I hope that OS-X *is* successful: it'll stop all the `What's the point of *BSD?' questions and the sourgrapes about Linux's success from the *BSD crowd.
I've been using reiserfs on my laptop for about 6 months (with the SuSE patch to kernel 2.2.14). A journalling filesystem is pretty much indispensable: I've probably had two dozen dirty shutdowns in that time, a couple during large file operations, and I would have hated to have been using ext2. The few bugs that have shown up seem pretty small fry compared to the risks of running fscks every week or so.
I wouldn't use it for a server at the moment, not until there are a few more dot-releases, but I'm using FreeBSD for my server in any case.
I think `hardening' a distribution is (partly) orthogonal to what TrustedBSD is up to: the TrustedBSD folks are aiming to provide tools to make it possible to ensure that a distribution satisfies a security policy, whilst Bastille is meant to check a given system for obvious holes. A Bastille project for a TrustedBSD system would make sense.
There seem to be a proliferating number of proposed extensions to *NIXes with ruleset-based mandatory access controls. Is standardisation important? What influence do you see of NSA's recently released `security enhanced linux' having on other systems (like that in TrustedBSD)?
The author makes two valid (but unoriginal points): firstly that trying to do everything with inheritance is bad (every non-novice OO programmer agrees), and secondly that you don't support for classes and objects to get the benefits of object orientation (but you need first-class functions and dynamic/existential/abstract types, obvious to the LISP world for over twenty years).
Object orientation makes it easier to manage changing interfaces. Support for object orientation makes it easier to achieve this in programming practice. The author seems so upset about bad designs he has encountered caused by an inflexible, deep inheretiance that he seems blind to the fact that OO support can facilitate good design.
I think the author has got the comparison between the NT users who switched and the Mac users wrong. I think he is about right with the Mac users, but I think very few organisations who ran MS-only shops switched to Linux. In my experience, the people who switched to Linux were people who were running heterogenous computer systems, were promised interoperability by Microsoft, and were delivered something rather different. Win2k changes very little from that point of view, and the advantages of Linux are: UNIX offers a better network glue for a heterogenous network, and open source means you don't get the `bait-and-switch' promises and double-speak of a proprietary vendor.
With.NET, what Microsoft means is that their MSIL (their Java byte-code like language), C# and some other technologies are being submitted to the ECMA standards organization (Note: This is more than Sun ever did with Java).
The submission of c# to a standards body is old news: I hadn't heard about MSIL. Which are the other technologies? Any APIs?
Which they couldn't do if they wren't broken up without very obviously being an anti-competitive cartel. The Economist article is just plain wrong: breaking up Microsoft cripples its `Embrace and Extend' strategy, with or without.NET.
If AMD are planning on redesigning the 32-bit emulation facilities for Sledgehammer, does that mean a later release date for the chip? How does that affect the likelihood of uptake for the new chip?
I don't know what to make about this story. My gut feeling is that AMD is casting about in its 64-bit strategy after not getting the support it had been counting on from Microsoft.
This is a specific exemption in the GPL and it applies only to libraries that are a part of the standard part of the OS (the convoluted section 3 again). It has been argued that Qt might fall under this: it isn't really plausible, and again, neither the FSF or Debian interpret the GPL this way.
You haven't followed the KDE/Qt fiasco in any detail. Formally the GPL does not permit combination with any license that applies further restrictions, as QPL did. The authors of some packages in the early KDE distributions made use of GPLed code and then released their code (which made use of Qt) under the GPL. Since they were not the only copyright owners, they were not able to waive enforcement of the clause (as, I believe, RMS has said that `pure' KDE code was able to).
I happen to believe that it is possible to interpret GPLv2 in a way that permits compatibility with QPL and other free-but-incompatible licenses: namely to say that the redistribution of source demanded by the GPL is to be interpreted as `separate free distributability' (the plausibility of my view comes from the fact that the GPLv2 does not directly require that the GPL is applied to the source redistributed, the relevant parts are sections 2 and 3, but this requirement is inferred from what I regard as ambiguous language in those sections). In fact, though, that is not how the FSF or Debian have interpreted it.
