Neal Stephenson's "The Great Simoleon Caper"
on
The Diamond Age
·
· Score: 1
There's one work by Neal Stephenson that almost never gets mentioned here, yet it's absolutely excellent and is perched right in the middle of Slashdot territory.
It's "The Great Simoleon Caper", a very funny yet "with it" story concerning digital currency, conspiracies and Jolt Cola.
If there is no skills shortage where you are then business is lucky and you're unlucky.
In Europe (and here in the UK in particular), there is an immense skills shortage in the relevant areas. It's not good enough to know just one or two things though, because the current skills shortage is caused by convergence of many areas, and your skills have to match. If you have extensive knowledge in Internet networking, Unixes of various kinds (especially Solaris), server hardware and software administration, programming in C/C++, shell and Perl scripting, and experience with SQL databases, then you will find everyone scrambling to pay you more to work for them.
And the shortage is so bad that everyone is poaching from everyone else, but that again drives up the contract rates. (I'm not complaining.)
So, stop moaning, go freelance, and move where the work is.
The situation you describe is precisely what has promoted the explosive growth of the contractor market. It allows you to practice your skills while negociating as a peer with those that pay you, instead of working as a slave.
Now it's up to you. Either go freelance or continue to moan about your plight, the cluelessness of PHBs, and the iniquities of the situation.
Oh dear, that line of speculation is worrying. If Amiga Inc had to change their plans because of Transmeta-related "assumptions that have turned out to be incorrect" then that could also mean that Transmeta has hit an iceberg on its maiden voyage.
I wish Transmeta would say *something* -- *anything*. I think they may be too quiet for their own good, because silence breeds hype which may end up being very difficult to live up to. It would be awful if they ended up tarred with the Amiga/Gateway vapourware brush just because they allowed total hype runaway instead of making their intentions clear.
It's odd, but I hate CPUs in slot-based cartidges vehemently, and will not buy one. I rationalize this to myself occasionally, but I don't find the reasons I give particularly convincing --- I expect that I'm just emotionally tied to the idea that a CPU is meant to live entirely on a single chip (despite the lessons of history), and "I don't want no crappy chunk of plastic". Yeah, I know, it's illogical.
Still, if AMD or Intel want my custom (and I can't be the only one so oddly predisposed), they'd better crank out ZIF-socketed CPUs. I'm currently on multiple 500MHz socket-370 Celerons, so it was great to hear that AMD will be offering me an upgrade path in the form of socket-423 Athlons. Hooray.
Nope, I'm not replying to that article because the radio amateur world does not operate to the same rules as ours here. It's pointless making an informed argument (except for self-satisfaction perhaps) because positions there are so heavily entrenched that reason does not prevail, *ever*.
Fortunately, there is a way out. Progressive radio amateurs don't talk, they *do* --- in other words, they build that better mousetrap (radio) and if it's good then others adopt it and the good idea prevails. So much so that experimentation always precedes the official law-making in the hobby by a decade, typically. It's kind of funny, knowing that the leading-edge experimenters that are quietly pushing *beyond* the legal boundaries set by today's FCC's rulings are precisely the same people that are creating the underpinnings for the legal framework of tomorrow.
I suppose Robin Hood would have said that it's a fine line between outlaw and criminal.:-)
Actually, I think you're 100% wrong (:-), in that I believe that opening software up is dramatically better for the product in all cases except very specific niche ones, which StarOffice is not.
Only time will tell though. It'll be fun to watch how it develops.
The article immediately following yours on technocrat.net was decidedly ill-informed though.
The guy was proposing a protectionist stance against SS use in narrowband segments, completely misunderstanding that power density drops as the SS bandwidth gets wider, and that therefore the best way of protecting narrowband operation would be to insist that all SS transmissions should fill the entire band or else be limited to a given power density (ie. per Hz).
I couldn't help thinking that this is not unlike traditional software companies totally misunderstanding how the free software/open source model is a *benefit*, not a problem.:-)
Sun gains nothing useful from this partial closure of the source, despite what they say. They know that the time their developers spend on improving this product in-house is time and money lost. They must know that they would just gain, gain, gain, if they opened StarOffice up completely.
Why then are they doing as they are? As anyone that works for a *big* company will tell you, the answer is simple: internal politics and inertia, development divisions holding grimly on to old methods of working, marketing divisions holding on to old methods of licensing, and so on, that's all.
There is an analogy that can be made here, referring to big brass in Sun management: big dinosaurs move and think slowly. The question is, will they see the light soon enough, change, and survive?
