implementing reiserfs in C++ for a new OS

implementing reiserfs in C++ for a new OS

[Ramachandra K]

Hello,

I am planning to implement ReiserFS on my operating system named Capital 
(www.mitpune.com/research/capital1.html). To give a brief background - 
Capital is a 32-bit Object Oriented operating system for the Intel
i386 range of
microprocessors. Capital's features include multithreading,virtual memory 
and a VFS interface.

The implementation language of Capital is C++ and hence I, along with my
team, would like to implement ReiserFS from scratch in C++. I know that
this is a daunting task requiring considerable effort. The main motivation,
in addition to implementing an efficient file system for my operating system,
is to acquire an in depth understanding of the principles behind the design
of ReiserFS and possibly come up with some new ideas.

It would be of great help if I can get some pointers about how do I go about
understanding ReiserFS.

Currently I have got a bird's eye view of ReiserFS (plugins, items,
dancing trees etc).
But I am still grappling with the following issues:

1)I have not been able to get any information about the on-disk structure
(i.e where is the superblock, bitmaps etc). I guess I can get it by
reading the source
 (probably mkreiserfs code would also be helpful). Is there any other
source of information ?

2)All the information that is available at namesys.com is about Reiser4 and
I would like to know where I could get more information about version 3.

I would greatly appreciate suggestions and pointers to more information.

(Please CC the replies to me as I am currently not subscribed to the
mailing list)

rgds
Ram

Re: implementing reiserfs in C++ for a new OS

[Redeeman]

i suggest implementing reiser4 instead

On Sat, 2004-08-07 at 18:41 +0530, Ramachandra K wrote:
> Hello,
> 
> I am planning to implement ReiserFS on my operating system named Capital 
> (www.mitpune.com/research/capital1.html). To give a brief background - 
> Capital is a 32-bit Object Oriented operating system for the Intel
> i386 range of
> microprocessors. Capital's features include multithreading,virtual memory 
> and a VFS interface.
> 
> The implementation language of Capital is C++ and hence I, along with my
> team, would like to implement ReiserFS from scratch in C++. I know that
> this is a daunting task requiring considerable effort. The main motivation,
> in addition to implementing an efficient file system for my operating system,
> is to acquire an in depth understanding of the principles behind the design
> of ReiserFS and possibly come up with some new ideas.
> 
> It would be of great help if I can get some pointers about how do I go about
> understanding ReiserFS.
> 
> Currently I have got a bird's eye view of ReiserFS (plugins, items,
> dancing trees etc).
> But I am still grappling with the following issues:
> 
> 1)I have not been able to get any information about the on-disk structure
> (i.e where is the superblock, bitmaps etc). I guess I can get it by
> reading the source
>  (probably mkreiserfs code would also be helpful). Is there any other
> source of information ?
> 
> 2)All the information that is available at namesys.com is about Reiser4 and
> I would like to know where I could get more information about version 3.
> 
> I would greatly appreciate suggestions and pointers to more information.
> 
> (Please CC the replies to me as I am currently not subscribed to the
> mailing list)
> 
> rgds
> Ram
> 
-- 
Redeeman <redeeman@metanurb.dk>

Re: implementing reiserfs in C++ for a new OS

[Dr. Giovanni A. Orlando]

Redeeman wrote:

Hi,

    I suppose you are using GNU/Linux ... and therefore, you don't need 
to re-write nothing
    in C++, because is connected with the kernel and the kernel had been 
written in C, not in C++.

    Second, the name is a little 'comic' for me, if it is OpenSource.

    If you will adopt a Kernel different than Linux ... good luck.

Thanks,
Giovanni

>i suggest implementing reiser4 instead
>
>On Sat, 2004-08-07 at 18:41 +0530, Ramachandra K wrote:
>  
>
>>Hello,
>>
>>I am planning to implement ReiserFS on my operating system named Capital 
>>(www.mitpune.com/research/capital1.html). To give a brief background - 
>>Capital is a 32-bit Object Oriented operating system for the Intel
>>i386 range of
>>microprocessors. Capital's features include multithreading,virtual memory 
>>and a VFS interface.
>>
>>The implementation language of Capital is C++ and hence I, along with my
>>team, would like to implement ReiserFS from scratch in C++. I know that
>>this is a daunting task requiring considerable effort. The main motivation,
>>in addition to implementing an efficient file system for my operating system,
>>is to acquire an in depth understanding of the principles behind the design
>>of ReiserFS and possibly come up with some new ideas.
>>
>>It would be of great help if I can get some pointers about how do I go about
>>understanding ReiserFS.
>>
>>Currently I have got a bird's eye view of ReiserFS (plugins, items,
>>dancing trees etc).
>>But I am still grappling with the following issues:
>>
>>1)I have not been able to get any information about the on-disk structure
>>(i.e where is the superblock, bitmaps etc). I guess I can get it by
>>reading the source
>> (probably mkreiserfs code would also be helpful). Is there any other
>>source of information ?
>>
>>2)All the information that is available at namesys.com is about Reiser4 and
>>I would like to know where I could get more information about version 3.
>>
>>I would greatly appreciate suggestions and pointers to more information.
>>
>>(Please CC the replies to me as I am currently not subscribed to the
>>mailing list)
>>
>>rgds
>>Ram
>>
>>    
>>


