The question is really just who the armies report to. Someone you can vote out of office, where your vote is as good as the next guy's, or someone who owns your labour and makes you buy your goods at the company store - or your water?
There are more ways of lethally threatening people than with guns.
"Getting a bunch of statistically random people to sit in a lab and drink from either plastic or non-plastic cups for 50 years isn't really going to be possible."
"Next... Mr Smith."
"Hi."
"Oops, looks like the seed material is recycling. Jenkins, retune the entropy pool, and get a cleaner in here."
Kraboom!
"Next... Senor Frobuzz."
"Purple souffle! Unicorn toffee!"
"Excellent, welcome to the team. Would you like to work on financial forecasting, cryptography or Situationism?"
I don't get why people are slagging Tim off over this. We already do have such mechanisms on the small scale: karma points for comments, reputation systems for online trading, blogrolls and 'social bookmarking services' for 'this unknown website is recommended/suggested by this other website I read'.
Remember Advogato's rankings?
The logical next step would be to have a generic way of talking about such rankings/recommendations such that I don't need to subscribe to a third party to do it. Use, oh, I don't know, how about RDF? We've already got FOAF - how about an 'Enemy Of My Enemy' protocol?
Yes, this will lead to 'ontology wars' as groups with different views of trustworthiness start formalising the metrics they already use informally. As long as the protocol itself remains open and interconnectable, I don't see this as a huge problem. At least people will be openly owning their philosophical bias rather than pretending it doesn't exist.
"Guns can be used to kill people. They can also be used to save people."
No, guns can only kill (or not-quite-lethally wound, if you're a *very* skilled and extremely lucky shot). That's why handgun safety rules say you should never point a gun at something you don't want dead.
The best case outcome is that in a combat situation, where one creature is about to die anyway, you can choose to kill something you don't care about; trade one life for another. That's the closest a gun can come to 'saving a life'.
And you can only get that fake 'life saving' effect if you are truly prepared to remain completely emotionally unattached to the impact of the death of the living creature you shot. Doing that doesn't come without psychological consequences.
A simple first aid kit, on the other hand - that can *really* save lives, in a non-zero-sum way.
It always boggles me that we still run X - a protocol designed from the ground up to separate 'server' and 'clients' - in the same process tree such that killing the X server *also* kills the clients.
I mean, that just shouldn't happen, right? If the X server were on a separate host, you could restart it without killing the clients. How come we can't separate them on the same machine?
There is a way to run multiple X servers on a single Ubuntu session - I forget what it's called now, but a year ago I was using it to run Dosbox, because Dosbox has a nasty habit of crashing when some old crashy DOS games crash, and then leaving its X server in an unusable graphics mode. Run it under a separate X server, and voila, your real X desktop is left alone.
So I'm still baffled why, out of the box on all Linux distros, for all that X has this huge advantage of separating clients and servers, we still force you to kill all your clients when you Ctrl-Alt-Backspace.
"In Program Files? Of course not. It installs into your profile. Thankfully your local profile, so if you're actually using roaming profiles you won't be transferring the entire browser back and forth."
Ahahahahaha!
That would be why it Fails on our Novell Netware network.
Anything installed into the local profile gets erased on reboot for us and recreated at next reboot. That's a Novell quirk, but it assumes quite rightly that if you roam, *all* your user data is on the network and the only local stuff is temporary working cache, not application bits. The last app that failed this bad was Palm Desktop, and is why we no longer authorise Palms for purchase.
Nice going, Google.
Seriously, that local profile thing is insane. It's like Microsoft has completely ignored the existence of roaming users and assumes everyone now has their own personal laptop. If I install a program to a workstation, I want it on that workstation for all roaming users, because it's a *program* and it's got all sorts of dependencies on the local system, but it darn well ought to be able to create default user settings when it sees a new user. If I have some personal data, I want it in my roaming profile wherever I am.
THERE IS NO CONCEIVABLE SANE CIRCUMSTANCE where I would want a piece of corporate software infrastructure which is ONLY runnable for 'one user on one workstation'. Not on a managed network. Anyone who uses that insane place to put a whole *application* just completely fails at Windows application development (and operating system design).
Being noisy blinking rubbish is EXACTLY my problem with ads, especially Flash ads. So not Adblock for me, but Flashblock.
When I have to use IE (or Chrome) I wince. All those huge Flash ads! I can't concentrate on reading a page when there's about 100 square centimetres of mini-movie playing right next to it.
