The expressive power of a language is not the
same as its implementative power.
Just because you implement Haskell in C and
conversely doesn't mean that both languages are
equivalently expressive.
Implementative power is what you can do with
a program written in the language, once it's
finished; if you're to produce closed-source
stuff, that's the only thing you're interested
in, and that's why people have been brainwashed
by forty years of closed source into not caring
about more than it. The language could be
"write-only", and "write-once", implementative
power wouldn't be less.
Expressive power is the information you can
exchange using the language with other people
with whom you work. When programming is an
incremental undertaking, what you care about is
expressive power. That's what people need in
the world of free software computing.
To be very expressive, the language must be
"read-write" and "write-indefinitely".
For instance, a language to describe finite state
machines operating on an indefinite tape
(Turing Machines) can implement
any one-shot computations from integers
to integers in asymptotically optimal space and time. But as a tool for interprogrammer communication, it is not nearly as expressive as
Cayenne that allows to describe arbitrary functions from arbitrary higher-order types to
any other, including the ability to define
statically enforced logical invariants.
In Cayenne (which compiles into Haskell,
which can be compiled into machine code,
bypassing C for everything but for system
interface and runtime support),
you can define a type whose elements
are precisely all sort functions, excluding
any buggy function that sometimes fail.
The language can thus express much more
than C or any other implemented language,
for that matter.
As was proposed several times on slashdot, the solution is not just more TLDs, but infinitely many TLDs, whereas anyone could dynamically create a new TLD so as to register a domain under it. If FuckingCompany registers FuckingCompany.sucks, then it's still time to register FuckingCompany.fuckingsucks or even FuckingCompany.reallyfuckingsucks, etc. If each registration is $1, it'll cost $17576 to register every TLA TLD, and $3M to register every 6-letter TLD, and $5G to register every 9-letter TLD, and so on.
To replicate a lot of disks, and to be able to switch broken disks (or just to make easy real-world tests), I strongly recommend the use of disk racks.
Since for obvious cost reasons, you'll most probably use IDE on your deployed machines, I recommend that your replicator be a mostly SCSI machine, with an IDE rack, and IDE compiled as kernel modules; this way, you can hot-plug and hot-unplug your IDE racked disks as you replicate them, by modprobing the IDE modules in and out (be sure to use the right hdparm/ioctl calls to flush caches and shutdown disks, though): no reboot needed!
Even without a SCSI machine, I've installed a company's computer rooms with IDE racks, and it's been a pleasure to replicate, install, test, fix, and reinstall machines there. (I also used debian as the basic distribution, which isn't perfect in a NFS- and NIS- sharing environment, but is the least horrible thing I tried.)
Finally, if your machines are heavily networked, you can use such thing as the Linbox Network Architecture (web site currently down for recreation; contact the people at linbox.com), which is basically a lots of diskless linux clients (you buy a fast ethernet and RAM instead of lame IDE disks), and a few badass servers for files and applications. Ten times less disks to maintain and clients without persistent state means much less administration.
Of course they are moral entities. Everything anyone chooses and does has moral implications. The only entities that are not moral are dead entities. Corporations are just groups of people. They can ignore their moral responsibilities; they can fail to be up to them; but they just cannot evade the fact that they are moral entities, the fact that their actions (or lack thereof) have moral implications.
Of course, in absence of the Rule of Law, a lot of big corporations have managed to become big because they behaved in immoral and unfair ways that ought to have been repressed but couldn't by a flaw of Law or of its Rule. But that's the problem with Legislation, Law and Liberty; it isn't in any way specific to corporations.
Re:Maybe public key encryption and signed DNS?
on
Pirate DNS?
·
· Score: 1
No, you only need do one request, for the DNSNG will send you all the signed certificates you need together with the data. And the keys for the central trust repository will be included in your DNSNG client. The signed certificates for a lookup on foo.bar will be
The signature of the.bar registry, as signed by the central registry.
