In Bruce Sterling's short story "Maneki Neko", everyone has a pda/cellphone thing with pervasive wireless networking and GPS. The folks in the story are part of a P2P network whose symbol is 'Maneki Neko', and whose function is to automates a gift economy.
Say you're in the coffee shop, buying a cup. The PDA buzzes, says 'buy two'. So you do. You walk out with two, it buzzes again: 'give it to the hung-over chap on the bench'. He's psyched, even though he didn't order it, it's what he needed. Since the network has some idea of what you have purchased, what you need, where you are, what you've been doing, and what you have extra of, it efficiently moves goods (and without spoiling the story, personal services) around without there being anyone in charge. And since we have databases, fourteen people don't show up with coffees for the poor lush.
In the story, the main character is having a baby. Unsolicited baby clothes (for the correct sex) show up in the mail, along with toys, etc, sent by total strangers, because their PDA told them to. Presumably they had extra, or their child had outgrown it, or whatever. And since the network often benefits them, they have an incentive to comply with its requests, when they can.
Now other than the rampant privacy problems involved in a world that has such devices and services working seamlessly on a global scale, doesn't it sound cool? And since we're going to end up with a world that has such devices and services working (we hope) seamlessly on a global scale, should we not make such a thing?
There's a particular aesthetic impression you get from jazz that you can identify and recognize right away. It's the same with SF -- once you get used to it, you just know.
- Neal Stephenson, 2004
I can't define pornography, but I know it when I see it.
- Supreme Court Justice Potter Stewart, 1964
I guess I'm not entirely satisfied with the 'know it if you see it' argument... certainly, the aesthetic he carries from earlier work is present in Quicksilver, but that could simply be his writing style, showing through.
If Quicksilver can be described as SciFi, it's definitely an outlier in the category. It is obviously fiction, of a historical bent, and it has a lot of science in it, but one could imagine writing imaginary conversations between Newton and Liebnitz, for example, without it being considered Science Fiction.
For inspiration I think of Walter Murphey's 1976 disco classic 'A Fifth of Beethoven'. This reworking of an historical classic is disco for recognizable reasons: the beat, the instrumentation, the structural changes, its length, etc. Similarly, Quicksilver can be seen as a SciFi riff on a historical material for recognizable reasons. Later in the article, he articulates one of those reasons:
How is it going to be used, not just by engineers who design products but by regular people who pick this stuff up and turn it to their own weird ends?
So, science fiction could perhaps be described as speculative writing in which science/technology plays a central role, and in which characters in the story turn the science to their own ends.
If you're being a good Peopleware people, and have spent the Mythical Man Month in the past rewriting crappy designs, you might want to start your project by writing a complete design, and releasing the design for comments BEFORE you write the code.
If you have written a coherent design, and it has been reviewed and hopefully added to by your peers, the code will come rapidly, and will be less painful to change over time. Also, I can't help noting that it's a lot more satisfying to do something technically sweet, even if it takes a little longer to get to the coding stage, and have some chance that it will actually be used by others.
If it is a commercial product that is to be open-sourced, make sure that your revenue model for the product is such that you DON'T CARE if other companies like you take the code (or the design) and run. If, for example, you are writing the world's greatest encrypted email server, you might want other entities to charge for the same thing you charge for, as long as they register their servers so that you can show that your tech is gaining market share. Open-source products with commercial possibilities can't rely on the "we have the code, you don't" advantage as a startup revenue model, it defeats the purpose of open source and will only lead to painful arguments with the business guys.
If there is one thing I would like to say to all potential mass-murderers out there who are in the final planning stages of their rampage, it is this:
Get good publicity photos made before you kill.
I am outraged at the poor quality of photos of the deceased and deranged murderers who perpetrate media-rich acts of wanton violence: mostly they are grainy, bad black-and-white scans of some Freshman Yearbook photo.
Hey, if I thought the photo that was going to immortalize me in the minds of millions on the evening news would be one of me, pimply, awkward, and forced to smile in braces, I would seriously consider going on a killing rampage myself.
There are a few things that I did not like when I recently switched from Slakware to RedHat, but overall, the experience was a positive one, so I now suggest that beginners install RedHat as their first distribution.
Particularly Annoying Bits (or PAB's:) :
No/usr/adm, not even a link to the/var/log. For those of us who grew up with 'tail/usr/adm/messages/' ingrained in our brainstem, this was annoying. The fix, obviously, is trivial.