-- 

-- 

--
Check FT Websites ... 
http://www.futuretg.com  - ftp://ftp.futuretg.com
http://www.FTLinuxCourse.com
	http://www.FTLinuxCourse.com/Certification
http://www.rpmparadaise.org
http://GNULinuxUtilities.com
http://www.YourPersonalOperatingSystem.com

--

Re: implementing reiserfs in C++ for a new OS

[Redeeman]

On Sat, 2004-08-07 at 17:05 +0200, Dr. Giovanni A. Orlando wrote:
> Redeeman wrote:
> 
> Hi,
> 
>     I suppose you are using GNU/Linux ... and therefore, you don't need 
> to re-write nothing
>     in C++, because is connected with the kernel and the kernel had been 
> written in C, not in C++.
> 
>     Second, the name is a little 'comic' for me, if it is OpenSource.
> 
>     If you will adopt a Kernel different than Linux ... good luck.
i just suggested implementing reiser4, i said nothing about c++ :)
> 
> Thanks,
> Giovanni
> 
> >i suggest implementing reiser4 instead
> >
> >On Sat, 2004-08-07 at 18:41 +0530, Ramachandra K wrote:
> >  
> >
> >>Hello,
> >>
> >>I am planning to implement ReiserFS on my operating system named Capital 
> >>(www.mitpune.com/research/capital1.html). To give a brief background - 
> >>Capital is a 32-bit Object Oriented operating system for the Intel
> >>i386 range of
> >>microprocessors. Capital's features include multithreading,virtual memory 
> >>and a VFS interface.
> >>
> >>The implementation language of Capital is C++ and hence I, along with my
> >>team, would like to implement ReiserFS from scratch in C++. I know that
> >>this is a daunting task requiring considerable effort. The main motivation,
> >>in addition to implementing an efficient file system for my operating system,
> >>is to acquire an in depth understanding of the principles behind the design
> >>of ReiserFS and possibly come up with some new ideas.
> >>
> >>It would be of great help if I can get some pointers about how do I go about
> >>understanding ReiserFS.
> >>
> >>Currently I have got a bird's eye view of ReiserFS (plugins, items,
> >>dancing trees etc).
> >>But I am still grappling with the following issues:
> >>
> >>1)I have not been able to get any information about the on-disk structure
> >>(i.e where is the superblock, bitmaps etc). I guess I can get it by
> >>reading the source
> >> (probably mkreiserfs code would also be helpful). Is there any other
> >>source of information ?
> >>
> >>2)All the information that is available at namesys.com is about Reiser4 and
> >>I would like to know where I could get more information about version 3.
> >>
> >>I would greatly appreciate suggestions and pointers to more information.
> >>
> >>(Please CC the replies to me as I am currently not subscribed to the
> >>mailing list)
> >>
> >>rgds
> >>Ram
> >>
> >>    
> >>
> 
> 
> -- 
> 
-- 
Redeeman <redeeman@metanurb.dk>

Re: implementing reiserfs in C++ for a new OS

[Hans Reiser]

Ramachandra K wrote:

>Hello,
>
>I am planning to implement ReiserFS on my operating system named Capital 
>(www.mitpune.com/research/capital1.html). To give a brief background - 
>Capital is a 32-bit Object Oriented operating system for the Intel
>i386 range of
>microprocessors. Capital's features include multithreading,virtual memory 
>and a VFS interface.
>
>The implementation language of Capital is C++ and hence I, along with my
>team, would like to implement ReiserFS from scratch in C++. I know that
>this is a daunting task requiring considerable effort. The main motivation,
>in addition to implementing an efficient file system for my operating system,
>is to acquire an in depth understanding of the principles behind the design
>of ReiserFS and possibly come up with some new ideas.
>
>It would be of great help if I can get some pointers about how do I go about
>understanding ReiserFS.
>
>Currently I have got a bird's eye view of ReiserFS (plugins, items,
>dancing trees etc).
>But I am still grappling with the following issues:
>
>1)I have not been able to get any information about the on-disk structure
>(i.e where is the superblock, bitmaps etc). I guess I can get it by
>reading the source
> (probably mkreiserfs code would also be helpful). Is there any other
>source of information ?
>  
>
Just the source.  I don't see why reading the source is difficult.  It 
should be easy to convert V4 C to C++ (C++ to C would be harder.....)

