Osty writes ... designed to protect the consumer, not a monopoly's competitors. Samba's the latter... companies that can no
longer compete in a market. There's nothing illegal with going out of business.
As it happens, Samba is not a compeditor, nor a company
that cannot compete in the market. It's a
effort to allow consumers to use the protocols,
something that is well within the lawfull aims of
antitrust.
...for Linux, that is!
We've had public key cryptography for a while,
thanks to Dr. Diffie and friends, I wonder if it's
time to prototype a real personal wallet
framework around PK and get someone like
Whitfield Diffie to push it as a privacy-friendly
form of magic authentication.
How about "if you don't have a PenguinCard we
can't look you up in our Oracle database,
so you can't get you on the plane".
Messers Hrebejk and Boudreau would like this to
be the case, but they're only argued that
one **might** occur, not that it must.
"Natural" monoplies are those which arise
because there cannot be anything else. For example, it's not credible to expect more
than one hydro cable to your house, or more
than one telephone company: there isn't enough
room on the poles for more than one each. Heck,
the cable folks often have to sue to get permission to use the other guy's poles...
These are existing natural monopolies, along
with water, sewers and the like.
They propose that Bill has a natural monopoly,
and that it will be broken and replaced with
one where we are the monopolists. Well, natural monopolies are not trivial to overturn, so by arguing the we will overturn Bill, they're
(accidentally!) arguing that he doesn't have
a natural monopoly.
They have argued that some of the prerequisites exist, but they've jumped from there to the conclusion that it does exist, without
offering proof.
A cople of references, findable via google: A Glossary of Political Economy Terms, and from
http://www1.oecd.org/daf/clp/non-member_activities/dnme10.htm,
Natural monopoly arises in sectors characterised by declining costs of production so that there is room for only
one firm to exploit available economies of scale. Typically, natural monopolies occur in industries characterised
by large distribution networks with substantial fixed costs, such as gas, electricity, water and railways. In
practice, however, it is as rare to find examples of industry-wide natural monopoly as it is examples of perfect
competition. Even if some parts of an industry have natural monopoly characteristics others may be potentially
competitive. For example electricity supply consists of natural monopoly in transmission but potential
competition in generation and the supply of user equipment. Telecommunications was for a long time considered
to be a natural monopoly at least for the basic telephone service, whereas value added services and the
equipment market are competitive. The creation of new networks for voice transmission has even eroded the
monopoly of the basic service.
It's a cool conceit to call virus-writing a form
of terrorism, but I don't expect it to be taken
all that seriously.
We had home-grown terrorists in Canada some years ago, panicked, declared martial law,
called in the army and started declaring all sorts of things
terrorism, not just blowing up mailboxes and murdering diplomats.
found the terrorists.
Six months later we were back to normal,
and all the quasi-terroristic things we prohibited (public gatherings, for one) were back legal again. I expect no less from the U.S.
I suspect it's FUD: no-one's announced a
cancellation...
Sun has announced that Gnome 2.0 (or later)
will be on a later version of Solaris,
and that OpenOffice (based on StarOffice)
wil be part of the desktop. Speak to your
Salesperson for specific plans.
Personal opinion: someone's suckered Linuxgram
with FUD.
As another commentator pointed out, this
can be hard if your C++ compiler gets in
the way (;-))
The general problem was solved back before
Unix (on Multics, to be precise), and an
equivalent problem was solved for databases by
the relational algebra folks.
You need a motivation to do it, though, and
that motivation's not there. You need
to be worried about stability without
preventing innovation. A few years
back a quite large
part of it was solved for Suns by My Evil
Twin, David J. Brown: a two-part version
number for every interface in the libraries,
checked at link time.
The U.S. constitution isn't a contract, it's
what it says it is, a constitution. It defines
what laws can be created.
It specifically controls the criminal law,
which in turn controls what one citizen can do to another,
by defining certain actions as criminal. Amoung
these are breaking and entering and the
malicious destruction of property.