I remember when Symantec meant 'natural language query database company', not 'the people who bought Peter Norton'.
Ah Q&A, how I miss you. You might have been a wonky little DOS-based thing with less power than dBASE II... but it was the smell of the late '80s, before the AI winter.
You've never handled corporate software licencing, I take it? Pure pain it is. An administrative nightmare. Timed licences, demo licences, restrictive feature sets, yearly licences, two-yearly licences, monthly licences, per-workstation licences, per-user licences, simultaneous-connection licences, per-team licences, per-role licences, per-organisation licences, special discount licences, academic or industry partner licences which may or may not apply depending on which sub-organisation you consider yourself to be working for, leases, purchases, suppliers going out of business, dongles, keys, codes (any or all of which may be allergic to any other if installed on the same machine), replacement codes, upgrade codes, Internet phone-home licences, lockouts which can bring your entire business to a screaming halt if you're so much as a day late in renewing.
Every licence is potentially a single point of failure for your entire business and they all multiply, they don't add.
Every open-source, free licence that doesn't mandate its own version of bureaucratic tracking overhead is simply one less moving part to break. Yes, you still have to track security patches but you don't have to be forced into upgrades by artificial accounting deadlines or restrictive shrink-wrap agreements.
What *wouldn't* a reliability-focused industry like spaceflight like about that?
"They got basic TCP/IP running, then email on it, then started building local network apps, which they discussed on the network in "Request for Comment" messages which spec'ed the new app/protocol."
Mostly true, except for the TCP/IP bit. RFC 1 specified a much older protocol. ARPANET didn't fully switch over to TCP/IP until 1983.
"After all the race for the stars should be for humanity's sake, not just one country."
Given nobody has even the first inkling of a theoretical approach toward starting to work out how to talk about designing a warp drive, and most physicists seem to think it's a priori impossible, that would be a rather slow race.
Did you mean "the race to be the second country to land massively inefficient but impressive looking human rated spacecraft on a bunch of inhospitable rocks that aren't actually useful for anything either militarily or economically, and haven't been for fifty years?"
Yeah, we'll get right on that.
Or we could keep spending money on boring little LEO and GEO comms satellites, which are the only actually *useful* applications we've so far found for space.
Reality != Star Trek, and the stars aren't necessarily ours in this or anyone's lifetime.
"f I flood your IP address, 250 GB can disappear pretty fast, and there's really nothing you can do about it. Whether your router drops the packets or not, they'll still be counted against your quota.
Similar if you fire up a p2p program, and download a video or game level or whatever. Once you end it, thousands of other people are still going to be sending packets to your IP address, checking whether you're back online and can share the file."
Yes.
So the correct answer is *fix the protocols* which incorrectly assume that Internet traffic costs nothing and send ridiculously huge numbers of packets, not sweep the problem under the carpet by hiding it with an all-you-can-eat charging regime.
If enough people in North America start hurting financially because TCP/IP is broken, maybe it'll get fixed, and that'll in turn fix a whole raft of serious efficiency and DOS issues, not just the short-term 'my cable bill is too big!' one.
"Dammit Australia, just because you have crap internet, the rest of the world shouldn't have to accept it!"
There's a little thing called 'living within your means' which used to be considered a virtue. That's why we laugh at people who have ten times as much stuff as us and yet feel more hard done by. Grow some restraint. It'll be good for you.
Also, if you guys have ten times as much bandwidth as us, you'll make websites loaded down with useless Flash and vidcasts which are ten times bigger, you'll write operating systems which are blithely unaware that Internet is not a free commodity for some of us and have no concept of restricting transmissions to the necessary, and we'll get locked out of the Web by all your bloat.
So it's in our interest for broadband speeds charging regimes to be roughly the same all around the world - otherwise we end up the wrong side of the data gap.
And it's not crap, it's metered. You don't get free all-you-can-eat electricity or petrol or food each month - why should Internet capacity be different?
If you really want absolutely unlimited Internet with a charging regime completely uncoupled from usage, that means you want to socialise the cost of communications infrastructure. Fine, that's a valid political position and it's got some merit to it, but in that case you guys should already have free healthcare and be advocating for a Universal Basic Income.
Schadenfreude, I think. We get so sick of Americans and British thinking they own the world and exporting their nasty foreign policy while we get all our TV, movie and book imports second hand, six months late and twice as expensive.
Plus it's that old pioneer attitude of 'you can darn well learn to put up with anything, young man, it'll build character'.