The public key of the foo.bar owner, as signed by the.bar registry.
The signature by the foo.bar owner of each chunk the foo.bar domain description.
Size of (compressed) chunks will be determined so as to optimize overall traffic. With a proper negociation protocol, no unneeded certificate need be sent over the wire. Finally, with a proper cache structure, only ISP's DNSNG caches need exchange certificates (sharing the checks for all of the ISP's users), while each user can hopefully trust the answers from his ISP's DNS without having to (systematically) check keys (he may check them once in a while if paranoid, or when connecting to e-business sites).
PS: Thanks, Mr Z, I was going to propose just the same thing as you did, and am glad I'm not the only (or first) one to think about it.
The TUNES project has an OS Review page that lists lots of code bases you may want to inspire from. It also provides lots of information on general OS concepts, so that you may produce original OSes, not just yet another toy kernel.
I could be totally misunderstanding what the goal of the protocol is.
Indeed you are, and your criticism is puny.
Your corporatist argumentum ad auctoritatem is most despicable. If anywhere, it should have been put at the end, with a much softer tone.
The ploy is not at all about secret-sharing, it's all about free publication. It is so that anyone can publish information without any of the individual participating servers being possibly held guilty.
The whole concept of "purely" random data is irrelevant hogwash. What counts is practically random data. White noise from your the kernel/soundcard/whatever, obfuscated by whatever cryptographic means of the day, is more than enough in practice.
The choice of pads can be automated according to any cold-minded (i.e. not you) cryptographer's criteria, if needed.
Alice doesn't have to claim she's the author. When she publishes a list of IDs, it is up to the police to prove that she is the original perpetrator, not just someone who repeats information given by someone else. Unless you make publishing of IDs a forbidden thing in itself. Welcome to the next DeCSS-like "everybody copies the list of IDs" contest!
The ploy indeed can be usefully combined with other ploys; David Madore makes no claim that it is a complete specification for a publication system. You will still need trust into the servers so as to keep no date information, you will still need crypto when communicating with servers if you fear sniffers and men-in-the-middle, etc.
Apart from the classical problems of client-server communication privacy (with classical solutions), and of trust in the servers (with classical lack of solutions), and can see a big difficulty in Madore's ploy: the challenge will be so that pads be reused a bit, but not too much. They must be reused a bit, so that no one can claim a particular pad as being culprit in a particular message; but they must not be reused too much, so as not to weaken the innocence of other pads in a message. Finally, it seems to me that servers should maintain some meta information, although (obviously) not publication date, but rather divulgation date (the date the message has first been observed in an ID list), so that one can avoid relying on pads all of which have been used, which would compromise the last message as "guilty". David Madore proved that you can sometimes prove pads "innocent", which allows to invoke the principle of presumption of innocence for pads. The danger is that proving too many pads innocent might result in finding the remaining ones guilty.
Ideally, no single pad (and thus no single pad publisher) can ever be held "guilty for sure" of holding any "forbidden" information. Maybe this ideal can be practically approximated; maybe it cannot. In either case, I'd be most interested in a proof, rather than in fallacious whining.
Disclaimer: I know David Madore personally, and although he's not a cryptographic academics, he is neither a crook nor a naive guy, and usually has interesting things to say; even when he is mistaken, it is most interesting to figure out how and why.
Why did you put a link to the presidency of the republic for the name "France" ? That's most insulting. France isn't its presidency anymore than the US is the white house; neither is Russia epitomized by the crooks from the Kremlin or China by the evil chinese communist party.
Government is an invention to oppress people; if you want to talk specifically about government, say "french government", not "France"; if you want to talk about the country, say "France", not "french government". I am french, yet I feel little solidarity with present, past and future french governments; I know for sure than in other countries were governments are even more on the loose than in France, people feel even less solidarity with their governments.