Goddamned runlevels./etc/rc.d is WAY too complicated for any purpose I can realistically see people desiring... and it is nontrivial for administrators to jump between distributions on different machines. The Slakware scheme was more than sufficient. Thankfully, VAResearch includes a cheat-sheet with all of their servers, making the transition easier if you have the means.
The Arch-PAB! The RPM for a later version of afterstep (no wm flames please) that I got FROM THE REDHAT WEB SITE placed the 'afterstep' executable in a different place from RedHat 5.2! Major annoyance to find this, and I suspect there are plenty of other RPMs out there doing similarly nasty (and transparent!) things.
... if we wish to truly make Linux and other open source elements the shiniest toys in our collective toybox. It is the transparency of the process, including the arguments and the flame wars, that allows us to truly form our own opinions, and to plot the course we each want to take on the Ocean of Open Software.
In Neal Stephenson's article mentioned yesterday on Slashdot [ http://www.cryptonomicon.com/beginning_print.html ] he notes that in the bug-tracking and fixing process, the fact that one can access the questions and comments of other users is essential to the process of open source initiatives creating truly superior software. I think the same is true of creating a truly superior culture surrounding the software, we need to see the "bugs" in the process, and hear the arguments back and forth about which are truly bugs, if we wish to move our tent to the section of the bazaar that suits us.
Our advantage lies not in the fact that we present a coherent illusion of the products we create communally, but in the exact opposite phenomenon: that such a diverse and undisciplined group of individuals creates something useful to all of us without any of us explicitly agreeing on what we are doing.
I don't think we need leaders, at least not in the sense that we need an analog to Jobs or Gates to advocate a coherent "product" that we are building for the common good or to battle MicroSoft. In fact, we tend to rail against such people, because (a) we aren't getting paid by them, and (b) we never agreed on anything in particular except that stable software is useful to us. Perhaps by trying to position himself as such a person, Mr. Raymond has inadvertently drawn the lightening that we are always sure to generate when presented with someone attempting to narrow, or to make coherent, a "movement" that is the antithesis of coherent, monolithic effort.
When you are building a Cathedral, you need an architect. I don't think we need Mr. Raymond or anyone else to come to the bazaar and try to impose (or even abstract) a structure on something that is by nature fluid and chaotic.
Slashdot Mirror
| I give this book an 8/10, meaning you can't live without a copy!
Apparently, a 10/10 would mean "If you don't already have this book, you have died."
In Bruce Sterling's short story "Maneki Neko", everyone has a pda/cellphone thing with pervasive wireless networking and GPS. The folks in the story are part of a P2P network whose symbol is 'Maneki Neko', and whose function is to automates a gift economy.
Say you're in the coffee shop, buying a cup. The PDA buzzes, says 'buy two'. So you do. You walk out with two, it buzzes again: 'give it to the hung-over chap on the bench'. He's psyched, even though he didn't order it, it's what he needed. Since the network has some idea of what you have purchased, what you need, where you are, what you've been doing, and what you have extra of, it efficiently moves goods (and without spoiling the story, personal services) around without there being anyone in charge. And since we have databases, fourteen people don't show up with coffees for the poor lush.
In the story, the main character is having a baby. Unsolicited baby clothes (for the correct sex) show up in the mail, along with toys, etc, sent by total strangers, because their PDA told them to. Presumably they had extra, or their child had outgrown it, or whatever. And since the network often benefits them, they have an incentive to comply with its requests, when they can.
Now other than the rampant privacy problems involved in a world that has such devices and services working seamlessly on a global scale, doesn't it sound cool? And since we're going to end up with a world that has such devices and services working (we hope) seamlessly on a global scale, should we not make such a thing?
I guess I'm not entirely satisfied with the 'know it if you see it' argument
If Quicksilver can be described as SciFi, it's definitely an outlier in the category. It is obviously fiction, of a historical bent, and it has a lot of science in it, but one could imagine writing imaginary conversations between Newton and Liebnitz, for example, without it being considered Science Fiction.