For the last 20 years we've been busy dismantling our welfare state and installing Thatcherite/Reaganite 'user pays' laissez faire, it hurt like hell, and if WE had to suffer for that free-market crap, you guys who INVENTED it can stand to apply it yourself.
But seriously, if you're using over 250GB a month, that's just greedy. And while *information* is not a zero-sum commodity, bandwidth over a shared cable *is*. Every gigabyte you use around is a gigabyte someone else can't.
It's just basic fairness. You should pay for what you use. Don't sweep use costs under the carpet of hidden externalities. The same principle underlies good economics, good ecology, and good neighbourliness.
Ted Stevens was bought by Big Cable, but he was right, as far as he went, and it won't help to mock him. The Internet *is* a series of data pipes with finite bandwidth, and that needs to be paid by someone. Should really be local user-owned broadband cooperatives rather than profit-driven companies, but even then it'll cost you a nonzero amount to move packets, and it should.
It's the only Firefox add-in I actually use all the time. Can't stand autorunning Flash ads (or even Youtube videos). Adblock is too aggressive for me, but Flashblock is just perfect. One click and I see an animation, otherwise my sanity is preserved.
"How can you trust anything science says? It revises itself to be accurate, that's why. An unchanging theory constructed in year X is one that is unlikely to become more correct with the collection of new evidence in the years to come, except by some fluke of blind luck."
Right. But if a theory was correct *to begin with*, it doesn't actually need revising. The more correct and trustworthy a piece of scientific knowledge, the *less* it will change over time. So logically speaking, we should be more skeptical of new "discoveries" and more trusting of old knowledge. And the more and deeper "scientific revolutions" we encounter the more cynical it should make us of any claims to knowledge of any kind.
After a couple of such revolutions, we should feel deeply uneasy, uncertain, bewildered and emotionally devastated. We should feel traumatised, like a war or disaster survivor, because learning that people you respected were teaching you falsehoods and can give you no confidence that they are telling you the ultimate truth now *should* feel like a deep personal betrayal.
Feel familiar?
Robert Persig describes this phenomenon quite well in 'Zen and the Art of Motorcycle Maintenance'. The nature of science is to generate and disprove hypotheses, not produce 'truth'. There's an illusion that science is progressing and converging toward truth. But the more hypotheses we generate, the more contradictory alternatives we actually have in our minds, and the less able to respond to the world we become.
It's a little like Windows patches. If you keep having to retroactively fix things, you're doing it wrong.
"People interested in science enter research careers and have access from their universities or research organisations. People who are not scientists but are interested in science will use these resources for entertainment purposes, not to advance the body of knowledge."
Sheesh. With that attitude, do you seriously wonder *why* people in the street think scientists are 'elitists'? No scruffy unwashed masses allowed in OUR little club, no sir. They might be ENTERTAINED by knowledge! The horror! Knowledge must be locked up and kept secure lest it be defiled by grubby minds.
Thank God for Wikipedia. I've actually learned things by reading it. Stuff they never taught me in high school science.
Presumably to the manufacturers of hardware which contains binary-only drivers.
The idea is that it's a deliberately stress-testing distribution designed to be 100% Free and to cause any hardware which isn't Free to fail. If nobody complains that broken stuff is broken, it won't get fixed. And requiring binary drivers *is* breakage. As soon as the kernel updates, potentially wham! go your drivers if there's no source code.
"The reasonable man adapts himself to the conditions that surround him... The unreasonable man adapts surrounding conditions to himself... All progress depends on the unreasonable man." -- George Bernard Shaw.
Slashdot Mirror
The robber barons already have armies.
The question is really just who the armies report to. Someone you can vote out of office, where your vote is as good as the next guy's, or someone who owns your labour and makes you buy your goods at the company store - or your water?
There are more ways of lethally threatening people than with guns.
"Getting a bunch of statistically random people to sit in a lab and drink from either plastic or non-plastic cups for 50 years isn't really going to be possible."
"Next... Mr Smith."
"Hi."
"Oops, looks like the seed material is recycling. Jenkins, retune the entropy pool, and get a cleaner in here."
Kraboom!
"Next... Senor Frobuzz."
"Purple souffle! Unicorn toffee!"
"Excellent, welcome to the team. Would you like to work on financial forecasting, cryptography or Situationism?"