I agree with the point you're making, but the GPLv2 position is coherent: the whole of the source for all binaries that you distribute should be freely available. This makes sense: users don't really have free use of your code if part of the source of the code is commercially tied with a restrictive license.
I lost sympathy with the GPL over the KDE/Qt fiasco (especially in its second, QPL, manifestation). Here the whole of the source was available under a free license, *but* the free license happened to be incompatible with the GPL. OK, old story, I'll just say that I think the direction the GPL v3 should be going is to make it tolerant of other free licenses.
I don't see how the dynamic linking problem can be avoided without spoiling what makes GPLv2 coherent: namely its focus upon the actual distributed binary.
An excellent case for the irreducibility of visual and spatial intuitions to linguistic ones is made by John Etchemendy in his paper "Computers, visualisation and the nature of reasoning" (PDF file). John Etchemendy is a smart chap, a formal logician working at the Stanford CSLI, and a close colleague of the late Jon Barwise.
So the answer is `not much'. One of the striking features of this case is that it is one of the handful of anti-trust cases that Bork has actually supported.
I think I can't resist asking another question: Microsoft seem to be basing their appeal on the argument that Jackson showed persitent bias throughout the trial, confirmed by the lack of consultation over the proposed remedy. Do you think they have much chance of success with this?
Richard: Good to see you here (I remember your excellent contribution to the DOJ vs. MS trial analysis). If Bush *were* (I think it unlikely) to try to get Microsoft of the hook, realistically, what do you think his options are?
George Bush doesn't have the power to stop the case against Microsoft. The only thing he can accomplish (at a big political cost) is to withdraw the DOJ, but the State DAs (the coplaintiffs with the DOJ) would still pursue the case.
Bush has also stated in an interview (in April) that he would not interfere in this case. Remember that Reagan when he was elected to a much stronger political position, and with a much stronger views against anti-trust than Bush, did nothing to stop the breakup of AT&T.
I don't really hate advocacy. I just hate the way we do it most of the time. We do it in a dumb way. And I think the discoursive habits we pick up as a result are going to impede the progress of programming languages for a long time.
Let me see if I understand you: you are saying that the rule of law depends upon people regularly engaging in bullying, lynching and illegal intimidation? And that this follows from the second law of thermodynamics?
TeX macros are very hairy, but LaTeX defines a clean interface to them (via \newcommand, etc.) and encourages a separation between the `ugly stuff' that is hidden in the backroom of class files and the clean stuff that appears in the document. It is quite easy to forbid \def macros from appearing anywhere in a document.
I have to say I don't think XML will live up to your hopes for it: whilst there are a number of content representation formats that it is well suited for (3D graphics, vector graphics, business charts, UML graphs, etc), the ones that achieve mass acceptance will surely be corrupted by formatting markup in just the same way as HTML and Latex are now. I'm afraid that when it comes to document processing `worse is better'.
It's a myth that SuSE doesn't target server boxes: they do and they say so (and they won `best server solution' earlier this year from some linux conference or another). What they don't do is try to make support their core revenue stream, and they try to follow organic growth (ie. build on your profits) rather than the IPO lottery.
Today that's looking like a smart strategy, though I think Red Hat's troubles are overstated (one learns as little about the fundamentals of a business from share prices in the middle of a run on tech stocks as one does from them in the middle of an IPO gold rush).
Environmental regulation is costly, but my comment was directed at the `it doesn't do anything' argument: this announcement reports on evidence that supports the effectiveness of environmental regulation.
It is funny that the costs of environmental regulation proposed at Kyoto are so closely examined, but fewer people seriously question the more costly `need' for the US to be able to fight two major offensives in two theatres at the same time.
I really don't care what the likes of Rush Limbaugh think. I think this announcement is positive ammunition against the "Environmental regulation is costly and doesn't do anything anyway" brigade; these people have much more influence, I think. (I hope...)
