The FSF primary goal is *not* to create good software. It is to create *moral* software - software for goodneighbourliness and sharing - the fact that it is good (high quality/few bugs) - is a welcome - but secondary effect.
FSF's beef with Microsoft is not that it produces poor software - but that it produces non-Free software.
Speaking as someone who spent 3 years trying (and failing) to do just that I can only agree with you - it is very hard - that's why people who have done it with particularly important systems have received Nobel Prizes (not all of them of course, these days (it *is* getting easier)).
These days I write automation programs to help these guys out. Much less frustrating:).
Sadly, I am a little late to see this posting, here here you go nevertheless....
Strangely, IMHO, missed by the slashdot editors (on second thoughts, perhaps I am not so surprised) and the article itself was the paper for which Peter Galbraith is famous. The paper is Lisp: Good News, Bad News, How to Win Big, which includes the section "The Rise of Worse is Better" which he wrote while at Lucid.
Peter Galbraith was respected and quoted by JWZ (or lucid/xemacs and Netscape fame).
It was the ideas in Worse is Better that ESR rehashed and become the Cathederal and the Bazaar.
i.e. Linux was developed using the "New Jersey" approach and GNU was developed using the MIT approach: The folowing passage illustrates this:
Two famous people, one from MIT and another from Berkeley (but working on Unix) once met to discuss operating system issues. The person from MIT was knowledgeable about ITS (the MIT AI Lab operating system) and had been reading the Unix sources. He was interested in how Unix solved the PC loser-ing problem. The PC loser-ing problem occurs when a user program invokes a system routine to perform a lengthy operation that might have significant state, such as IO buffers. If an interrupt occurs during the operation, the state of the user program must be saved. Because the invocation of the system routine is usually a single instruction, the PC of the user program does not adequately capture the state of the process. The system routine must either back out or press forward. The right thing is to back out and restore the user program PC to the instruction that invoked the system routine so that resumption of the user program after the interrupt, for example, re-enters the system routine. It is called PC loser-ing because the PC is being coerced into loser mode, where loser is the affectionate name for user at MIT.
The MIT guy did not see any code that handled this case and asked the New Jersey guy how the problem was handled. The New Jersey guy said that the Unix folks were aware of the problem, but the solution was for the system routine to always finish, but sometimes an error code would be returned that signaled that the system routine had failed to complete its action. A correct user program, then, had to check the error code to determine whether to simply try the system routine again. The MIT guy did not like this solution because it was not the right thing.
The New Jersey guy said that the Unix solution was right because the design philosophy of Unix was simplicity and that the right thing was too complex. Besides, programmers could easily insert this extra test and loop. The MIT guy pointed out that the implementation was simple but the interface to the functionality was complex. The New Jersey guy said that the right tradeoff has been selected in Unix: namely, implementation simplicity was more important than interface simplicity.
i.e. Linus used the "Worse is Better" method and RMS (ahem...:) did not, thus the GNU Kernel, however good it is is delayed somewhat while they Do The Right Thing.
I encourage you to read the whole of Good News, Bad News - it contains insightful material on things other than Lisp (I should declare an interest in that I am a scheme programmer).
Re:We are emotional and not rational??!!!
on
Halloween VII
·
· Score: 1
What the (MS) author presumably meant was that anyone who has actually read the GNU Manifesto (or its derivatives) and approves of it (and the ideas of sharing and good neighbourliness) are making an *emotional* decision - not rational.
Just as in many scientific experiment (and especially psychology experiments) it is very important to have (at least) one control.
A control is where the experiment is performed expect without the difference being tested for. The results of the control show (or not, as the case may be) that it is the thing being tested for that causes differences, not the experiment itself.
Here, there is no control - crappy experiment.
The participants could have just been scoring on "this is different to the unencoded track, therefore it must be worse".
So put a copy of the unencoded track as a test track and see if it gets marked down (and also, of course do NOT tell the participants that it is there).
Relatively safe that is if you calculate it on fatal_accidents/person/mile (as the industry does, of course).
