"Your best friend, or your mom... preys on some base human voyeuristic instinct..."
You take that back about my mom!
s/base/natural. It's *normal* for humans to be social beings. Since when did 'has a social life' become an accusation? Where did the stereotype of the 'heroic lonely individual' come from? That's the dysfunctional syndrone, not the online chatter.
XENIX is a fascinating story and it seems that if things had worked out a little differently at Microsoft, we might have been running Unix on the desktop since the early 90s. Then Linux would have had a much easier job of challenging the status quo.
I wonder why that didn't happen? Was NT an attempt to leapfrog *both* DOS and XENIX in a single bound? That seems sensible, but iIn retrospect, other than creating lock-in dominance, did it work? NT's ACLs are nicer than Unix's, and Linux is still struggling to adopt capabilities in a standard way - but how did silly things like 'you can't replace an open file' get through? Was it all the fault of the Win32 compatibility box?
And since MS was starting from a clean slate with NT, and Windows was needing an object model, why the heck didn't they build objects into the OS kernel rather than hacking up such a weird kludge as COM? How did 'objects' get perverted into 'components' and why should we ever need two separate, incompatible, ways of saying roughly the same thing? Why did they use C++ to build NT even though C++ had no decent ABI?
I have all these questions about the history of modern computing that just don't make sense. It is sad that even in 2009, 40-year-old Unix remains 'the best we can do' in open systems. What the heck went wrong, and why?
Please don't tell me 'this current architectural mess is the best we can ever hope for'. That would be just too depressing.
Indeed. The Commodores ran a modified MS-BASIC - was it OUT 256, x which would display "Microsoft!" x times?
By the standards of the era, MS-BASIC wasn't the worst - but it was pretty crap. BBC BASIC was a revelation in comparison. Multiline IFs, REPEAT loops, proper named functions (IIRC).
But MS-BASIC grew up and became BASICA, BASICE, GW-BASIC, QuickBASIC, Visual BASIC, VB.NET. And it's still as broken as it was back then.
NZ-er here, but I miss the whole 80s UK chip scene. There was a lot of really innovative stuff like the Transputer which never made it to the 90s. At least the ARM survived as an embedded platform, but Intel pretty much ate a lot of companies who deserved a better chance.
No, IBM were just as big a bastard as Microsoft was. They were two robber barons who sort of cooperated then fell out with each other. IBM made a major misstep when they underestimated personal computing and platform-independent software - but not because they were altruistic.
Yes, they created an open 'Personal Computer' (tm) standard by accident (turning a generic industry-wide name into a trademark in the process - some of us haven't forgotten that), but they sure didn't go gently into the good open-standards night.
Remember the lawsuits against clone manufacturers? The 'BIOS wars'?
Then remember PS/2 in the late '80s? Micro Channel? How OS/2 was going to be much more strictly tied to proprietary PS/2 hardware, a proprietary expansion bus?
There were reasons why OS/2 didn't succeed and Windows did that weren't just 'poor unworldly IBM'. The OS/2 - PS/2 one-two punch was a desperate, naked power-grab to try to claw back proprietary vertical integration over the entire desktop right down to the hardware, and it alienated a generation of developers.
Windows 95 and NT were evil, but they were just *slightly* less evil than OS/2 in the early 90s; at least you could run it on slightly more hardware. Then having destroyed OS/2, Windows went the same lock-in way and meanwhile BillG was killing off all the software competition, and until Linux and OSX and Firefox he had a clear run at it - and we're still paying the price for our lack of vision.
And let's not consider Novell completely the nice guys either. They once owned the networking market, lost their iron grip to NT, and then made a last-ditch play for the desktop with Wordperfect / Paradox / Quattro. Remember Novell Office? Remember NEST? They wanted 'Novell Everywhere'.
And then there's Sun, and Oracle... same story, they just didn't win the lock-in game, but not for want of trying to be evil enough.
Frankly I'm sick of our computing destiny being decided by people who just don't understand the word 'interoperate'.
"Microsoft brought desktop computing to the home user."
In what sense?
Some of us are old enough to remember when the words 'Personal Computer', 'DOS', 'BASIC' and even 'Windows' were not names of Microsoft products.