That said, like many french people, I think that the "anti-racist" law is itself a very racist law (apart from being most inappropriate), since it makes a special case for the jews; of course, I can't say it on any french media, least I be censored by this very law.
As a side note, I found a text by the same author against patents, but one for copyrights (none on the 'Net in english, AFAIK). Not that I agree with him on the latter point...
I'm very angry at university eggheads, who both accept to endorse responsibility for personal student pages (since they censor them on demand) and refuse to ever act responsibly by pondering all points of views when faced with this responsibility.
Even ISPs don't censor at the whim of the first come complainant; they ask for backed arguments, and have someone check the validity of the claim, before they take any action. Why don't universities just disclaim responsibility for the contents of personal pages? In any case, censorship is not something to use lightly, and at the very least, the accused students ought to be able to defend themself before being censored.
Ok, so AC allegedly owns his comments; but property only extends as far as the right to sue. Now, who'll sue Slashdot if an AC post is published? Nobody. Therefore, for all that matters, Slashdot (or anyone, if Slashdot denies property) can publish AC posts however it likes.
That said,/. may lose anonymous writership if it publishes AC posts w/o permission, so a preference-defaulted checkbox to allow or disallow republication on other media is a good idea. On the other hand, if people write on such fora, it is in the hope of being read; publication on other media only enhances the public; hence, the default should be to allow republication (with proper authorship acknowledgement, of course), since it only amplifies the displayed goal of the poster.
Anyway, down with intellectual property! Long live acknowledgement of authorship!
This suggests that DDoS attacks against 800 numbers could be made: you can only, say, phone once per received spam message; but then, everyone in the anti-spam club can phone once, too, for every message of every spammer received. So there could be an automated system that would collect spammer 800 numbers, messages to send to them, and client software so that people equipped with proper voice modems can connect and gripe.
I don't have any voice modem, and don't know if some such modems exist with enough documentation; also, it would require a small recording software to make messages tailored to 800 answer boxes (i.e. wait the right time, emit tone, wait, emit tone, wait, speak). Voice-Synthetizing the messages could also save bandwidth for the client connections.
Darn! For once, I post a message that's moderated up, and I'm hit with a slashbug that makes it AC. Then I get Error:Undefined subroutine &Slash::selectStories called at (eval 72) line 2. and now an internal server error. Looks like the new server has got problems. Is it the $5000 in my message confusing some slash perl script?
Assuming the publisher is responsible, if there is Microsoft-copyrighted material published on MSN, this is no copyright infringement. Of course, MS may remove it nonetheless if it wishes, since it publishes the pages and is not forced to keep it. But again, this is no copyright infringement.
Assuming the publisher isn't responsible, then he is not forced to remove even copyrighted material from his site.
This all suggests another copyright fighting tactic: posting lots of copyrighted material on information boards published by copyright-holder...
If what you did was copy the look and feel, while rewriting the underlying code, then precedent exist: the dreaded look and feel litigations against which the LPF has fought so long.
There's been for a very long time a free OS whose design is close from that of QNX: VSTa. If you want a light, somewhat POSIXish, microkernel-based OS, you know where to find it.
Would those companies have even come to life to research such things if they knew that the moment they figured out how to do what they do, every other company would be able to use their technology? Of course they would. Of course they do. Same goes for software? Would people write and publish software even without a monopoly afterwards? Of course they would. Of course they do. As long as there is a USE for innovation, there will be a MARKET for innovation. If you remove these monopolies, you get a FREE market. Patents are EVIL. They consist in getting paid for PREVENTING innovation, instead of getting paid for constantly renewed innovation. The free software model shows how the only thing that makes a company successful is its ability to stay ahead technologically. What do we have? The strongest incentive to innovate, ever. Free Software! Free Information!
Slashdot Mirror
Implementative power is what you can do with a program written in the language, once it's finished; if you're to produce closed-source stuff, that's the only thing you're interested in, and that's why people have been brainwashed by forty years of closed source into not caring about more than it. The language could be "write-only", and "write-once", implementative power wouldn't be less.