For inspiration I think of Walter Murphey's 1976 disco classic 'A Fifth of Beethoven'. This reworking of an historical classic is disco for recognizable reasons: the beat, the instrumentation, the structural changes, its length, etc. Similarly, Quicksilver can be seen as a SciFi riff on a historical material for recognizable reasons. Later in the article, he articulates one of those reasons:
So, science fiction could perhaps be described as speculative writing in which science/technology plays a central role, and in which characters in the story turn the science to their own ends.If you're being a good Peopleware people, and have spent the Mythical Man Month in the past rewriting crappy designs, you might want to start your project by writing a complete design, and releasing the design for comments BEFORE you write the code.
If you have written a coherent design, and it has been reviewed and hopefully added to by your peers, the code will come rapidly, and will be less painful to change over time. Also, I can't help noting that it's a lot more satisfying to do something technically sweet, even if it takes a little longer to get to the coding stage, and have some chance that it will actually be used by others.
If it is a commercial product that is to be open-sourced, make sure that your revenue model for the product is such that you DON'T CARE if other companies like you take the code (or the design) and run. If, for example, you are writing the world's greatest encrypted email server, you might want other entities to charge for the same thing you charge for, as long as they register their servers so that you can show that your tech is gaining market share. Open-source products with commercial possibilities can't rely on the "we have the code, you don't" advantage as a startup revenue model, it defeats the purpose of open source and will only lead to painful arguments with the business guys.
If there is one thing I would like to say to all potential mass-murderers out there who are in the final planning stages of their rampage, it is this:
Get good publicity photos made before you kill.
I am outraged at the poor quality of photos of the deceased and deranged murderers who perpetrate media-rich acts of wanton violence: mostly they are grainy, bad black-and-white scans of some Freshman Yearbook photo.
Hey, if I thought the photo that was going to immortalize me in the minds of millions on the evening news would be one of me, pimply, awkward, and forced to smile in braces, I would seriously consider going on a killing rampage myself.
There are a few things that I did not like when I recently switched from Slakware to RedHat, but overall, the experience was a positive one, so I now suggest that beginners install RedHat as their first distribution.
:) :
/usr/adm, not even a link to the /var/log. For those of us who grew up with 'tail /usr/adm/messages/' ingrained in our brainstem, this was annoying. The fix, obviously, is trivial.
/etc/rc.d is WAY too complicated for any purpose I can realistically see people desiring ... and it is nontrivial for administrators to jump between distributions on different machines. The Slakware scheme was more than sufficient. Thankfully, VAResearch includes a cheat-sheet with all of their servers, making the transition easier if you have the means.
Particularly Annoying Bits (or PAB's
No
Goddamned runlevels.
The Arch-PAB! The RPM for a later version of afterstep (no wm flames please) that I got FROM THE REDHAT WEB SITE placed the 'afterstep' executable in a different place from RedHat 5.2! Major annoyance to find this, and I suspect there are plenty of other RPMs out there doing similarly nasty (and transparent!) things.
... if we wish to truly make Linux and other open source elements the shiniest toys in our collective toybox. It is the transparency of the process, including the arguments and the flame wars, that allows us to truly form our own opinions, and to plot the course we each want to take on the Ocean of Open Software.
In Neal Stephenson's article mentioned yesterday on Slashdot [ http://www.cryptonomicon.com/beginning_print.html ] he notes that in the bug-tracking and fixing process, the fact that one can access the questions and comments of other users is essential to the process of open source initiatives creating truly superior software. I think the same is true of creating a truly superior culture surrounding the software, we need to see the "bugs" in the process, and hear the arguments back and forth about which are truly bugs, if we wish to move our tent to the section of the bazaar that suits us.
Our advantage lies not in the fact that we present a coherent illusion of the products we create communally, but in the exact opposite phenomenon: that such a diverse and undisciplined group of individuals creates something useful to all of us without any of us explicitly agreeing on what we are doing.
I don't think we need leaders, at least not in the sense that we need an analog to Jobs or Gates to advocate a coherent "product" that we are building for the common good or to battle MicroSoft. In fact, we tend to rail against such people, because (a) we aren't getting paid by them, and (b) we never agreed on anything in particular except that stable software is useful to us. Perhaps by trying to position himself as such a person, Mr. Raymond has inadvertently drawn the lightening that we are always sure to generate when presented with someone attempting to narrow, or to make coherent, a "movement" that is the antithesis of coherent, monolithic effort.
When you are building a Cathedral, you need an architect. I don't think we need Mr. Raymond or anyone else to come to the bazaar and try to impose (or even abstract) a structure on something that is by nature fluid and chaotic.