I'd say the essence of microkernels is message-passing, which is a kind of abstraction of OS services that makes a lot of OS tasks easier to coordinate, like locking of devices. It should be covered in any good intro-to-OS-design course, and it's a bit like the `command' design pattern.
p.Linux isn't even modular: Linus has frequently resisted any attempt to standardise linux APIs, citing the need to allow the implementation to evolve. Linux's kernel modules have little to do with modularity: they are just a trick for switching bits of code in and out of kernel space.
Re:Damned if you do, Damned if you don't
on
Golden Rice
·
· Score: 2
Why don't we hear the same screaming and yelling about pesticides and fertilisers, which are *proven* to cause long-term damage to the environment compared to GM, where the degree of harm is mostly alleged, and where there is evidence, it is less.
Well, according to such a restrictive definition of microkernel (ie. there must be at least two context changes whenever you use a core service), there aren't many microkernels out there (Mach, Chorus, are there any others?), because context changes really are expensive, especially on an x86. I don't suppose that when people say they that microkernels won the OS design wars they meant that the OS must be doing endless context switches to process the most basic service...
If by a microkernel OS you mean that there is a principled distinction between core OS services and other OS services, with a well-defined message passing interface between the two, then NT is a microkernel.
The example is illuminating...but bug tolerance? It makes converting between formats easier, but if your code isn't what you meant, but works becuase of little-endianness then it makes bugs harder to spot, which in my book is bad.
I don't understand your post. Are you suggesting OSX supports the Win32 API? I'm afraid that's not the case: the best it does is it's virtual PC mode, which is more the equivalent of VMware, and isn't really all that interesting. I heard some talk of porting WINE to the OSX a while back, though...
I don't suppose that Apple needs to give away the source to OS X to get the support of the open source community: instead providing support and commitment for the equivalent of WINE for OS X on a free software platform (Linux and/or BSD) would be enough. This would do a lot to get support for the platform from the OS cumminty. Check out this article by Jordan Hubbard on why Apple might want to do this (though Jordan advocates Apple actually opening up the source).
Solaris scales great on Sun hardware. If you happen to actually like Solaris, and you have lots of Sun hardware, then no doubt about it, go for Solaris over FreeBSD.
We had some Sun folks here offering free Sun boxes. We took some boxes as clients, but my research group decided to stick with FreeBSD on Intel as our main server...
I think hey! was referring to Adam Smith's puzzle over why diamonds should be valued more highly than water, when the first is a dispensable trinket and the last is the most urgent necessity. The answer is that value has more to do with availability than usefulness.
I thought the point of the manifesto bit of the GPL was that software should be treated like a service and not a commodity. Surely MS is moving *closer* to the way that the Open Source community sees things...
Interestingly, ESR predicted (TMC) that smaller companies would be forced to move to subscription models, since otherwise offering technical support is like writing a blank cheque. The surprise is that MS is the first to make this a centrepiece of their strategy...
It makes sense to change to a subscription model if you are already a MS shop.
Sure. Most site placement experts rate a placement at Yahoo higher than a placement at DMoz, though, so that suggests Yahoo's reputation is worth more than all the others put together.
There are a small handful of folks there who do benchmarks I trust (eg. C't and Ars Technica). It isn't always so easy to do your own benchmarks since one needs access to all the sets of hardware you want to compare.
The complaint about Kaplan is not that he applied the DMCA, but that he constantly went for interpretations of the evidence that bent the case most in the plaintiff's favour. The stuff he said about the intent of the authors of DeCSS being piracy is simply rubbish.
I used to edit at dmoz (I resigned as editor cas just two days ago), and while I agree that there is a lot to be desired in terms of quality, I think there are examples of excellence in the Science hierarchy. Yahoo, by contrast, is usually poor quality in Science.