Transputer boards were available for many different types of personal computer, including IBM-PC and compatibles, Macs, Amiga, and Atari. Most transputers ended up embedded in things like laser printers though, allegedly, and quite a few went into fancy research machines, hundreds and thousands at a time.
I've got one in my junk box somewhere. I should embed it in acrylic and turn it into a topical paperweight.:-)
Real-time OSs should be integrated on-chip in microcontrollers so that the cost goes down to rock bottom and so that the integration with on-chip hardware is very tight. A new, extensible microcontroller OS would be welcome if it achieved this, but otherwise it's "yet another OS".
Transputers had at least 3 features that made them so far ahead of their time that they died through lack of applications. Or maybe it was just because the Brits and the French are lousy at bringing things to market. If only Intel had bought out Inmos...
Anyway, getting back to the features:
- A process scheduler implemented in hardware (an outer loop outside the usual inner instruction fetch and execute loop), which allowed transputers to implement concurrency with very fine granularity because the context switch time was exceptionally low. (And the process scheduler was directly driven by I/O events at the transputer links, below.)
- Four high-speed serial "Inmos links" on-chip through which the transputer could be linked directly to other transputers and to other peripherals without further glue logic, so that building multiprocessors was very inexpensive and scaled linearly. Furthermore, these links ran not only extremely quickly (for their time) despite being serial, but more importantly they ran under DMA power all simultaneously and at the same time as the processor was doing its own thing independently.
- The above two features made the transputer exceptional for multiprocessing, but I think its instruction set was also far, far ahead of its time: not only ultra-RISC, but highly extensible too. For example, numeric literals in instructions were only as long as needed, because an extension bit would (if present) indicate that more bits were to follow if needed. This made code *extremely* tight. The scheme also allowed extensions to the instruction set to be made in a fully backward-compatible manner.
The transputer was ultra-cool, and the world hasn't seen anything like it since. No doubt somebody will reinvent this approach some day, but probably in a US or Japanese lab, as usual, and they'll take the credit for exceptional design ideas made in an earlier age. Sigh.
The subject line doesn't match the body of your reply: the subject suggests that variable-length IP addresses are already here, but then you clarify in the body that it's only domain names that have variable length.
Well, that isn't good enough. We have not hit any limits with domain names precisely because they are totally extensible overall and also in each component part, whereas the limits on IPv4 addresses are a daily headache right now, and the limits on IPv6 addressing will be a daily headache for us in the future, sooner than we think.
The suggestion that extensible IP addresses should be avoided because of the difficulty it might cause router manufacturers is so short-sighted that it's just a joke. If the existing crop of manufacturers can't come up with an efficient routing algorithm for extensible addresses then they should die and leave the profits to newer outfits that are willing to engage a few brain cells to solve the problem.
Seven computers will be sufficient to satisfy the world's entire computing requirement. And 640 Kbytes is more memory than the biggest PC user will ever need.
If there's one thing that computer people should have learned by now, it's that limits have a habit of being reached sooner than expected. Len Kleinrock is right to question the lack of extensibility in IPv6 addressing. It *will* bite us one day, and sooner than we think.
Maybe the more visionary kernel coders should get together and develop a stack for Linux or BSD based on one of the designs for extensible IP addressing. It's still early days for IPv6, and there is still ample opportunity for an upstart alternative design and implementation to make it big and become a de facto standard. And that would be good for us in the long run, which may be a lot less long than we think.
Having worked with perhaps a dozen different Unixes over the years, I think I can make a fairly reasonable assessment as to the merits of SCO as a Unix.
In my opinion, SCO is the biggest pile of rubbish on the Unix scene. I remember it as unstable, non-standard, lacking in features, sysadmin-unfriendly, user-hostile, expensive, poorly supported, and dead slow. The only things that it had going for it was that there were a lot of applications listed in its catalogue, and that after all it is a Unix. Personally I wouldn't touch it again with a barge pole though.
As for SCO's corporate stance re Linux, it's laughable so ignore it. Linux and the other free Unixes are the death of SCO, and if they're going to babble silly things as they die then good riddance.
> Linux is NOT a threat to Microsoft (or any user-friendly OS maker) at the *consumer level* until it is re-designed from the ground up to be USER-FRIENDLY.
User-friendliness has nothing to do with it. Windows is extraordinarily user-hostile, overly complex and appallingly obscure, but it benefits from one key fact: it comes preloaded on PCs, as a result of which prospective users don't have to go through the utterly user-unfriendly Windows installation procedure. If you put Windows and Linux head-to-head in front of a computer-unaware beginner with the goal of installing exactly the same feature set, Windows would be so far behind that it wouldn't even figure on the user-friendly map.