--dave
[Some folks who've never read history
(or Ayn Rand!) keep
making this "all law is contract law"
claim. In fact, contract law came
into existance long after criminal...
try looking up "hanseatic league"
or "emergent law". Sheesh!]
This is utterly typical behavior of someone who's
convinced themselves they're "right".
A parallel kind of behavior is the tendancy of
professional criminals to get caught on
traffic violations: the get blase' about
breaking laws and then do something dumb right
in front of a traffic cop.
Both are variations on the psycology that
makes "time wounds all heels" true (;-))
Mr Coward wrote: N\It's obvious (to me, at least) that you can't build a sucessful
open-source application based on a closed source platform like Java. Sooner or later, the virus of commercialism will invade the mindset of all the layers above it. Immagine if Apache, or Emacs were written in a closed source language. They would not be what they are today.
I quite disagree: all sorts of open source
offerings were written on top of closed source
platforms, notably Unix operating systems, by writing to stable and published interfaces. For example, Emacs!
I also think Mr. Coward (Noel?) is confusing closed
source platforms (eg, OSs, runtimes) and languages (eg, C, Java).
Finally, he suggests that closed source
is a virus: if that were true, the converse would be true, and Microsoft would be "justified" in
trying to prevent the use of open source operating systems and software with their closed-source
OSs... for fear it will take over the
minds of their lawyers.
I, for one, am not prepared to agree to
that kind of arguement: I suspect Noel's
not going to want to either!
--dave
Quite, but we do have one advantage that's
reasonably technological: logistics.
With modern cargo aircraft and shiping, a force
operating inside Afganistan in support of the
Northern Alliance will be able to use
light armour, self-propelled artillery
and wheeled vehicles in offensive operations
without worrying about:
- running out of ammunition
- freezing to death
- running out of gas in mid-battle.
And the core force will arguably be
a line infantry unit of moderate size, with
arty and some integral anti-tank capability.
I'd guess at a reinforced brigade group.
Anything larger would tend to trip ove
its own feet, if not those of the Alliance.
I know there are specific courts for certain technical matters, as I stumbled over an
admiralty court once, and it was ferociously
well-informed.
I wonder if there are equally capable
committees or subcomittees in the house
or senate, that are good forums for hard
problems...
In Canada, where our Senate is appointed
for life (and is "called the house of cronies"
by the sarcastic), there are some very capable
standing comittees, but they have limited powers
to write legislation...
In effect, I'm asking where in the
U.S. government one applies pressure by
bringing forward facts, not just opinion
or numbers...
nachoworld wrote:
Why is the U.S. such a big believer in precedents? No other country...
Sorry, but you're labouring under a misapprehension: the entire english-speaking
world works from precedent. It's inherited
from Britain, the same place the U.S. inherited its common law from.
The same qustion comes up in each generation:
my father had a book on the value of what was
then called a liberal education.
For the record, a univeristy education pays
off in dollars.
The things I expected
I'd never use, I use every day. The things
I studied with intention to use, I never did.
A friend who went to community college
to get training in programming... makes
exactly half what I do per year!
So now I'm a well-rounded (fat???), well-paid
performance engineer at a three-letter-name
company, having a heck of a good time and getting
to tease the people who thought a philosophy
degree would lead to a job driving taxi...
And I write nerd books on the side, which I
never could have done with the grasp of english
I had before going to university.
I disagree: You need to distinguish betwen the
public action of mailing or emailing a letter,
and the private contents of the letter.
This is well understood in business:
to quote Shanti Atkins (a business
connsultant) speaking in Wired
"if an employee is led to expect something is private, such as
e-mail communications, then that privacy cannot be violated. But, if the
company informs its employees that, for example, e-mail sent over the
company's network is monitored, then the employee can no longer claim an
"expectation of privacy." In short, once the company stakes its claim over its
cyber-dominion, its employees have no right to privacy there.
I expect the fact of the mailing to be public,
and the contents to be private. I do have
some duties, though: sealing the message in an envelope is more secure than sending a postcard,
and encrypting it is stronger still. But that's
a different discussion...