I don't get why people are slagging Tim off over this. We already do have such mechanisms on the small scale: karma points for comments, reputation systems for online trading, blogrolls and 'social bookmarking services' for 'this unknown website is recommended/suggested by this other website I read'.
Remember Advogato's rankings?
The logical next step would be to have a generic way of talking about such rankings/recommendations such that I don't need to subscribe to a third party to do it. Use, oh, I don't know, how about RDF? We've already got FOAF - how about an 'Enemy Of My Enemy' protocol?
Yes, this will lead to 'ontology wars' as groups with different views of trustworthiness start formalising the metrics they already use informally. As long as the protocol itself remains open and interconnectable, I don't see this as a huge problem. At least people will be openly owning their philosophical bias rather than pretending it doesn't exist.
"Guns can be used to kill people. They can also be used to save people."
No, guns can only kill (or not-quite-lethally wound, if you're a *very* skilled and extremely lucky shot). That's why handgun safety rules say you should never point a gun at something you don't want dead.
The best case outcome is that in a combat situation, where one creature is about to die anyway, you can choose to kill something you don't care about; trade one life for another. That's the closest a gun can come to 'saving a life'.
And you can only get that fake 'life saving' effect if you are truly prepared to remain completely emotionally unattached to the impact of the death of the living creature you shot. Doing that doesn't come without psychological consequences.
A simple first aid kit, on the other hand - that can *really* save lives, in a non-zero-sum way.
Yes!
It always boggles me that we still run X - a protocol designed from the ground up to separate 'server' and 'clients' - in the same process tree such that killing the X server *also* kills the clients.
I mean, that just shouldn't happen, right? If the X server were on a separate host, you could restart it without killing the clients. How come we can't separate them on the same machine?
There is a way to run multiple X servers on a single Ubuntu session - I forget what it's called now, but a year ago I was using it to run Dosbox, because Dosbox has a nasty habit of crashing when some old crashy DOS games crash, and then leaving its X server in an unusable graphics mode. Run it under a separate X server, and voila, your real X desktop is left alone.
So I'm still baffled why, out of the box on all Linux distros, for all that X has this huge advantage of separating clients and servers, we still force you to kill all your clients when you Ctrl-Alt-Backspace.
"In Program Files? Of course not. It installs into your profile. Thankfully your local profile, so if you're actually using roaming profiles you won't be transferring the entire browser back and forth."
Ahahahahaha!
That would be why it Fails on our Novell Netware network.
Anything installed into the local profile gets erased on reboot for us and recreated at next reboot. That's a Novell quirk, but it assumes quite rightly that if you roam, *all* your user data is on the network and the only local stuff is temporary working cache, not application bits. The last app that failed this bad was Palm Desktop, and is why we no longer authorise Palms for purchase.
Nice going, Google.
Seriously, that local profile thing is insane. It's like Microsoft has completely ignored the existence of roaming users and assumes everyone now has their own personal laptop. If I install a program to a workstation, I want it on that workstation for all roaming users, because it's a *program* and it's got all sorts of dependencies on the local system, but it darn well ought to be able to create default user settings when it sees a new user. If I have some personal data, I want it in my roaming profile wherever I am.
THERE IS NO CONCEIVABLE SANE CIRCUMSTANCE where I would want a piece of corporate software infrastructure which is ONLY runnable for 'one user on one workstation'. Not on a managed network. Anyone who uses that insane place to put a whole *application* just completely fails at Windows application development (and operating system design).
Being noisy blinking rubbish is EXACTLY my problem with ads, especially Flash ads. So not Adblock for me, but Flashblock.
When I have to use IE (or Chrome) I wince. All those huge Flash ads! I can't concentrate on reading a page when there's about 100 square centimetres of mini-movie playing right next to it.
I remember when Symantec meant 'natural language query database company', not 'the people who bought Peter Norton'.
Ah Q&A, how I miss you. You might have been a wonky little DOS-based thing with less power than dBASE II... but it was the smell of the late '80s, before the AI winter.
"And you know the old saying about bring a bat'leth to a phaser fight."
"It's hard to pull a trigger when your severed arm's on the other side of the holodeck?"
Or was it "big bat'leth, big... shoes?"
First time I saw a Borg I went 'what, cheap Cybermen ripoff! Get your own monster, you Yanks!'
Do we get the funky hats and robes from Encounter At Farpoint as well?