However, if you calculate safety using fatal_accidents/person/journey (which is just as, if not more legitimate) then flying is about the same level as riding a motorcycle.
That would be surprising since he almost gave his life fighting for it in Spain (Homage to Catalonia). What Orwell was against was Stalinism, not socialism.
1984, if you have read it, is about what happens when the unions have been crushed.
Palladium + ISP snooping on customers without consent or knowledge and without a search warrent.
We are getting there.
Orwell was just wrong about the year.
(as you can tell by my name, I am no fan of Stalinism either).
Re:GNU/GPL/Linux were all carried along ...
on
The Stallman Factor
·
· Score: 1
Alright then I agree with you. Your comments did strike me as somewhat one-sided, which is why I replied.
GNU and Linux were mutually benificial - and remain so.
Re:Linux was carried along by a larger revolution
on
The Stallman Factor
·
· Score: 3, Insightful
You've got it arse about face...
With the ideas and coding contributions from Linus Torvalds, where would GNU be today?
Summary: Nature abhors a vacuum, something else would have filled the niche occupied by the Linux kernel.
Long version:
GNU would be about the same. We would be using a different kernel,
that's about it. Free kernels existed before Linux, free editors
before emacs. Linux did not create anything new, it was just more
convenient than the others. If Linux did not exist something
else would have filled the niche, attracted users, and thereby gained
more attention, maintenance, support, and use. Linux is good
but free kernels existed before and will exist after.
The real revolution was not the kernel, the real revolution was in
communications of ideas via the net and distribution of tools (aka the
operating system) via the net. Linux was at the right place at
the right time and road the net wave from it's early academic-oriented
days.
Free software and the sharing of code existed long before the
Linux. The GPL has a unique spin compared to older licenses but the
truth is few programmers really care about BSD'ish vs. GPL'ish. People
merely tend towards the license of OS they are using, or maybe the
GPL's viral nature has twisted some arms:-).
> It's not retarded. Anybody who has ported applications to various UNIX platforms knows that it can be non-trivial.
Quite right. In recent years much of the traffic on the Debian Hurd ML has been about package porting issues see for example Heimdal issues.
A new OS is a new partition, a new disk, totally new way of doing things
Actually, if you can spare a gig or so, you can make a file and make it a filesystem and install GNU/Hurd in that file (I believe that this is the way WINE works with DOS partitions?) (that is the plan anyway). So repartitioning will not be necessary to try out GNU/Hurd and everyone had a spare gig or two these days, right?:).
It is not really a new way of doing things, I found. Debian GNU/Hurd is as different to Debian GNU/Linux as Debian GNU/Linux is to RedHat Linux (roughly).
So, although not every package is ported without effort, many packages can be got working with a little effort. Many (most?) of the porting problems are because the packages are Linuxified and not strictly POSIX.
Actually, the concept of multiple Hurd-based computers was thought out long before Beowulf was a twinkle in Donald Becker's eye (in Hurd-speak, we talk about a "multicomputer" instead of a Beowulf cluster).
Also note that the Hurd is built with consideration of portability (witness the recent porting to the PowerPC architecture using the MkLinux kernel (i.e. the Hurd is not tied only to GNU Mach).
And also note the Hurd is build with consideration for SMP. The reason that there are not SMP Hurd-based computers right now is because the *microkernel* does not support SMP. When it does, then so, automatically, will the Hurd. There is discussion on the ML recently addressing this issue.
Despite Nusphere's comment, the point is that it is *not* about linking (perhaps Nusphere where trying to cloud the issue).
Nusphere lost the right to redistibute MySQL because they made a modified version without the source code for that version being available: "Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License."
The MySQL authors therefore have the right to stop Nusphere redistributing their software.
MySQL AB undoubtedly have their own reasons for being so stiff necked.
I like the dinosaur book and bought it years ago. My copy is battered and dog-eared now. I have not seen it mentioned in Slashdot before however.
Incidently, in the new (6th) edition, they never mention the Hurd and the chapter on Mach has been deleted and replaced by Linux. Sign of the times, heh. Andy eat your heart out:).