There was a whole ecosystem of small machines in which Microsoft was one of a number of players. Microsoft wasn't first to the 'desktop' (defined as a windowed operating system), they weren't first to the microcomputer, and their products weren't the best of the bunch.
Personal computing was already affordable before BillG (and IBM) got involved. He just rode the wave.
They were just the ones who achieved 'market leadership' by crowding out better designed products, and reaped the financial rewards of 'winning' that game, and those of us who sadly remember, say, the Amiga and the Archimedes, or even OpenDoc, find that hard to forgive.
Desktop computing might never recover from the blow it was dealt by atrocities like MS-DOS's mixing Unix paths and CP/M's drive letters, say, or the Windows Registry. Or COM. Yes, there could have been worse technologies that got adopted, but there could have been far, far, better. We could have had a functioning, sane, system.
I dunno about the Mother Country but here in Her Majesty's Antipodean Dominions the 'pint' has officially been precisely 600 millilitres since around the 1960s.
I'm wondering what might happen if, as a transitional stage, we took Javascript in the Web browser as a given and then had a rethought protocol over or replacing HTTP which distributed persistent/cached Javascript objects rather than pages. Like AJAX, but all the time, without even the overlaying concept of 'page'.
I suppose that still wouldn't work without some way of creating a DOM; but if we had a standard form of representing XML in JSON (like S-XML does for S-Expressions)...
I'd love to have some kind of standardised publish/subscribe push protocol over TCP, akin to HTTP but purely data-based, and automatically taking care of issues like version tracking, querying and resending if an update is missed, and sending source-code objects which can do computation (preferably pure-functional for safety, but untyped - if you have the source code/parse tree you can do your own type checking, and more).
There are so many applications today - email lists, instant messaging, message queueing, Facebook, Twitter - which all sort of implement roughly generalised 'publish/subscribe' but do so at the application level, incompatibly, that it seems like a simple standard store/forward data/compute block protocol would simplify things enormously.
"What does the OS, *any* OS, have to do with playing back menuing items from a DVD?"
Because the OS/application divide is not a real thing.
The user doesn't care about the difference, the computer doesn't care about the difference, and therefore ultimately, there is no actual difference.
It's all just software. It's shipped together. Therefore, it's not an 'operating system plus applications', it's just a *system*. It's supposed to do stuff the user wants done; that's what it exists for.
When your system can't play DVDs, yet it's inhabiting the desktop niche on a piece of hardware that has a DVD drive in a world where DVDs are a popular media, that's a problem which needs to be fixed.
If the solution is 'pay extra and buy an add-on', that's only a solution to a third-party *company*'s problem of 'how do I get people to give me money?' , not to the user's original problem of 'how do I play DVDs?'
A large part of the problem of computing today, IMO, is that architects and designers put so much effort into maintaining an inherited abstract mental *infrastructure* - categories like 'OS' and 'application', for instance, which are ideas that made sense in the days of batch timesharing systems where a program was literally a 'job' , but have almost zero relevance to the desktop, let alone the Internet.
But this infrastructure is imaginary. These concepts aren't real. They're ways we've invented of thinking about the problem of computing, but they're not computing itself.
The user doesn't care. The user is *right* not to care. And software engineers need to start caring less about maintaining the 'correct' mental infrastructure they've been taught and asking 'how can we think about this whole thing from a better viewpoint?'
"And, IMO, it's insanely stupid to think that if the U.S. were just more popular with the worlds' countries, those countries would like the U.S. better."
Actually, um, that's not only easy to believe but a tautology - if the US were more popular, by definition it would be more well-liked.....
I'm trying to parse what you really mean and I can't figure it. Do you mean 'it's stupid to believe that if the US were less aggressive on the world stage, it would be more popular?' Or 'it's stupid to believe that if the US were more popular with the majority of countries, that it would also be more popular with the hard-core minority'?
"If someone creates a model that predicts where the oil futures will go based on past performance in similar circumstances, you think that person should be required by law to reveal his model?"
Yes.