The interesting ones are the Apache Software Foundation and O'Reilly and Associates, who
are likely to be concerned about all our
concerns. One produces a certain heavily-used
program which could use or reject untrustworthy suppliers, and the other might well publicise
misbehavior...
marm notes I think the major technical stumbling block to allowing applets or scripting in any language is
finding a language-neutral way of restricting execution along the lines of Java's security
model.
I agree, but building such security
regiemes is well within the state of the art.
It was state-of-practice in Multics, used to control sysadmin.daemon's access to your files
for backup purposes.
Although it fell out of use with the creation of Unix, it's slowly coming back into our
consciousness once more.
You need a set of access controls on
what a particular "surrogate user" (the entension)
is allowed to use, a way of authenticating the user
such as a digital signature, and a way of associating
the two. I can easily see /tmp/$$.$Z/* Microsoft.Passport rwe
$HOME/.passport Microsoft.Passport r
giving a downloaded MS Passport extension
access to a per-process, per-extension directory
in/tmp and a config file in my home directory.
(Although Passport would probably want rwe on root (;-))
Have a look for "mandatory access" (MAC) and
"role-based access control" (RBAC) in sites on Java security and the NSA's secure Linux.
They wish a scalable OS with location
transparency, which implies small and bounded
times independant of location. This little
problem was satirized by a physicist friend[1],
who once explained to an overenthusiastic
customer that "the system will perform as you
desire only if we can increase the speed
of light".
Actually I expect the opposite: same kind of reaction as in Canada after the October Crisis of 1970
or in the the U.S. after the last several wars.
Our American Cousins have a tradition of
taking action, solving their problems and then
putting their lives and country back together.
It might have to do with their first president
rejecting the crown and going home after the revolutionary war...
--dave
[Canada declared martial law in October 1970
after an internal terrorist attack on Pierre
Laporte, Quebec's labour and immigration minister
and James Cross, Britain's trade commissioner.
About a month later, those who had kidnapped and
killed Laporte were arrested, and the innocent victims of the martial law period were reimbursed.
The problem is a particular instance of the general
"continuous maintenance problem", which is now reasonably
well understood (i.e., it's not academically interesting any
more). This has been solved in particular cases as far back
as Multics and, more topically, in the relational database world.
I have one of Paul Stachour's papers on the subject, although
it needs to be OCR'd...
Slashdot Mirror
Osty writes ... designed to protect the consumer, not a monopoly's competitors. Samba's the latter. .. companies that can no
longer compete in a market. There's nothing illegal with going out of business.
As it happens, Samba is not a compeditor, nor a company
that cannot compete in the market. It's a
effort to allow consumers to use the protocols,
something that is well within the lawfull aims of
antitrust.
...for Linux, that is! We've had public key cryptography for a while, thanks to Dr. Diffie and friends, I wonder if it's time to prototype a real personal wallet framework around PK and get someone like Whitfield Diffie to push it as a privacy-friendly form of magic authentication. How about "if you don't have a PenguinCard we can't look you up in our Oracle database, so you can't get you on the plane".
"Natural" monoplies are those which arise because there cannot be anything else. For example, it's not credible to expect more than one hydro cable to your house, or more than one telephone company: there isn't enough room on the poles for more than one each. Heck, the cable folks often have to sue to get permission to use the other guy's poles... These are existing natural monopolies, along with water, sewers and the like.
They propose that Bill has a natural monopoly, and that it will be broken and replaced with one where we are the monopolists. Well, natural monopolies are not trivial to overturn, so by arguing the we will overturn Bill, they're (accidentally!) arguing that he doesn't have a natural monopoly.
They have argued that some of the prerequisites exist, but they've jumped from there to the conclusion that it does exist, without offering proof.