Don't flatter Microsoft with a "user-friendly" label. It's not. Their horrid product is just preloaded, and Linux isn't.
I hope this event gives paranoia a new lease of life, before it becomes accepted practice for hardware manufacturers to supply binary-only drivers for Linux.
It's bad enough to run binary-only applications, but to some extent their impact can be controlled. In contrast, once you add an opaque binary-only module into the kernel then all bets are off.
Although the article raises some interesting points, it paints with too broad a brush when saying that computer users are becoming disempowered. It's yet another case of statistical generalization, which may delight journalists and politicians but is always very annoying to those that don't follow others like sheep nor benefit from it. Some users are disempowered, yes, namely those that are not able to assess for themselves whether relying on a service like Hotmail or a company like Microsoft is a good idea, and those who are not able to make the right evaluation and move to other pastures. But does it disempower you, as Slashdot reader? Almost universally, no, because for the most part people who use this forum are competent enough to know when to leave a sinking ship or not to expose themselves to the hazard in the first place. We're not the Borg. We're individuals, and just because statistically something appears to be happening to some computer users doesn't mean that it is happening to computer users in general. There always will be people who are challanged in one or more areas and who as a result are prone to some group-specific ailment, but you can't extrapolate from that to the universe of people when that universe is as diverse as that of computer users.
That's not a function of a normal MAC-level switch.
There are boxes that perform this kind of function (we use 'em), like the Cisco Local Director (works like a modified bridge) and various layer 4 switches too. That's why I was asking. What is Slashdot using?
Yes, I hear the difference too, and I mean at the highest normal bitrates (I haven't tried anything beyond 256K though). I wish somebody would do a proper scientific blind A-B test and publish the results to debunk the myth of no loss of fidelity.
I wouldn't say it sounds like crap though, just not hifi, and the high frequency artifacts are probably more irritating then FM radio.
'Pity that Shorten gives you only a 60% inprovement. That's so poor that MP3 isn't threatened at all.
Slashdot Mirror
There's one work by Neal Stephenson that almost never gets mentioned here, yet it's absolutely excellent and is perched right in the middle of Slashdot territory.
/ simoleon.html
It's "The Great Simoleon Caper", a very funny yet "with it" story concerning digital currency, conspiracies and Jolt Cola.
Here's a rare link:
http://kuoi.asui.uidaho.edu/~kamikaze/documents
Somebody please mirror it before it gets Slashdotted or is otherwise lost!
If there is no skills shortage where you are then business is lucky and you're unlucky.
In Europe (and here in the UK in particular), there is an immense skills shortage in the relevant areas. It's not good enough to know just one or two things though, because the current skills shortage is caused by convergence of many areas, and your skills have to match. If you have extensive knowledge in Internet networking, Unixes of various kinds (especially Solaris), server hardware and software administration, programming in C/C++, shell and Perl scripting, and experience with SQL databases, then you will find everyone scrambling to pay you more to work for them.
And the shortage is so bad that everyone is poaching from everyone else, but that again drives up the contract rates. (I'm not complaining.)
So, stop moaning, go freelance, and move where the work is.
The situation you describe is precisely what has promoted the explosive growth of the contractor market. It allows you to practice your skills while negociating as a peer with those that pay you, instead of working as a slave.
Now it's up to you. Either go freelance or continue to moan about your plight, the cluelessness of PHBs, and the iniquities of the situation.
Oh dear, that line of speculation is worrying. If Amiga Inc had to change their plans because of Transmeta-related "assumptions that have turned out to be incorrect" then that could also mean that Transmeta has hit an iceberg on its maiden voyage.
I wish Transmeta would say *something* -- *anything*. I think they may be too quiet for their own good, because silence breeds hype which may end up being very difficult to live up to. It would be awful if they ended up tarred with the Amiga/Gateway vapourware brush just because they allowed total hype runaway instead of making their intentions clear.
It's odd, but I hate CPUs in slot-based cartidges vehemently, and will not buy one. I rationalize this to myself occasionally, but I don't find the reasons I give particularly convincing --- I expect that I'm just emotionally tied to the idea that a CPU is meant to live entirely on a single chip (despite the lessons of history), and "I don't want no crappy chunk of plastic". Yeah, I know, it's illogical.
Still, if AMD or Intel want my custom (and I can't be the only one so oddly predisposed), they'd better crank out ZIF-socketed CPUs. I'm currently on multiple 500MHz socket-370 Celerons, so it was great to hear that AMD will be offering me an upgrade path in the form of socket-423 Athlons. Hooray.
Re your sig, death is an engineering problem, and so is sleep. I dare say that nanotech will address both.