You've never handled corporate software licencing, I take it? Pure pain it is. An administrative nightmare. Timed licences, demo licences, restrictive feature sets, yearly licences, two-yearly licences, monthly licences, per-workstation licences, per-user licences, simultaneous-connection licences, per-team licences, per-role licences, per-organisation licences, special discount licences, academic or industry partner licences which may or may not apply depending on which sub-organisation you consider yourself to be working for, leases, purchases, suppliers going out of business, dongles, keys, codes (any or all of which may be allergic to any other if installed on the same machine), replacement codes, upgrade codes, Internet phone-home licences, lockouts which can bring your entire business to a screaming halt if you're so much as a day late in renewing.
Every licence is potentially a single point of failure for your entire business and they all multiply, they don't add.
Every open-source, free licence that doesn't mandate its own version of bureaucratic tracking overhead is simply one less moving part to break. Yes, you still have to track security patches but you don't have to be forced into upgrades by artificial accounting deadlines or restrictive shrink-wrap agreements.
What *wouldn't* a reliability-focused industry like spaceflight like about that?
"They got basic TCP/IP running, then email on it, then started building local network apps, which they discussed on the network in "Request for Comment" messages which spec'ed the new app/protocol."
Mostly true, except for the TCP/IP bit. RFC 1 specified a much older protocol. ARPANET didn't fully switch over to TCP/IP until 1983.
Those early RFCs are fascinating reading actually. http://www.faqs.org/rfcs/
"This is America... This isn't a... country where we care..."
Would make a nice national epitaph.
"After all the race for the stars should be for humanity's sake, not just one country."
Given nobody has even the first inkling of a theoretical approach toward starting to work out how to talk about designing a warp drive, and most physicists seem to think it's a priori impossible, that would be a rather slow race.
Did you mean "the race to be the second country to land massively inefficient but impressive looking human rated spacecraft on a bunch of inhospitable rocks that aren't actually useful for anything either militarily or economically, and haven't been for fifty years?"
Yeah, we'll get right on that.
Or we could keep spending money on boring little LEO and GEO comms satellites, which are the only actually *useful* applications we've so far found for space.
Reality != Star Trek, and the stars aren't necessarily ours in this or anyone's lifetime.
"f I flood your IP address, 250 GB can disappear pretty fast, and there's really nothing you can do about it. Whether your router drops the packets or not, they'll still be counted against your quota.
Similar if you fire up a p2p program, and download a video or game level or whatever. Once you end it, thousands of other people are still going to be sending packets to your IP address, checking whether you're back online and can share the file."
Yes.
So the correct answer is *fix the protocols* which incorrectly assume that Internet traffic costs nothing and send ridiculously huge numbers of packets, not sweep the problem under the carpet by hiding it with an all-you-can-eat charging regime.
If enough people in North America start hurting financially because TCP/IP is broken, maybe it'll get fixed, and that'll in turn fix a whole raft of serious efficiency and DOS issues, not just the short-term 'my cable bill is too big!' one.
"Dammit Australia, just because you have crap internet, the rest of the world shouldn't have to accept it!"
There's a little thing called 'living within your means' which used to be considered a virtue. That's why we laugh at people who have ten times as much stuff as us and yet feel more hard done by. Grow some restraint. It'll be good for you.
Also, if you guys have ten times as much bandwidth as us, you'll make websites loaded down with useless Flash and vidcasts which are ten times bigger, you'll write operating systems which are blithely unaware that Internet is not a free commodity for some of us and have no concept of restricting transmissions to the necessary, and we'll get locked out of the Web by all your bloat.
So it's in our interest for broadband speeds charging regimes to be roughly the same all around the world - otherwise we end up the wrong side of the data gap.
And it's not crap, it's metered. You don't get free all-you-can-eat electricity or petrol or food each month - why should Internet capacity be different?
If you really want absolutely unlimited Internet with a charging regime completely uncoupled from usage, that means you want to socialise the cost of communications infrastructure. Fine, that's a valid political position and it's got some merit to it, but in that case you guys should already have free healthcare and be advocating for a Universal Basic Income.
Schadenfreude, I think. We get so sick of Americans and British thinking they own the world and exporting their nasty foreign policy while we get all our TV, movie and book imports second hand, six months late and twice as expensive.
Plus it's that old pioneer attitude of 'you can darn well learn to put up with anything, young man, it'll build character'.
For the last 20 years we've been busy dismantling our welfare state and installing Thatcherite/Reaganite 'user pays' laissez faire, it hurt like hell, and if WE had to suffer for that free-market crap, you guys who INVENTED it can stand to apply it yourself.