Expressive power is the information you can exchange using the language with other people with whom you work. When programming is an incremental undertaking, what you care about is expressive power. That's what people need in the world of free software computing. To be very expressive, the language must be "read-write" and "write-indefinitely".
For instance, a language to describe finite state machines operating on an indefinite tape (Turing Machines) can implement any one-shot computations from integers to integers in asymptotically optimal space and time. But as a tool for interprogrammer communication, it is not nearly as expressive as Cayenne that allows to describe arbitrary functions from arbitrary higher-order types to any other, including the ability to define statically enforced logical invariants. In Cayenne (which compiles into Haskell, which can be compiled into machine code, bypassing C for everything but for system interface and runtime support), you can define a type whose elements are precisely all sort functions, excluding any buggy function that sometimes fail. The language can thus express much more than C or any other implemented language, for that matter.
-- Faré @ TUNES.org
-- Faré @ TUNES.org
-- Faré @ TUNES.org
Since for obvious cost reasons, you'll most probably use IDE on your deployed machines, I recommend that your replicator be a mostly SCSI machine, with an IDE rack, and IDE compiled as kernel modules; this way, you can hot-plug and hot-unplug your IDE racked disks as you replicate them, by modprobing the IDE modules in and out (be sure to use the right hdparm/ioctl calls to flush caches and shutdown disks, though): no reboot needed!
Even without a SCSI machine, I've installed a company's computer rooms with IDE racks, and it's been a pleasure to replicate, install, test, fix, and reinstall machines there. (I also used debian as the basic distribution, which isn't perfect in a NFS- and NIS- sharing environment, but is the least horrible thing I tried.)
Finally, if your machines are heavily networked, you can use such thing as the Linbox Network Architecture (web site currently down for recreation; contact the people at linbox.com), which is basically a lots of diskless linux clients (you buy a fast ethernet and RAM instead of lame IDE disks), and a few badass servers for files and applications. Ten times less disks to maintain and clients without persistent state means much less administration.
Yours freely,
-- Faré @ TUNES.org
Of course they are moral entities. Everything anyone chooses and does has moral implications. The only entities that are not moral are dead entities. Corporations are just groups of people. They can ignore their moral responsibilities; they can fail to be up to them; but they just cannot evade the fact that they are moral entities, the fact that their actions (or lack thereof) have moral implications.
Of course, in absence of the Rule of Law, a lot of big corporations have managed to become big because they behaved in immoral and unfair ways that ought to have been repressed but couldn't by a flaw of Law or of its Rule. But that's the problem with Legislation, Law and Liberty; it isn't in any way specific to corporations.
-- Faré @ TUNES.org
- The signature of the
.bar registry, as signed by the central registry. - The public key of the foo.bar owner, as signed by the
.bar registry. - The signature by the foo.bar owner of each chunk the foo.bar domain description.
Size of (compressed) chunks will be determined so as to optimize overall traffic. With a proper negociation protocol, no unneeded certificate need be sent over the wire. Finally, with a proper cache structure, only ISP's DNSNG caches need exchange certificates (sharing the checks for all of the ISP's users), while each user can hopefully trust the answers from his ISP's DNS without having to (systematically) check keys (he may check them once in a while if paranoid, or when connecting to e-business sites).PS: Thanks, Mr Z, I was going to propose just the same thing as you did, and am glad I'm not the only (or first) one to think about it.
-- Faré @ TUNES.org
-- Faré @ TUNES.org
Is that some suggestion as to a way to assassinate said victim? Being inserted into a slashdot comment must be a painful death indeed!
-- Faré @ TUNES.org
Indeed you are, and your criticism is puny.