Slashdot Mirror
I hope that OS-X *is* successful: it'll stop all the `What's the point of *BSD?' questions and the sourgrapes about Linux's success from the *BSD crowd.
SuSE patch to kernel 2.2.14). A journalling filesystem is pretty much
indispensable: I've probably had two dozen dirty shutdowns in that
time, a couple during large file operations, and I would have hated to
have been using ext2. The few bugs that have shown up seem pretty
small fry compared to the risks of running fscks every week or so.
I wouldn't use it for a server at the moment, not until there are a
few more dot-releases, but I'm using FreeBSD for my server in any
case.
I think `hardening' a distribution is (partly) orthogonal to what
TrustedBSD is up to: the TrustedBSD folks are aiming to provide tools
to make it possible to ensure that a distribution satisfies a security
policy, whilst Bastille is meant to check a given system for obvious
holes. A Bastille project for a TrustedBSD system would make sense.
There seem to be a proliferating number of proposed extensions to
*NIXes with ruleset-based mandatory access controls. Is
standardisation important? What influence do you see of NSA's
recently released `security enhanced linux' having on other systems
(like that in TrustedBSD)?
trying to do everything with inheritance is bad (every non-novice OO
programmer agrees), and secondly that you don't support for classes
and objects to get the benefits of object orientation (but you need
first-class functions and dynamic/existential/abstract types, obvious
to the LISP world for over twenty years).
Object orientation makes it easier to manage changing interfaces.
Support for object orientation makes it easier to achieve this in
programming practice. The author seems so upset about bad designs he
has encountered caused by an inflexible, deep inheretiance that he
seems blind to the fact that OO support can facilitate good design.
I think the author has got the comparison between the NT users who
switched and the Mac users wrong. I think he is about right with the
Mac users, but I think very few organisations who ran MS-only shops
switched to Linux. In my experience, the people who switched to Linux
were people who were running heterogenous computer systems, were
promised interoperability by Microsoft, and were delivered something
rather different. Win2k changes very little from that point of view,
and the advantages of Linux are: UNIX offers a better network glue for
a heterogenous network, and open source means you don't get the
`bait-and-switch' promises and double-speak of a proprietary vendor.
byte-code like language), C# and some other technologies are being
submitted to the ECMA standards organization (Note: This is more than
Sun ever did with Java).
The submission of c# to a standards body is old news: I hadn't
heard about MSIL. Which are the other technologies? Any APIs?
Which they couldn't do if they wren't broken up without very obviously being an anti-competitive cartel. The Economist article is just plain wrong: breaking up Microsoft cripples its `Embrace and Extend' strategy, with or without .NET.
Sledgehammer, does that mean a later release date for the chip? How
does that affect the likelihood of uptake for the new chip?
I don't know what to make about this story. My gut feeling is that
AMD is casting about in its 64-bit strategy after not getting the
support it had been counting on from Microsoft.
Company executives who knowingly break the law on their companies behalf are personally liable for their actions.
This is a specific exemption in the GPL and it applies only to
libraries that are a part of the standard part of the OS (the
convoluted section 3 again). It has been argued that Qt might fall
under this: it isn't really plausible, and again, neither the FSF or
Debian interpret the GPL this way.
GPL does not permit combination with any license that applies further
restrictions, as QPL did. The authors of some packages in the early
KDE distributions made use of GPLed code and then released their code
(which made use of Qt) under the GPL. Since they were not the only
copyright owners, they were not able to waive enforcement of the
clause (as, I believe, RMS has said that `pure' KDE code was able to).
I happen to believe that it is possible to interpret GPLv2 in a way
that permits compatibility with QPL and other free-but-incompatible
licenses: namely to say that the redistribution of source demanded by
the GPL is to be interpreted as `separate free distributability' (the
plausibility of my view comes from the fact that the GPLv2 does not
directly require that the GPL is applied to the source redistributed,
the relevant parts are sections 2 and 3, but this requirement is
inferred from what I regard as ambiguous language in those sections).