But seriously, if you're using over 250GB a month, that's just greedy. And while *information* is not a zero-sum commodity, bandwidth over a shared cable *is*. Every gigabyte you use around is a gigabyte someone else can't.
It's just basic fairness. You should pay for what you use. Don't sweep use costs under the carpet of hidden externalities. The same principle underlies good economics, good ecology, and good neighbourliness.
Ted Stevens was bought by Big Cable, but he was right, as far as he went, and it won't help to mock him. The Internet *is* a series of data pipes with finite bandwidth, and that needs to be paid by someone. Should really be local user-owned broadband cooperatives rather than profit-driven companies, but even then it'll cost you a nonzero amount to move packets, and it should.
Here in the Antipodes we pay $80/month for 20G.
Uphill. Both ways. In the snow. And we like it.
http://www.telstraclear.co.nz/residential/inhome/internet/cable-broadband/plans.cfm
You can't download terabytes of pirated movies every day for free over someone else's lines? Oppression, I tell you. Oppression.
Why no Flashblock?
It's the only Firefox add-in I actually use all the time. Can't stand autorunning Flash ads (or even Youtube videos). Adblock is too aggressive for me, but Flashblock is just perfect. One click and I see an animation, otherwise my sanity is preserved.
"TOTAL TAX 11.50%"
What, only 11.50%? In New Zealand we've had a flat mandatory 12.5% Goods and Services Tax on all consumer transactions since 1989.
Introduced - and then increased - by the economic hard right wing, by the way.
(Who happened to be running a Labour government at the time. It's complicated. It was the 80s.)
http://en.wikipedia.org/wiki/Goods_and_Services_Tax_(New_Zealand)
For reminding me to remove 'Idle' from my sections list.
First time I've needed to do it. You might be interested to know for future reference. Consider this an exit interview.
Reasons: These book reviews, and that horrible green header.
Bye and good luck. And my sympathies for having to read the slush pile.
(Yeah, I know this is filed in Books, not Idle. But the last one was correctly filed and was just as bad. Don't make me delist Books too, okay?)
"How can you trust anything science says? It revises itself to be accurate, that's why. An unchanging theory constructed in year X is one that is unlikely to become more correct with the collection of new evidence in the years to come, except by some fluke of blind luck."
Right. But if a theory was correct *to begin with*, it doesn't actually need revising. The more correct and trustworthy a piece of scientific knowledge, the *less* it will change over time. So logically speaking, we should be more skeptical of new "discoveries" and more trusting of old knowledge. And the more and deeper "scientific revolutions" we encounter the more cynical it should make us of any claims to knowledge of any kind.
After a couple of such revolutions, we should feel deeply uneasy, uncertain, bewildered and emotionally devastated. We should feel traumatised, like a war or disaster survivor, because learning that people you respected were teaching you falsehoods and can give you no confidence that they are telling you the ultimate truth now *should* feel like a deep personal betrayal.
Feel familiar?
Robert Persig describes this phenomenon quite well in 'Zen and the Art of Motorcycle Maintenance'. The nature of science is to generate and disprove hypotheses, not produce 'truth'. There's an illusion that science is progressing and converging toward truth. But the more hypotheses we generate, the more contradictory alternatives we actually have in our minds, and the less able to respond to the world we become.
It's a little like Windows patches. If you keep having to retroactively fix things, you're doing it wrong.
"People interested in science enter research careers and have access from their universities or research organisations. People who are not scientists but are interested in science will use these resources for entertainment purposes, not to advance the body of knowledge."
Sheesh. With that attitude, do you seriously wonder *why* people in the street think scientists are 'elitists'? No scruffy unwashed masses allowed in OUR little club, no sir. They might be ENTERTAINED by knowledge! The horror! Knowledge must be locked up and kept secure lest it be defiled by grubby minds.
Thank God for Wikipedia. I've actually learned things by reading it. Stuff they never taught me in high school science.
Presumably to the manufacturers of hardware which contains binary-only drivers.
The idea is that it's a deliberately stress-testing distribution designed to be 100% Free and to cause any hardware which isn't Free to fail. If nobody complains that broken stuff is broken, it won't get fixed. And requiring binary drivers *is* breakage. As soon as the kernel updates, potentially wham! go your drivers if there's no source code.
"The reasonable man adapts himself to the conditions that surround him... The unreasonable man adapts surrounding conditions to himself... All progress depends on the unreasonable man." -- George Bernard Shaw.