A cople of references, findable via google: A Glossary of Political Economy Terms, and from http://www1.oecd.org/daf/clp/non-member_activities /dnme10.htm,
Natural monopoly arises in sectors characterised by declining costs of production so that there is room for only one firm to exploit available economies of scale. Typically, natural monopolies occur in industries characterised by large distribution networks with substantial fixed costs, such as gas, electricity, water and railways. In practice, however, it is as rare to find examples of industry-wide natural monopoly as it is examples of perfect competition. Even if some parts of an industry have natural monopoly characteristics others may be potentially competitive. For example electricity supply consists of natural monopoly in transmission but potential competition in generation and the supply of user equipment. Telecommunications was for a long time considered to be a natural monopoly at least for the basic telephone service, whereas value added services and the equipment market are competitive. The creation of new networks for voice transmission has even eroded the monopoly of the basic service.
We had home-grown terrorists in Canada some years ago, panicked, declared martial law, called in the army and started declaring all sorts of things terrorism, not just blowing up mailboxes and murdering diplomats. found the terrorists.
Six months later we were back to normal, and all the quasi-terroristic things we prohibited (public gatherings, for one) were back legal again. I expect no less from the U.S.
Actually he works for a U.S. company, so he
just might want to visit the hom e office some time.
If she was having trouble with ApplixWare's spell-checker, maybe she should have investigated a Free Software solution
But surely the helpful person who deleted her work should have been fired for destroying company property without permission...
I suspect it's FUD: no-one's announced a
cancellation...
Sun has announced that Gnome 2.0 (or later)
will be on a later version of Solaris,
and that OpenOffice (based on StarOffice)
wil be part of the desktop. Speak to your
Salesperson for specific plans.
Personal opinion: someone's suckered Linuxgram
with FUD.
--dave
The general problem was solved back before Unix (on Multics, to be precise), and an equivalent problem was solved for databases by the relational algebra folks.
You need a motivation to do it, though, and that motivation's not there. You need to be worried about stability without preventing innovation. A few years back a quite large part of it was solved for Suns by My Evil Twin, David J. Brown: a two-part version number for every interface in the libraries, checked at link time.
--daveIt specifically controls the criminal law, which in turn controls what one citizen can do to another, by defining certain actions as criminal. Amoung these are breaking and entering and the malicious destruction of property.
--dave[Some folks who've never read history (or Ayn Rand!) keep making this "all law is contract law" claim. In fact, contract law came into existance long after criminal... try looking up "hanseatic league" or "emergent law". Sheesh!]
The organizers usually demand you provide
powerpoint files... I had to borrow a
windows laptop to give a Linux presentation
at a conference last year!
This is utterly typical behavior of someone who's convinced themselves they're "right". A parallel kind of behavior is the tendancy of professional criminals to get caught on traffic violations: the get blase' about breaking laws and then do something dumb right in front of a traffic cop. Both are variations on the psycology that makes "time wounds all heels" true (;-))
I'm using 6 beta, and it's better than
5.x and less buggy: the only things I
can break are broken in 5.2 as well.
Don't worry, it's a microbe, not a virus (:-))
--dave
I quite disagree: all sorts of open source offerings were written on top of closed source platforms, notably Unix operating systems, by writing to stable and published interfaces. For example, Emacs!
I also think Mr. Coward (Noel?) is confusing closed source platforms (eg, OSs, runtimes) and languages (eg, C, Java).
Finally, he suggests that closed source is a virus: if that were true, the converse would be true, and Microsoft would be "justified" in trying to prevent the use of open source operating systems and software with their closed-source OSs... for fear it will take over the minds of their lawyers.
I, for one, am not prepared to agree to that kind of arguement: I suspect Noel's not going to want to either!
--dave
With modern cargo aircraft and shiping, a force operating inside Afganistan in support of the Northern Alliance will be able to use light armour, self-propelled artillery and wheeled vehicles in offensive operations without worrying about:
- running out of ammunition
- freezing to death
- running out of gas in mid-battle.
And the core force will arguably be a line infantry unit of moderate size, with arty and some integral anti-tank capability. I'd guess at a reinforced brigade group. Anything larger would tend to trip ove its own feet, if not those of the Alliance.
I wonder if there are equally capable committees or subcomittees in the house or senate, that are good forums for hard problems...