Apart from the classical problems of client-server communication privacy (with classical solutions), and of trust in the servers (with classical lack of solutions), and can see a big difficulty in Madore's ploy: the challenge will be so that pads be reused a bit, but not too much. They must be reused a bit, so that no one can claim a particular pad as being culprit in a particular message; but they must not be reused too much, so as not to weaken the innocence of other pads in a message. Finally, it seems to me that servers should maintain some meta information, although (obviously) not publication date, but rather divulgation date (the date the message has first been observed in an ID list), so that one can avoid relying on pads all of which have been used, which would compromise the last message as "guilty". David Madore proved that you can sometimes prove pads "innocent", which allows to invoke the principle of presumption of innocence for pads. The danger is that proving too many pads innocent might result in finding the remaining ones guilty.
Ideally, no single pad (and thus no single pad publisher) can ever be held "guilty for sure" of holding any "forbidden" information. Maybe this ideal can be practically approximated; maybe it cannot. In either case, I'd be most interested in a proof, rather than in fallacious whining.
Disclaimer: I know David Madore personally, and although he's not a cryptographic academics, he is neither a crook nor a naive guy, and usually has interesting things to say; even when he is mistaken, it is most interesting to figure out how and why.
-- Faré @ TUNES.org
-- Faré @ TUNES.org
Bzzzttt. What about Genera, the operating system of ages-old LISP Machines?
-- Faré @ TUNES.org
Government is an invention to oppress people; if you want to talk specifically about government, say "french government", not "France"; if you want to talk about the country, say "France", not "french government". I am french, yet I feel little solidarity with present, past and future french governments; I know for sure than in other countries were governments are even more on the loose than in France, people feel even less solidarity with their governments.
That said, like many french people, I think that the "anti-racist" law is itself a very racist law (apart from being most inappropriate), since it makes a special case for the jews; of course, I can't say it on any french media, least I be censored by this very law.
-- Faré @ TUNES.org
As a side note, I found a text by the same author against patents, but one for copyrights (none on the 'Net in english, AFAIK). Not that I agree with him on the latter point...
-- Faré @ TUNES.org
-- Faré @ TUNES.org
Even ISPs don't censor at the whim of the first come complainant; they ask for backed arguments, and have someone check the validity of the claim, before they take any action. Why don't universities just disclaim responsibility for the contents of personal pages? In any case, censorship is not something to use lightly, and at the very least, the accused students ought to be able to defend themself before being censored.
-- Faré @ TUNES.org
That said, /. may lose anonymous writership if it publishes AC posts w/o permission, so a preference-defaulted checkbox to allow or disallow republication on other media is a good idea. On the other hand, if people write on such fora, it is in the hope of being read; publication on other media only enhances the public; hence, the default should be to allow republication (with proper authorship acknowledgement, of course), since it only amplifies the displayed goal of the poster.
Anyway, down with intellectual property! Long live acknowledgement of authorship!
-- Faré @ TUNES.org
Error Occurred While Processing Request
Error Diagnostic Information
WaitNamedPipe returned FALSE. Windows NT error number 121 occurred.
Hum. Can we trust such a buggy NT-based service to provide secure spam recycling services?
-- Faré @ TUNES.org
I don't have any voice modem, and don't know if some such modems exist with enough documentation; also, it would require a small recording software to make messages tailored to 800 answer boxes (i.e. wait the right time, emit tone, wait, emit tone, wait, speak). Voice-Synthetizing the messages could also save bandwidth for the client connections.
Just my 2 eurocents worth...
-- Faré @ TUNES.org
-- Faré @ TUNES.org
How dare you insult gorillas? They are nice beasts that mean no harm, not ruthless brutes that will attack anyone who stands in their way.
-- Faré @ TUNES.org
This all suggests another copyright fighting tactic: posting lots of copyrighted material on information boards published by copyright-holder...
-- Faré @ TUNES.org
-- Faré @ TUNES.org
Nice oxymoron.
-- Faré @ TUNES.org
Of course microkernels suck, anyway.
-- Faré @ TUNES.org
-- Faré @ TUNES.org