The words that I meant to put in Robin Hood's mouth were of course that it's a fine line between hero and criminal. :-)
Nope, I'm not replying to that article because the radio amateur world does not operate to the same rules as ours here. It's pointless making an informed argument (except for self-satisfaction perhaps) because positions there are so heavily entrenched that reason does not prevail, *ever*.
:-)
Fortunately, there is a way out. Progressive radio amateurs don't talk, they *do* --- in other words, they build that better mousetrap (radio) and if it's good then others adopt it and the good idea prevails. So much so that experimentation always precedes the official law-making in the hobby by a decade, typically. It's kind of funny, knowing that the leading-edge experimenters that are quietly pushing *beyond* the legal boundaries set by today's FCC's rulings are precisely the same people that are creating the underpinnings for the legal framework of tomorrow.
I suppose Robin Hood would have said that it's a fine line between outlaw and criminal.
Actually, I think you're 100% wrong (:-), in that I believe that opening software up is dramatically better for the product in all cases except very specific niche ones, which StarOffice is not.
Only time will tell though. It'll be fun to watch how it develops.
The article immediately following yours on technocrat.net was decidedly ill-informed though.
:-)
The guy was proposing a protectionist stance against SS use in narrowband segments, completely misunderstanding that power density drops as the SS bandwidth gets wider, and that therefore the best way of protecting narrowband operation would be to insist that all SS transmissions should fill the entire band or else be limited to a given power density (ie. per Hz).
I couldn't help thinking that this is not unlike traditional software companies totally misunderstanding how the free software/open source model is a *benefit*, not a problem.
Sun gains nothing useful from this partial closure of the source, despite what they say. They know that the time their developers spend on improving this product in-house is time and money lost. They must know that they would just gain, gain, gain, if they opened StarOffice up completely.
Why then are they doing as they are? As anyone that works for a *big* company will tell you, the answer is simple: internal politics and inertia, development divisions holding grimly on to old methods of working, marketing divisions holding on to old methods of licensing, and so on, that's all.
There is an analogy that can be made here, referring to big brass in Sun management: big dinosaurs move and think slowly. The question is, will they see the light soon enough, change, and survive?
Re your last paragraph, that was Richard Stallman, not Linus. RMS is the free software visionary, Linus is the pragmatic and supreme expert engineer.
We're damn lucky to have such a pair, and indeed others like Alan Cox that stand on the very same pedestal in my book.
Transputer boards were available for many different types of personal computer, including IBM-PC and compatibles, Macs, Amiga, and Atari. Most transputers ended up embedded in things like laser printers though, allegedly, and quite a few went into fancy research machines, hundreds and thousands at a time.
:-)
I've got one in my junk box somewhere. I should embed it in acrylic and turn it into a topical paperweight.
Real-time OSs should be integrated on-chip in microcontrollers so that the cost goes down to rock bottom and so that the integration with on-chip hardware is very tight. A new, extensible microcontroller OS would be welcome if it achieved this, but otherwise it's "yet another OS".
Transputers had at least 3 features that made them so far ahead of their time that they died through lack of applications. Or maybe it was just because the Brits and the French are lousy at bringing things to market. If only Intel had bought out Inmos ...
Anyway, getting back to the features:
- A process scheduler implemented in hardware (an outer loop outside the usual inner instruction fetch and execute loop), which allowed transputers to implement concurrency with very fine granularity because the context switch time was exceptionally low. (And the process scheduler was directly driven by I/O events at the transputer links, below.)
- Four high-speed serial "Inmos links" on-chip through which the transputer could be linked directly to other transputers and to other peripherals without further glue logic, so that building multiprocessors was very inexpensive and scaled linearly. Furthermore, these links ran not only extremely quickly (for their time) despite being serial, but more importantly they ran under DMA power all simultaneously and at the same time as the processor was doing its own thing independently.
- The above two features made the transputer exceptional for multiprocessing, but I think its instruction set was also far, far ahead of its time: not only ultra-RISC, but highly extensible too. For example, numeric literals in instructions were only as long as needed, because an extension bit would (if present) indicate that more bits were to follow if needed. This made code *extremely* tight. The scheme also allowed extensions to the instruction set to be made in a fully backward-compatible manner.
The transputer was ultra-cool, and the world hasn't seen anything like it since. No doubt somebody will reinvent this approach some day, but probably in a US or Japanese lab, as usual, and they'll take the credit for exceptional design ideas made in an earlier age. Sigh.
The subject line doesn't match the body of your reply: the subject suggests that variable-length IP addresses are already here, but then you clarify in the body that it's only domain names that have variable length.