In Canada, where our Senate is appointed for life (and is "called the house of cronies" by the sarcastic), there are some very capable standing comittees, but they have limited powers to write legislation...
In effect, I'm asking where in the U.S. government one applies pressure by bringing forward facts, not just opinion or numbers...
I quite agree: I can influence the
IETF, but it's hard to influence
a formal government. Suggestions
would be appreciated!
(And yes, I have written to U.S.
Sentors of border states)
Sorry, but you're labouring under a misapprehension: the entire english-speaking world works from precedent. It's inherited from Britain, the same place the U.S. inherited its common law from.
For the record, a univeristy education pays off in dollars.
The things I expected I'd never use, I use every day. The things I studied with intention to use, I never did. A friend who went to community college to get training in programming... makes exactly half what I do per year!
So now I'm a well-rounded (fat???), well-paid performance engineer at a three-letter-name company, having a heck of a good time and getting to tease the people who thought a philosophy degree would lead to a job driving taxi...
And I write nerd books on the side, which I never could have done with the grasp of english I had before going to university.
--dave
I disagree: You need to distinguish betwen the public action of mailing or emailing a letter, and the private contents of the letter. This is well understood in business: to quote Shanti Atkins (a business connsultant) speaking in Wired "if an employee is led to expect something is private, such as e-mail communications, then that privacy cannot be violated. But, if the company informs its employees that, for example, e-mail sent over the company's network is monitored, then the employee can no longer claim an "expectation of privacy." In short, once the company stakes its claim over its cyber-dominion, its employees have no right to privacy there. I expect the fact of the mailing to be public, and the contents to be private. I do have some duties, though: sealing the message in an envelope is more secure than sending a postcard, and encrypting it is stronger still. But that's a different discussion...
The interesting ones are the Apache Software Foundation and O'Reilly and Associates, who are likely to be concerned about all our concerns. One produces a certain heavily-used program which could use or reject untrustworthy suppliers, and the other might well publicise misbehavior...
I think the major technical stumbling block to allowing applets or scripting in any language is finding a language-neutral way of restricting execution along the lines of Java's security model.
I agree, but building such security regiemes is well within the state of the art. It was state-of-practice in Multics, used to control sysadmin.daemon's access to your files for backup purposes.
Although it fell out of use with the creation of Unix, it's slowly coming back into our consciousness once more.
You need a set of access controls on what a particular "surrogate user" (the entension) is allowed to use, a way of authenticating the user such as a digital signature, and a way of associating the two. I can easily see
/tmp/$$.$Z/* Microsoft.Passport rwe /tmp and a config file in my home directory.
(Although Passport would probably want rwe on root (;-))
$HOME/.passport Microsoft.Passport r
giving a downloaded MS Passport extension access to a per-process, per-extension directory in
Have a look for "mandatory access" (MAC) and "role-based access control" (RBAC) in sites on Java security and the NSA's secure Linux.
--daveThey wish a scalable OS with location
transparency, which implies small and bounded
times independant of location. This little
problem was satirized by a physicist friend[1],
who once explained to an overenthusiastic
customer that "the system will perform as you
desire only if we can increase the speed
of light".
--dave
[1. Russel Crook]
Our American Cousins have a tradition of taking action, solving their problems and then putting their lives and country back together. It might have to do with their first president rejecting the crown and going home after the revolutionary war...
--dave
[Canada declared martial law in October 1970 after an internal terrorist attack on Pierre Laporte, Quebec's labour and immigration minister and James Cross, Britain's trade commissioner. About a month later, those who had kidnapped and killed Laporte were arrested, and the innocent victims of the martial law period were reimbursed.
The problem is a particular instance of the general "continuous maintenance problem", which is now reasonably well understood (i.e., it's not academically interesting any more). This has been solved in particular cases as far back as Multics and, more topically, in the relational database world.
I have one of Paul Stachour's papers on the subject, although it needs to be OCR'd...
What's a suitable place/list to discuss this?
dave (formerly DRBrown.TSDC@Hi-Multics./ARPA) c-b