As for your analysis of what is wrong with dmoz, I couldn't agree more. It has also been getting worse quite quickly in recent months. Unfortunately, the rivals to it that I know of are not credible, the least bad being Dave Winer's ultra-lightweight HTML directories. Do you know of any better projects?
Dmoz has different strengths than Yahoo. For commercial listings, Yahoo is quite a bit better than DMoz, whilst for specialist areas (eg. in science and computing) DMoz absolutely trounces Yahoo.
DMoz has more `raw' listings than Yahoo, and is growing faster, but it's Business category is fraught with issues of editor abuse, and Yahoo has much more influence on the face of the net (yahoo.com is the most well known website, according to many surveys). It's going to be a long time before DMoz is seen as anything but second best for plain commercial listings.
Ouch! Sorry, Ian, I don't think you're ignorant. I stand by the content of my other post, though. Especially I think the point about longevity is important: if I want to keep some large and boring-but-important historical archives around, I don't need to win a beauty contest with MojoNation, and so I think it is better for this kind of application. Freenet deals with some free speech issues, but it doesn't deal with them all.
Ah. A number of myths. There is a bank in MojoNation, but it is only an arbitrage service for mojo, allowing you to convert between dollars and mojo. Anyone else can set up a rival arbitrage service. You don't ever need to interact with the bank, even to set up a server.
There is no central server either: anyone can run any service, and evryone has to do *something* to obtain mojo. The fundamental difference between MojoNation and Freenet is the different ways they seek to tackle the free rider problem. I think MojoNation does it in a way that better ensures longevity of unpopular data.
I suggest you check the FAQ to avoid spreading myths.
Mojo Nation deserves a plug. It has an ingenious solution to the freeloader problem, namely an internal currency system, which may make the system more scaleable than Freenet. Advogato also runs some good discussions of these issues.
You are still very confused. All governments that aren't monarchies are republics, and all governments that aren't republics are monrachies. End of story.
Slashdot Mirror
I'd say the essence of microkernels is message-passing, which is a
kind of abstraction of OS services that makes a lot of OS tasks easier
to coordinate, like locking of devices. It should be covered in any
good intro-to-OS-design course, and it's a bit like the `command'
design pattern.
p.Linux isn't even modular: Linus has frequently resisted any attempt
to standardise linux APIs, citing the need to allow the implementation
to evolve. Linux's kernel modules have little to do with modularity:
they are just a trick for switching bits of code in and out of kernel
space.
Why don't we hear the same screaming and yelling about pesticides and
fertilisers, which are *proven* to cause long-term damage to the
environment compared to GM, where the degree of harm is mostly
alleged, and where there is evidence, it is less.
(ie. there must be at least two context changes whenever you use a
core service), there aren't many microkernels out there (Mach,
Chorus, are there any others?), because context changes really are
expensive, especially on an x86. I don't suppose that when people say
they that microkernels won the OS design wars they meant that the OS
must be doing endless context switches to process the most basic
service...
If by a microkernel OS you mean that there is a principled
distinction between core OS services and other OS services, with a
well-defined message passing interface between the two, then NT is a
microkernel.
The example is illuminating...but bug tolerance? It makes converting
between formats easier, but if your code isn't what you meant, but
works becuase of little-endianness then it makes bugs harder to spot,
which in my book is bad.
I don't understand your post. Are you suggesting OSX supports the
Win32 API? I'm afraid that's not the case: the best it does is it's
virtual PC mode, which is more the equivalent of VMware, and isn't
really all that interesting. I heard some talk of porting WINE to the
OSX a while back, though...
Why not? It has a message passing architecture. What's missing?
I don't suppose that Apple needs to give away the source to OS X to
get the support of the open source community: instead providing
support and commitment for the equivalent of WINE for OS X on a free
software platform (Linux and/or BSD) would be enough. This would do a
lot to get support for the platform from the OS cumminty. Check out
this article by Jordan Hubbard on why Apple might want to do this
(though Jordan advocates Apple actually opening up the source).