In fact, though, that is not how the FSF or Debian have interpreted
it.
Nicely put. Thanks again for replying.
coherent: the whole of the source for all binaries that you distribute
should be freely available. This makes sense: users don't really have
free use of your code if part of the source of the code is
commercially tied with a restrictive license.
I lost sympathy with the GPL over the KDE/Qt fiasco (especially in
its second, QPL, manifestation). Here the whole of the source was
available under a free license, *but* the free license happened to be
incompatible with the GPL. OK, old story, I'll just say that I think
the direction the GPL v3 should be going is to make it tolerant of
other free licenses.
I don't see how the dynamic linking problem can be avoided without
spoiling what makes GPLv2 coherent: namely its focus upon the actual
distributed binary.
An excellent case for the irreducibility of visual and spatial
intuitions to linguistic ones is made by John Etchemendy in his paper
"Computers, visualisation and the nature of reasoning" (PDF file).
John Etchemendy is a smart chap, a formal logician working at the
Stanford CSLI, and a close colleague of the late Jon Barwise.
case is that it is one of the handful of anti-trust cases that Bork
has actually supported.
I think I can't resist asking another question: Microsoft seem to
be basing their appeal on the argument that Jackson showed persitent
bias throughout the trial, confirmed by the lack of consultation over
the proposed remedy. Do you think they have much chance of success
with this?
Richard: Good to see you here (I remember your excellent contribution
to the DOJ vs. MS trial analysis). If Bush *were* (I think it
unlikely) to try to get Microsoft of the hook, realistically, what do
you think his options are?
The only thing he can accomplish (at a big political cost) is to
withdraw the DOJ, but the State DAs (the coplaintiffs with the DOJ)
would still pursue the case.
Bush has also stated in an interview (in April) that he would not
interfere in this case. Remember that Reagan when he was elected to a
much stronger political position, and with a much stronger views
against anti-trust than Bush, did nothing to stop the breakup of AT&T.
Conclusion
I don't really hate advocacy. I just hate the way we do it most
of the time. We do it in a dumb way. And I think the discoursive
habits we pick up as a result are going to impede the progress of
programming languages for a long time.
Let me see if I understand you: you are saying that the rule of law
depends upon people regularly engaging in bullying, lynching and
illegal intimidation? And that this follows from the second law of
thermodynamics?
TeX macros are very hairy, but LaTeX defines a clean interface to them
(via \newcommand, etc.) and encourages a separation between the `ugly
stuff' that is hidden in the backroom of class files and the clean
stuff that appears in the document. It is quite easy to forbid \def
macros from appearing anywhere in a document.
I have to say I don't think XML will live up to your hopes for it:
whilst there are a number of content representation formats that it is
well suited for (3D graphics, vector graphics, business charts, UML
graphs, etc), the ones that achieve mass acceptance will surely be
corrupted by formatting markup in just the same way as HTML and Latex
are now. I'm afraid that when it comes to document processing `worse
is better'.
say so (and they won `best server solution' earlier this year from
some linux conference or another). What they don't do is try to make
support their core revenue stream, and they try to follow organic
growth (ie. build on your profits) rather than the IPO lottery.
Today that's looking like a smart strategy, though I think Red
Hat's troubles are overstated (one learns as little about the
fundamentals of a business from share prices in the middle of a run on
tech stocks as one does from them in the middle of an IPO gold rush).
`it doesn't do anything' argument: this announcement reports on
evidence that supports the effectiveness of environmental regulation.
It is funny that the costs of environmental regulation proposed at
Kyoto are so closely examined, but fewer people seriously question the
more costly `need' for the US to be able to fight two major offensives
in two theatres at the same time.
I really don't care what the likes of Rush Limbaugh think. I think
this announcement is positive ammunition against the "Environmental
regulation is costly and doesn't do anything anyway" brigade; these
people have much more influence, I think. (I hope...)