Well, that isn't good enough. We have not hit any limits with domain names precisely because they are totally extensible overall and also in each component part, whereas the limits on IPv4 addresses are a daily headache right now, and the limits on IPv6 addressing will be a daily headache for us in the future, sooner than we think.
The suggestion that extensible IP addresses should be avoided because of the difficulty it might cause router manufacturers is so short-sighted that it's just a joke. If the existing crop of manufacturers can't come up with an efficient routing algorithm for extensible addresses then they should die and leave the profits to newer outfits that are willing to engage a few brain cells to solve the problem.
Seven computers will be sufficient to satisfy the world's entire computing requirement. And 640 Kbytes is more memory than the biggest PC user will ever need.
If there's one thing that computer people should have learned by now, it's that limits have a habit of being reached sooner than expected. Len Kleinrock is right to question the lack of extensibility in IPv6 addressing. It *will* bite us one day, and sooner than we think.
Maybe the more visionary kernel coders should get together and develop a stack for Linux or BSD based on one of the designs for extensible IP addressing. It's still early days for IPv6, and there is still ample opportunity for an upstart alternative design and implementation to make it big and become a de facto standard. And that would be good for us in the long run, which may be a lot less long than we think.
Having worked with perhaps a dozen different Unixes over the years, I think I can make a fairly reasonable assessment as to the merits of SCO as a Unix.
In my opinion, SCO is the biggest pile of rubbish on the Unix scene. I remember it as unstable, non-standard, lacking in features, sysadmin-unfriendly, user-hostile, expensive, poorly supported, and dead slow. The only things that it had going for it was that there were a lot of applications listed in its catalogue, and that after all it is a Unix. Personally I wouldn't touch it again with a barge pole though.
As for SCO's corporate stance re Linux, it's laughable so ignore it. Linux and the other free Unixes are the death of SCO, and if they're going to babble silly things as they die then good riddance.
Probably the top site for information on business applications for Linux is:
http://www.hex.net/~cbbrowne/finances.ht ml
User-friendliness has nothing to do with it. Windows is extraordinarily user-hostile, overly complex and appallingly obscure, but it benefits from one key fact: it comes preloaded on PCs, as a result of which prospective users don't have to go through the utterly user-unfriendly Windows installation procedure. If you put Windows and Linux head-to-head in front of a computer-unaware beginner with the goal of installing exactly the same feature set, Windows would be so far behind that it wouldn't even figure on the user-friendly map.
Don't flatter Microsoft with a "user-friendly" label. It's not. Their horrid product is just preloaded, and Linux isn't.
I hope this event gives paranoia a new lease of life, before it becomes accepted practice for hardware manufacturers to supply binary-only drivers for Linux.
It's bad enough to run binary-only applications, but to some extent their impact can be controlled. In contrast, once you add an opaque binary-only module into the kernel then all bets are off.
Although the article raises some interesting points, it paints with too broad a brush when saying that computer users are becoming disempowered. It's yet another case of statistical generalization, which may delight journalists and politicians but is always very annoying to those that don't follow others like sheep nor benefit from it. Some users are disempowered, yes, namely those that are not able to assess for themselves whether relying on a service like Hotmail or a company like Microsoft is a good idea, and those who are not able to make the right evaluation and move to other pastures. But does it disempower you, as Slashdot reader? Almost universally, no, because for the most part people who use this forum are competent enough to know when to leave a sinking ship or not to expose themselves to the hazard in the first place. We're not the Borg. We're individuals, and just because statistically something appears to be happening to some computer users doesn't mean that it is happening to computer users in general. There always will be people who are challanged in one or more areas and who as a result are prone to some group-specific ailment, but you can't extrapolate from that to the universe of people when that universe is as diverse as that of computer users.
The link to core News For Nerds material is actually very strong because of the Linux angle and because it affects nerds in Turkey directly.
It's certainly much more relevant than a lot of other stuff that was deemed acceptable -- the Blair Witch film comes to mind.
That's not a function of a normal MAC-level switch.
There are boxes that perform this kind of function (we use 'em), like the Cisco Local Director (works like a modified bridge) and various layer 4 switches too. That's why I was asking. What is Slashdot using?
Yes, I hear the difference too, and I mean at the highest normal bitrates (I haven't tried anything beyond 256K though). I wish somebody would do a proper scientific blind A-B test and publish the results to debunk the myth of no loss of fidelity.
I wouldn't say it sounds like crap though, just not hifi, and the high frequency artifacts are probably more irritating then FM radio.
'Pity that Shorten gives you only a 60% inprovement. That's so poor that MP3 isn't threatened at all.