre-announcements have the effect of depressing sales as folks decide to wait for the upgraded version. That generally makes them a bad idea, but in this case it's exactly the results desired. It will help eke out the remaining inventory such that fewer people are left unhappy -- those that need the latest and greatest will wait, with the limited inventory going to those who can't wait.
Yeah, gotta love those NATO monikers. I always thought names like "flogger" and "foxbat" were great names for fighter aircraft (miles ahead of the official MIG designation).
But "fishbed?"
"One of these things is not like the others..."
This is why you set up an alternate MX domain. Have a friend set up his SMTP server to relay your mail (and perhaps vice versa in a barter arrangement), and add an MX record to your DNS. Your ISP can disappear from the face of the earth and your mail will just queue up on your friend's server until you get your domain set up again.
Of course, this doesn't work if you share ISP's. Also, it's important that you have the same sort of redundancy with DNS (perhaps via a similar arrangement). But if you're hardcore enough to set up your own SMTP server it's important to arrange for a backup -- you will need it, eventually.
Your post doesn't make much sense to me.
When Linux came out it needed a MINIMUM of a '386. In fact, that was one of Linus's main motivations -- Tanenbaum had refused to create a '386-based version of Minux, insisting that most students could only afford a '286-based machine. At the same point in time, BSD 2.X ran on PDP-11/70's with less than a megabyte of main memory -- and unlike the '386, the PDP-11 was entirely a 16-bit machine and didn't have demand paging. "Minicomputers" were already on the way out when Linux and 386BSD appeared since the chief performance differences between a mini and the '386's were (1) data-center grade peripherals and (2) a price tag more than an order of magnitude larger.
Of course, the 2.X series of BSD Unix was a dead end (who wants 16-bit Unix based on a platform that was EOL'd nearly twenty years ago?), but it shows that BSD once ran with resources that are probably quite a bit less than the average PDA of today.
To make this a little closer to the article's topic: the ABI of Bell Labs Unix was so widely known (and not just via the Lions book -- it's all in the man pages) that neither Tanenbaum nor Linus needed any knowledge of Unix internals whatsoever to replicate it. Furthermore, substantially all the details of those internals were published in Bell System Technical Journal articles and elsewhere. This is why the court found in the USL vs. Regents (BSD) suit that the cat was already out of the bag.
Album reviews can be useful if they are written by someone who has similar tastes to yours. Thus I find it useful to go back and look for good reviews of albums I especially like, and then look up other reviews written by the same reviewer. This doesn't always work -- musical taste is multidimensional. Thus I tend to associate a given genre with a favored reviewer and not automatically trust his/her reviews in other genres.
I agree that most reviewers, even ones whose tastes I share, don't seem to capture very well what makes a given track "good" or "bad" or even "different." But it's still worthwhile seeking reviewers that seem to like what I like even if they can't explain why (and I) like it.
Use of assembly doesn't preclude thinking in terms of functions and algorithms. Like nearly any form of programming, it pretty much requires such thought -- in abundance. But given that I have a limited amount of attention to spend on each line of code, focusing on registers, branches, instruction sets and memory layout takes away from time better spent on clarity, modularity, and algorithmic sophistication.
It's much easier to take code written for clarity and correctness and make it fast, than take code written for speed and make it clear and correct. That's what profilers and coverage tools are for. Once you've measured, code your inner loops and bit-fiddling in assembler if you must, but only after your program is working and well-tested.
I'd have expected more complaints by now if the article were for-pay. I wonder if the article started out as free but folks at Fortune decided to make it subscriber-only as a way of dealing with the slashdot effect?
Doubtful. After all, some of the original authors of BSD (e.g. Bill Joy) founded Sun. They'd have to explain away some of their own history to do that sort of thing.
The Solaris OS itself contributes relatively little to their income compared to hardware sales and support of both hardware and software. The threat to them isn't so much Linux, per se, as Linux on cheap but powerful i386 hardware. Companies that were dropping $1m on huge servers from Sun are now spending $80,000 on a rack of 1U servers and getting much better performance in the bargin.
Your comment:
Simply doesn't make sense -- you're acting like Linux somehow isn't a competitor to XP just because (you assert) no one would convert from *NIX to Microsoft. Not only is this false, but it ignores entirely Microsoft's problem with people converting in the other direction -- from Windows to Linux. They are the most directly threatened. They've said publicly and repeatedly that Linux is their biggest threat. Why dismiss them so cavalierly?