"Wouldn't that make it nearly impossible to make money in the stock market"
Yes. That would mean the stock market would be an *efficient market* and would do the job it's supposed to do: direct investment to sources of real value (long-term, planet-wide improvement in social conditions) rather than short-term Prisoner's Dilemma-style scams and ripoffs.
"and the only buying and selling would be people looking to begin investing in blue chips for the long haul"
YES. This would redirect the attention of the economy to solving the vitally important long-term problems of the world.
"thus reducing investment in companies"
No, only reducing investment in short-term rip-offs.
"thus reducing R&D, thus reducing innovation?"
No, it would expose the true sources of R&D investment, which remain what they've always actually been: groups like DARPA with funds and a long-term vision and commitment.
If your long-term R&D funding model is driven solely by expectation of short-term returns and REQUIRES obfuscation and deception between investors seeking self-interest rather than honest and transparent public dissemination of scientific knowledge, your society is already screwed no matter how you try to cut the cake.
Long-term, planetary scale R&D requires long-term, planetary scale wisdom and cooperation. There's no way around this. You can't boost the system by making individuals fight each other like starving rats in a fog of ignorance and think somehow that will generate good vibes of positivity and constructive progress. It won't.
"All of which means you can make more money without being detected - and you haven't attacked the software itself, you haven't changed how the code works, you have stayed within the boundaries that the software creates. All because you knew *precisely* how the code works."
Maybe I'm just naive... but wasn't having *precisely specified rules* for financial transactions, published and known by all players, once upon a time considered the bedrock of what finance and banking were all about? And why capitalism was the most perfect economic system known to mankind?
Yeah, I know, it's never actually been like that, that was probably way back before the Dutch invented the stock market or the Egyptians started building pyramids or the first hunter-gatherers traded some dodgy furs for some suspicious berries... but still.
Sorry, I don't have much sympathy when the guys playing liars' poker have their hands shown to the world. Something as critical as banking shouldn't be the preserve of proprietary commercial secrecy and deception. It should be a publically operated service, open and transparent and trustworthy and boring and above all manipulation, and that goes for all critical planetary infrastructure.
Oh say can you see By the nebula's light What so proudly we hailed At the comet's last gleaming...... o'er the land of the flneep And the home of the g'znarbilywarblave!
"Wikipedia cannot be trusted. For anything. Ever. There is no way whatsoever of knowing who controls the flow of information, or what their intent is, on any page. Wikipedia and its admins have no interest in the truth; only in their ability to control it."
Except that there are still those pesky citations which link to off-site resources. Kinda hard for even the omnipotent, evil Jimmy Wales to fake all those, hmm?
Yes, Wikipedia isn't completely free-range. Yes, it's done at least negative censorship. But frankly, the reason that censorship exists in the first place is because the genre is beseiged by trolls. Do you go to Everything2.com looking for anything sensible? What about consumerpedia.org? That was promising once, a few years ago... then the trolls invaded and decided it was all 'GFDL Corpus' this and that and how dare anyone remove any post, ever.
Don't trust Wikipedia to have *all* the information you want, sure. Don't trust anything which doesn't have citations. But... don't trust any other media outlet, either. And one person's 'information' is another person's 'mindless moronic garbage hazardous to life and sanity', isn't it? See also: Ufology, 9/11 Truth Movement.
Slashdot Mirror
"Your best friend, or your mom... preys on some base human voyeuristic instinct..."
You take that back about my mom!
s/base/natural. It's *normal* for humans to be social beings. Since when did 'has a social life' become an accusation? Where did the stereotype of the 'heroic lonely individual' come from? That's the dysfunctional syndrone, not the online chatter.
Correction: it was WAIT 6502, x
http://www.eeggs.com/items/12571.html
http://web.archive.org/web/20011211233332/www.rjh.org.uk/altair/4k/index2.html
Wow, floating point and arrays in 4K. That's actually quite impressive compared to Tiny Basic.
I still wish that the first environment for the Altair had been Forth. The Jupiter Ace tried, bless its silicon soul.
XENIX is a fascinating story and it seems that if things had worked out a little differently at Microsoft, we might have been running Unix on the desktop since the early 90s. Then Linux would have had a much easier job of challenging the status quo.