Solaris, and you have lots of Sun hardware, then no doubt about it, go
for Solaris over FreeBSD.
We had some Sun folks here offering free Sun boxes. We took some
boxes as clients, but my research group decided to stick with FreeBSD
on Intel as our main server...
I think hey! was referring to Adam Smith's puzzle over why diamonds
should be valued more highly than water, when the first is a dispensable
trinket and the last is the most urgent necessity. The answer is that
value has more to do with availability than usefulness.
Personally, I am happy to see MS moving to a service model. It
makes the distinction between commercial software and open source more
natural.
I thought the point of the manifesto bit of the GPL was that software
should be treated like a service and not a commodity. Surely MS is
moving *closer* to the way that the Open Source community sees
things...
forced to move to subscription models, since otherwise offering
technical support is like writing a blank cheque. The surprise is
that MS is the first to make this a centrepiece of their strategy...
It makes sense to change to a subscription model if you are already
a MS shop.
Sure. Most site placement experts rate a placement at Yahoo higher
than a placement at DMoz, though, so that suggests Yahoo's reputation
is worth more than all the others put together.
There are a small handful of folks there who do benchmarks I trust
(eg. C't and Ars Technica). It isn't always so easy to do your own
benchmarks since one needs access to all the sets of hardware you want
to compare.
The complaint about Kaplan is not that he applied the DMCA, but that
he constantly went for interpretations of the evidence that bent the
case most in the plaintiff's favour. The stuff he said about the
intent of the authors of DeCSS being piracy is simply rubbish.
and while I agree that there is a lot to be desired in terms of
quality, I think there are examples of excellence in the Science
hierarchy. Yahoo, by contrast, is usually poor quality in Science.
As for your analysis of what is wrong with dmoz, I couldn't agree
more. It has also been getting worse quite quickly in recent months.
Unfortunately, the rivals to it that I know of are not credible, the
least bad being Dave Winer's ultra-lightweight HTML directories. Do
you know of any better projects?
Yahoo is quite a bit better than DMoz, whilst for specialist
areas (eg. in science and computing) DMoz absolutely trounces Yahoo.
DMoz has more `raw' listings than Yahoo, and is growing faster, but
it's Business category is fraught with issues of editor abuse, and
Yahoo has much more influence on the face of the net (yahoo.com is the
most well known website, according to many surveys). It's going to be
a long time before DMoz is seen as anything but second best for plain
commercial listings.
No, it did them. Which idiot moderated the above post `insightful'? He didn't even read the slashdot banner, let alone the story.
Ouch! Sorry, Ian, I don't think you're ignorant. I stand by the
content of my other post, though. Especially I think the point about
longevity is important: if I want to keep some large and
boring-but-important historical archives around, I don't need to win a
beauty contest with MojoNation, and so I think it is better for this
kind of application. Freenet deals with some free speech issues, but
it doesn't deal with them all.
an arbitrage service for mojo, allowing you to convert between dollars
and mojo. Anyone else can set up a rival arbitrage service. You
don't ever need to interact with the bank, even to set up a server.
There is no central server either: anyone can run any service, and
evryone has to do *something* to obtain mojo. The fundamental
difference between MojoNation and Freenet is the different ways they
seek to tackle the free rider problem. I think MojoNation does it in
a way that better ensures longevity of unpopular data.
I suggest you check the FAQ to avoid spreading myths.
Umm, no. What's hotline? And how did you get a negative Karma? You're not a troll!
Mojo Nation deserves a plug.
It has an ingenious solution to the freeloader problem, namely an
internal currency system, which may make the system more scaleable
than Freenet. Advogato also
runs some good discussions of these issues.
You are still very confused. All governments that aren't monarchies
are republics, and all governments that aren't republics are
monrachies. End of story.
Umm, countries that aren't republics are monarchies. Aristocratic
government is the opposite of mob rule...
The interoperability promise, which I take to be the difference
between an operating system and a platform, is vaporware.