The guy's surname is "Wilson," not "Plame." If you'd actually read the book you'd know this -- it's on the cover. His wife, Valerie Plame, "worked" for a CIA front organization. Her name, the fact that she was married to Joe Wilson, and the name of her front organization weren't secrets. But the fact that the organization was a CIA front and Plame was a CIA operative and not just an international business consultant were secret. Revealing this meant that all of her contacts, some of which were involved in illegal arms sales and other activities of potential aid to terrorists, now know that she was a CIA plant. Whether she was presently active or not was immaterial -- The people she had dealt with now knew they had been compromised, and might well shut down ongoing operations and could "out" other operatives based on the chain of contacts. Also, consider that the claim she was no longer active could itself be a cover (think "damage control").
In other words, someone decided that getting revenge on Joe Wilson was worth sacrificing part of the war on terror. IMHO that's politics at its slimiest.
As for Iraq's attempts to obtain nuclear materials, it's well known that as of 1991 Iraq had a sophisticated nuclear program with capabilities for uranium enrichment. They were certainly attempting to obtain uranium from several sources, inclusing Nigeria. But we destroyed Iraq's nuclear program during and after Desert Storm, and there was little evidence that they had made significant progress in reconstructing it. Just because Joe Wilson mentions these earlier, pre-1991 attempts to obtain uranium doesn't contradict his claim that there were no such attempts ten years later.
People aren't going to have a uniform reaction to an ad; one segment may consider it quite interesting, another offensive. A lot of folks might be quite grateful to see an ad for flower delivery attached to a message saying someone has died. The clickthrough rate could be quite high. Nonetheless, a lot of folks might be offended. AdSense isn't going to be smart enough to separate the two (it depends on the recipient and the recipient's relation to the deceased, among other things).
I use death as just the most obvious example of a trouble area; there are lots of others (just about anything having to do with relationships and social mores, for example). This isn't an unsolvable problem by any means. Google might have to screen ads very carefully (a lot more carefully than they do for AdSense currently), and turn down perfectly good ads as a result. But I suspect there will be a learning period, and a few PR snafus to overcome in the meantime.
I've learned of the deaths of people close to me via email, twice. I also first learned of the cancer that ultimately killed my father, and my mother's alzheimers, via email. People use email for the same sort of things they used to use snailmail and even phone calls for, and that includes delivering bad news.
I spent an hour or so yesterday going through news about the Columbine 5th aniversary. (There's a family connection that ties me to the tragedy.) Twice I encountered Google-based ads for shooting schools -- not exactly what I wanted to see. I hope their ad selection for email is a bit more sensitive.
Another thing: you and I know quite well that keyword-based ads are just the result of some algorithm and not some faceless person perusing the text. But I suspect that a significant fraction of the public is going to find it creepy even if Google manages to avoid the negatives. Five years from now when direct exposure to AI-based phenomena is more common this won't be as much of an issue. But it might be one now.
IANAE
That would be funny, yes. However, I've been signing posts/email/whatever with "-Ed" for longer than many slashdotters have been alive. I even sign handwritten letters that way. The time to start to worry is if I change it to add a period at the end...
Actually, just a cursory examination of TFA shows that it's from Yahoo!'s Associated Press feed; I doubt if Yahoo! has an opinion on the matter. I'm sure a lot of folks will wind up reading the same article in the business section of their newspaper, or via Google News, or wherever they might read Associated Press articles.
I'm not flaming you specifically (the mis-attribution was in the article summary, after all), but Slashdot in general tends to be incredibly sloppy in getting attributions correct.
I think Charles Steinmetz had at least as much influence. His development of alternating current theory and the law of electromagnetic hysteresis were crucial underpinnings of AC power distribution engnineering. Tesla may have been an inspired inventor, but as a theoretician he was a lightweight compared to Steinmetz. For example, Tesla may have invented the induction motor, but it was Steinmetz's theories which showed how to make it efficient.
The first customer for ICs was the military (for the Minuteman Missile project, IIRC). Later, NASA was another early adopter of the technology. The government is often the only one with deep enough pockets to buy expensive but unproven technologies. And it almost always contracts with private industry to develop them. Your "computer chips" might not even have been developed without the Air Force and NASA, since who else would have paid Fairchild, etc. to make them? A simple logic gate once cost over a hundred pre-inflation dollars...
That said, the bureaucratic monstrosity known as "NASA" is a pale, bumbling and bloated organization with little resemblance to the group that ran the Apollo project.
Sad, sad, sad...
Even away from the coasts old-growth trees thrived because the larger, healthy trees typically survived the fires. One issue with fire-suppression is that it allows so much fuel accumulation that even the big trees die when fire eventually happens.