I wonder why that didn't happen? Was NT an attempt to leapfrog *both* DOS and XENIX in a single bound? That seems sensible, but iIn retrospect, other than creating lock-in dominance, did it work? NT's ACLs are nicer than Unix's, and Linux is still struggling to adopt capabilities in a standard way - but how did silly things like 'you can't replace an open file' get through? Was it all the fault of the Win32 compatibility box?
And since MS was starting from a clean slate with NT, and Windows was needing an object model, why the heck didn't they build objects into the OS kernel rather than hacking up such a weird kludge as COM? How did 'objects' get perverted into 'components' and why should we ever need two separate, incompatible, ways of saying roughly the same thing? Why did they use C++ to build NT even though C++ had no decent ABI?
I have all these questions about the history of modern computing that just don't make sense. It is sad that even in 2009, 40-year-old Unix remains 'the best we can do' in open systems. What the heck went wrong, and why?
Please don't tell me 'this current architectural mess is the best we can ever hope for'. That would be just too depressing.
"Apple was for some schools and a few geeks with no lives and way too much money."
Apple did do a lot of educational promotion, yes. But I'm guessing you don't remember Choplifter?
Indeed. The Commodores ran a modified MS-BASIC - was it OUT 256, x which would display "Microsoft!" x times?
By the standards of the era, MS-BASIC wasn't the worst - but it was pretty crap. BBC BASIC was a revelation in comparison. Multiline IFs, REPEAT loops, proper named functions (IIRC).
But MS-BASIC grew up and became BASICA, BASICE, GW-BASIC, QuickBASIC, Visual BASIC, VB.NET. And it's still as broken as it was back then.
BBC Represent!
NZ-er here, but I miss the whole 80s UK chip scene. There was a lot of really innovative stuff like the Transputer which never made it to the 90s. At least the ARM survived as an embedded platform, but Intel pretty much ate a lot of companies who deserved a better chance.
Heck yeah!
I went from 8K Commodore CBM, to 32K BBC Model B (built-in synth and triangle draw primitives!) to Atari ST to... PC.
Huge step back.
If Excel == the economic crash
and Powerpoint == Columbia
what horror will Word visit on us?
"You bastards! You finally did it! You auto-numbered the bullet points!"
"IBM, not being very worldly either"
*cough*
No, IBM were just as big a bastard as Microsoft was. They were two robber barons who sort of cooperated then fell out with each other. IBM made a major misstep when they underestimated personal computing and platform-independent software - but not because they were altruistic.
Yes, they created an open 'Personal Computer' (tm) standard by accident (turning a generic industry-wide name into a trademark in the process - some of us haven't forgotten that), but they sure didn't go gently into the good open-standards night.
Remember the lawsuits against clone manufacturers? The 'BIOS wars'?
Then remember PS/2 in the late '80s? Micro Channel? How OS/2 was going to be much more strictly tied to proprietary PS/2 hardware, a proprietary expansion bus?
There were reasons why OS/2 didn't succeed and Windows did that weren't just 'poor unworldly IBM'. The OS/2 - PS/2 one-two punch was a desperate, naked power-grab to try to claw back proprietary vertical integration over the entire desktop right down to the hardware, and it alienated a generation of developers.
Windows 95 and NT were evil, but they were just *slightly* less evil than OS/2 in the early 90s; at least you could run it on slightly more hardware. Then having destroyed OS/2, Windows went the same lock-in way and meanwhile BillG was killing off all the software competition, and until Linux and OSX and Firefox he had a clear run at it - and we're still paying the price for our lack of vision.
And let's not consider Novell completely the nice guys either. They once owned the networking market, lost their iron grip to NT, and then made a last-ditch play for the desktop with Wordperfect / Paradox / Quattro. Remember Novell Office? Remember NEST? They wanted 'Novell Everywhere'.
And then there's Sun, and Oracle... same story, they just didn't win the lock-in game, but not for want of trying to be evil enough.
Frankly I'm sick of our computing destiny being decided by people who just don't understand the word 'interoperate'.
"Microsoft brought desktop computing to the home user."
In what sense?
Some of us are old enough to remember when the words 'Personal Computer', 'DOS', 'BASIC' and even 'Windows' were not names of Microsoft products.