Slashdot Mirror
The FSF primary goal is *not* to create good software. It is to create *moral* software - software for goodneighbourliness and sharing - the fact that it is good (high quality/few bugs) - is a welcome - but secondary effect.
FSF's beef with Microsoft is not that it produces poor software - but that it produces non-Free software.
Speaking as someone who spent 3 years trying (and failing) to do just that I can only agree with you - it is very hard - that's why people who have done it with particularly important systems have received Nobel Prizes (not all of them of course, these days (it *is* getting easier)).
:).
These days I write automation programs to help these guys out. Much less frustrating
Yeah! Good idea! Let's stop buying products which include Transmeta chips so they almost go out of business!
Ummm....
Hang on a minute...
Strangely, IMHO, missed by the slashdot editors (on second thoughts, perhaps I am not so surprised) and the article itself was the paper for which Peter Galbraith is famous. The paper is Lisp: Good News, Bad News, How to Win Big, which includes the section "The Rise of Worse is Better" which he wrote while at Lucid.
Peter Galbraith was respected and quoted by JWZ (or lucid/xemacs and Netscape fame).
It was the ideas in Worse is Better that ESR rehashed and become the Cathederal and the Bazaar.
i.e. Linux was developed using the "New Jersey" approach and GNU was developed using the MIT approach: The folowing passage illustrates this:
i.e. Linus used the "Worse is Better" method and RMS (ahem...
I encourage you to read the whole of Good News, Bad News - it contains insightful material on things other than Lisp (I should declare an interest in that I am a scheme programmer).
What the (MS) author presumably meant was that anyone who has actually read the GNU Manifesto (or its derivatives) and approves of it (and the ideas of sharing and good neighbourliness) are making an *emotional* decision - not rational.
Pah.
November 5th is not Halloween...it's something else.
:)
What we have is is not Halloween VII, but Bonfire Night I.
Just as in many scientific experiment (and especially psychology experiments) it is very important to have (at least) one control.
A control is where the experiment is performed expect without the difference being tested for. The results of the control show (or not, as the case may be) that it is the thing being tested for that causes differences, not the experiment itself.
Here, there is no control - crappy experiment.
The participants could have just been scoring on "this is different to the unencoded track, therefore it must be worse".
So put a copy of the unencoded track as a test track and see if it gets marked down (and also, of course do NOT tell the participants that it is there).
Relatively safe that is if you calculate it on fatal_accidents/person/mile (as the industry does, of course).
However, if you calculate safety using fatal_accidents/person/journey (which is just as, if not more legitimate) then flying is about the same level as riding a motorcycle.
(Yikes).
Well, indeed. There are a whole host of things to worry about.
Those were the first 2 that I could think of at that moment. Other comments expand on the issue.
That would be surprising since he almost gave his life fighting for it in Spain (Homage to Catalonia). What Orwell was against was Stalinism, not socialism.
1984, if you have read it, is about what happens when the unions have been crushed.
Palladium + ISP snooping on customers without consent or knowledge and without a search warrent.
We are getting there.
Orwell was just wrong about the year.
(as you can tell by my name, I am no fan of Stalinism either).
Alright then I agree with you. Your comments did strike me as somewhat one-sided, which is why I replied.
GNU and Linux were mutually benificial - and remain so.
You've got it arse about face...
With the ideas and coding contributions from Linus Torvalds, where would GNU be today?
Summary: Nature abhors a vacuum, something else would have filled the niche occupied by the Linux kernel.
Long version:
GNU would be about the same. We would be using a different kernel, that's about it. Free kernels existed before Linux, free editors before emacs. Linux did not create anything new, it was just more convenient than the others. If Linux did not exist something else would have filled the niche, attracted users, and thereby gained more attention, maintenance, support, and use. Linux is good but free kernels existed before and will exist after.
The real revolution was not the kernel, the real revolution was in communications of ideas via the net and distribution of tools (aka the operating system) via the net. Linux was at the right place at the right time and road the net wave from it's early academic-oriented days.