There was a whole ecosystem of small machines in which Microsoft was one of a number of players. Microsoft wasn't first to the 'desktop' (defined as a windowed operating system), they weren't first to the microcomputer, and their products weren't the best of the bunch.
Personal computing was already affordable before BillG (and IBM) got involved. He just rode the wave.
They were just the ones who achieved 'market leadership' by crowding out better designed products, and reaped the financial rewards of 'winning' that game, and those of us who sadly remember, say, the Amiga and the Archimedes, or even OpenDoc, find that hard to forgive.
Desktop computing might never recover from the blow it was dealt by atrocities like MS-DOS's mixing Unix paths and CP/M's drive letters, say, or the Windows Registry. Or COM. Yes, there could have been worse technologies that got adopted, but there could have been far, far, better. We could have had a functioning, sane, system.
But we didn't. We got Windows.
"But what scientists use for their daily work has nothing to do with what common people use in their daily lives."
Except for us here in the rest of the world, who've been using Centigrade temperatures since before I was born.
Mid 20s is nice and warm. 30 is hot. 0 is precisely freezing. Right now it's a chilly 4 degrees at Christchurch Airport.
Works just fine for us.
I dunno about the Mother Country but here in Her Majesty's Antipodean Dominions the 'pint' has officially been precisely 600 millilitres since around the 1960s.
Metric beer: the drink of the future, today!
Excellent point.
I'm wondering what might happen if, as a transitional stage, we took Javascript in the Web browser as a given and then had a rethought protocol over or replacing HTTP which distributed persistent/cached Javascript objects rather than pages. Like AJAX, but all the time, without even the overlaying concept of 'page'.
I suppose that still wouldn't work without some way of creating a DOM; but if we had a standard form of representing XML in JSON (like S-XML does for S-Expressions)...
I'd love to have some kind of standardised publish/subscribe push protocol over TCP, akin to HTTP but purely data-based, and automatically taking care of issues like version tracking, querying and resending if an update is missed, and sending source-code objects which can do computation (preferably pure-functional for safety, but untyped - if you have the source code/parse tree you can do your own type checking, and more).
There are so many applications today - email lists, instant messaging, message queueing, Facebook, Twitter - which all sort of implement roughly generalised 'publish/subscribe' but do so at the application level, incompatibly, that it seems like a simple standard store/forward data/compute block protocol would simplify things enormously.
"Brings to mind that quote "People who don't understand Unix are condemned to repeat it, poorly.""
The same claim is often made for Lisp, but the Lisp Machines didn't run Unix.
Do Lisp and Unix spin eternally in a sort of co-recursive loop, like Vi and Emacs?
Also I presume they had all this technology inherited from NeXT, which did not use X as its window manager.
So THAT'S where the name "Dr. Strangelove" came from! *finally clicks*
"What does the OS, *any* OS, have to do with playing back menuing items from a DVD?"
Because the OS/application divide is not a real thing.
The user doesn't care about the difference, the computer doesn't care about the difference, and therefore ultimately, there is no actual difference.
It's all just software. It's shipped together. Therefore, it's not an 'operating system plus applications', it's just a *system*. It's supposed to do stuff the user wants done; that's what it exists for.
When your system can't play DVDs, yet it's inhabiting the desktop niche on a piece of hardware that has a DVD drive in a world where DVDs are a popular media, that's a problem which needs to be fixed.
If the solution is 'pay extra and buy an add-on', that's only a solution to a third-party *company*'s problem of 'how do I get people to give me money?' , not to the user's original problem of 'how do I play DVDs?'
A large part of the problem of computing today, IMO, is that architects and designers put so much effort into maintaining an inherited abstract mental *infrastructure* - categories like 'OS' and 'application', for instance, which are ideas that made sense in the days of batch timesharing systems where a program was literally a 'job' , but have almost zero relevance to the desktop, let alone the Internet.
But this infrastructure is imaginary. These concepts aren't real. They're ways we've invented of thinking about the problem of computing, but they're not computing itself.
The user doesn't care. The user is *right* not to care. And software engineers need to start caring less about maintaining the 'correct' mental infrastructure they've been taught and asking 'how can we think about this whole thing from a better viewpoint?'