Free software and the sharing of code existed long before the Linux. The GPL has a unique spin compared to older licenses but the truth is few programmers really care about BSD'ish vs. GPL'ish. People merely tend towards the license of OS they are using, or maybe the GPL's viral nature has twisted some arms :-).
The threshhold for Bill Gates is about $500. Anything less is not worth his time picking up.
> It's not retarded. Anybody who has ported applications to various UNIX platforms knows that it can be non-trivial.
:).
Quite right. In recent years much of the traffic on the Debian Hurd ML has been about package porting issues see for example Heimdal issues.
A new OS is a new partition, a new disk, totally new way of doing things
Actually, if you can spare a gig or so, you can make a file and make it a filesystem and install GNU/Hurd in that file (I believe that this is the way WINE works with DOS partitions?) (that is the plan anyway). So repartitioning will not be necessary to try out GNU/Hurd and everyone had a spare gig or two these days, right?
It is not really a new way of doing things, I found. Debian GNU/Hurd is as different to Debian GNU/Linux as Debian GNU/Linux is to RedHat Linux (roughly).
select() has indeed been a problem - e.g see this
So, although not every package is ported without effort, many packages can be got working with a little effort. Many (most?) of the porting problems are because the packages are Linuxified and not strictly POSIX.
Thanks for you interest in the Hurd.
Actually, the concept of multiple Hurd-based computers was thought out long before Beowulf was a twinkle in Donald Becker's eye (in Hurd-speak, we talk about a "multicomputer" instead of a Beowulf cluster).
Also note that the Hurd is built with consideration of portability (witness the recent porting to the PowerPC architecture using the MkLinux kernel (i.e. the Hurd is not tied only to GNU Mach).
And also note the Hurd is build with consideration for SMP. The reason that there are not SMP Hurd-based computers right now is because the *microkernel* does not support SMP. When it does, then so, automatically, will the Hurd. There is discussion on the ML recently addressing this issue.
Despite Nusphere's comment, the point is that it is *not* about linking (perhaps Nusphere where trying to cloud the issue).
Nusphere lost the right to redistibute MySQL because they made a modified version without the source code for that version being available: "Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License."
The MySQL authors therefore have the right to stop Nusphere redistributing their software.
MySQL AB undoubtedly have their own reasons for being so stiff necked.
Sounds a bit like Thomas Bushnell's Hurd design paper with the technicalities stripped out and made buzword compliant.
Yeah... well you know what happened when they made Samson get a haircut...
Don't Forget:
"We like the DMCA," says Jill Lesser, AOL Time Warner's senior vice president for domestic public policy.
By owning RedHat, AOL can force copy control into the heart of the most popular Linux distribution (and therefore onto millions of desktops).
(Just how long RedHat would remain most popular when they do so is an interesting question).
I like the dinosaur book and bought it years ago. My copy is battered and dog-eared now. I have not seen it mentioned in Slashdot before however.
:).
Incidently, in the new (6th) edition, they never mention the Hurd and the chapter on Mach has been deleted and replaced by Linux. Sign of the times, heh. Andy eat your heart out
With comments like this, I start to wonder if RMS really believes in Open Source.
:)
RMS does not believe in Open Source. He believes (if he believes in anything) in Free Software.
Where the shell is concerned, wouldn't it be better to have some OSS bundled into a commware product??
Before you start bashing RMS, get your facts right. What did he actually say? Don't rely on some mixed-up heresay of the previous poster.
I personally believe that he would not consider it a problem if the source code to bash is distributed. And why would Apple want to hide that?
"I can use Linux, it's got BASH!"
Just how mind-bogglingly ironic that would be?
Different Glorfindel, actually. (The first one died.)
Up early for some reason today, Tim?
Well, if it was me choosing an OS to organize a bunch of reindeer, I would have to trust to instinct and run with the Hurd.
:-)
(Groan: -1 Corny
Ah yes, so you did.
Such is the price we pay for anonymous cowardice.