"And, IMO, it's insanely stupid to think that if the U.S. were just more popular with the worlds' countries, those countries would like the U.S. better."
Actually, um, that's not only easy to believe but a tautology - if the US were more popular, by definition it would be more well-liked.....
I'm trying to parse what you really mean and I can't figure it. Do you mean 'it's stupid to believe that if the US were less aggressive on the world stage, it would be more popular?' Or 'it's stupid to believe that if the US were more popular with the majority of countries, that it would also be more popular with the hard-core minority'?
Those are very different claims though.
Now there's a whole new genre of romance novels begging to be written.
"If someone creates a model that predicts where the oil futures will go based on past performance in similar circumstances, you think that person should be required by law to reveal his model?"
Yes.
"Wouldn't that make it nearly impossible to make money in the stock market"
Yes. That would mean the stock market would be an *efficient market* and would do the job it's supposed to do: direct investment to sources of real value (long-term, planet-wide improvement in social conditions) rather than short-term Prisoner's Dilemma-style scams and ripoffs.
"and the only buying and selling would be people looking to begin investing in blue chips for the long haul"
YES. This would redirect the attention of the economy to solving the vitally important long-term problems of the world.
"thus reducing investment in companies"
No, only reducing investment in short-term rip-offs.
"thus reducing R&D, thus reducing innovation?"
No, it would expose the true sources of R&D investment, which remain what they've always actually been: groups like DARPA with funds and a long-term vision and commitment.
If your long-term R&D funding model is driven solely by expectation of short-term returns and REQUIRES obfuscation and deception between investors seeking self-interest rather than honest and transparent public dissemination of scientific knowledge, your society is already screwed no matter how you try to cut the cake.
Long-term, planetary scale R&D requires long-term, planetary scale wisdom and cooperation. There's no way around this. You can't boost the system by making individuals fight each other like starving rats in a fog of ignorance and think somehow that will generate good vibes of positivity and constructive progress. It won't.
"All of which means you can make more money without being detected - and you haven't attacked the software itself, you haven't changed how the code works, you have stayed within the boundaries that the software creates. All because you knew *precisely* how the code works."
Maybe I'm just naive... but wasn't having *precisely specified rules* for financial transactions, published and known by all players, once upon a time considered the bedrock of what finance and banking were all about? And why capitalism was the most perfect economic system known to mankind?
Yeah, I know, it's never actually been like that, that was probably way back before the Dutch invented the stock market or the Egyptians started building pyramids or the first hunter-gatherers traded some dodgy furs for some suspicious berries... but still.
Sorry, I don't have much sympathy when the guys playing liars' poker have their hands shown to the world. Something as critical as banking shouldn't be the preserve of proprietary commercial secrecy and deception. It should be a publically operated service, open and transparent and trustworthy and boring and above all manipulation, and that goes for all critical planetary infrastructure.
Oh say can you see ... o'er the land of the flneep
By the nebula's light
What so proudly we hailed
At the comet's last gleaming...
And the home of the g'znarbilywarblave!
"Melancholy Elephants". http://www.spiderrobinson.com/melancholyelephants.html
"Wikipedia cannot be trusted. For anything. Ever. There is no way whatsoever of knowing who controls the flow of information, or what their intent is, on any page. Wikipedia and its admins have no interest in the truth; only in their ability to control it."
Except that there are still those pesky citations which link to off-site resources. Kinda hard for even the omnipotent, evil Jimmy Wales to fake all those, hmm?
Yes, Wikipedia isn't completely free-range. Yes, it's done at least negative censorship. But frankly, the reason that censorship exists in the first place is because the genre is beseiged by trolls. Do you go to Everything2.com looking for anything sensible? What about consumerpedia.org? That was promising once, a few years ago... then the trolls invaded and decided it was all 'GFDL Corpus' this and that and how dare anyone remove any post, ever.
Don't trust Wikipedia to have *all* the information you want, sure. Don't trust anything which doesn't have citations. But... don't trust any other media outlet, either. And one person's 'information' is another person's 'mindless moronic garbage hazardous to life and sanity', isn't it? See also: Ufology, 9/11 Truth Movement.