>2)All the information that is available at namesys.com is about Reiser4 and
>I would like to know where I could get more information about version 3.
>  
>
Using V3 for a new OS is crazy.  Use V4.  V3 is obsolete.

>I would greatly appreciate suggestions and pointers to more information.
>
>(Please CC the replies to me as I am currently not subscribed to the
>mailing list)
>
>rgds
>Ram
>
>
>  
>
Is it a free OS, or do you need to buy a license from us in addition to 
the GPL?

Re: implementing reiserfs in C++ for a new OS

[Ramachandra K]

Hans Reiser <reiser@namesys.com> wrote:

> Is it a free OS, or do you need to buy a license from us in
> addition to  the GPL?
> 

ReiserFS would be implemented FROM SCRATCH for Capital
in C++ with the help of existing literature/documentation etc.
By this I mean that no existing code would be used, that is
ported or modified in any way. This should mean that GPL 
would not be applicable to the code written for my OS, Capital.

Is my understanding of the license correct ? Are there any
other licenses that I should be aware of ?

Regarding Capital OS, the source would be made available,
but under a different license and not GPL.

And thanks for all the information.

rgds,
Ram

Re: implementing reiserfs in C++ for a new OS

[Dr. Giovanni A. Orlando]

Ramachandra K wrote:

>Hans Reiser <reiser@namesys.com> wrote:
>
>  
>
>>Is it a free OS, or do you need to buy a license from us in
>>addition to  the GPL?
>>
>>    
>>
>
>ReiserFS would be implemented FROM SCRATCH for Capital
>in C++ with the help of existing literature/documentation etc.
>By this I mean that no existing code would be used, that is
>ported or modified in any way. This should mean that GPL 
>would not be applicable to the code written for my OS, Capital.
>  
>
Opps. There are something wrong here.

Seems that Capital OS, will stole the ReiserFS idea and re-implement.

Very honest, very capital like.

Thanks,
Giovanni

>Is my understanding of the license correct ? Are there any
>other licenses that I should be aware of ?
>
>Regarding Capital OS, the source would be made available,
>but under a different license and not GPL.
>
>And thanks for all the information.
>
>rgds,
>Ram
>
>  
>


-- 

-- 

--
Check FT Websites ... 
http://www.futuretg.com  - ftp://ftp.futuretg.com
http://www.FTLinuxCourse.com
	http://www.FTLinuxCourse.com/Certification
http://www.rpmparadaise.org
http://GNULinuxUtilities.com
http://www.YourPersonalOperatingSystem.com

--

Re: implementing reiserfs in C++ for a new OS

[Marcelo Pacheco]

That's the dirtiest thing you can possibily do.

Em Seg 09 Ago 2004 11:52, Dr. Giovanni A. Orlando escreveu:
> Ramachandra K wrote:
> >Hans Reiser <reiser@namesys.com> wrote:
> >>Is it a free OS, or do you need to buy a license from us in
> >>addition to  the GPL?
> >
> >ReiserFS would be implemented FROM SCRATCH for Capital
> >in C++ with the help of existing literature/documentation etc.
> >By this I mean that no existing code would be used, that is
> >ported or modified in any way. This should mean that GPL
> >would not be applicable to the code written for my OS, Capital.
>
> Opps. There are something wrong here.
>
> Seems that Capital OS, will stole the ReiserFS idea and re-implement.
>
> Very honest, very capital like.
>
> Thanks,
> Giovanni
>
> >Is my understanding of the license correct ? Are there any
> >other licenses that I should be aware of ?
> >
> >Regarding Capital OS, the source would be made available,
> >but under a different license and not GPL.
> >
> >And thanks for all the information.
> >
> >rgds,
> >Ram
>
> --

Re: implementing reiserfs in C++ for a new OS

[mjt]

On Mon, Aug 09, 2004 at 04:52:24PM +0200, Dr. Giovanni A. Orlando wrote:

>Opps. There are something wrong here.

No, there is not.

>Seems that Capital OS, will stole the ReiserFS idea and re-implement.

No, Capital OS is _re-implementing_ a free (in every sense of the word)
file system. Is that so very different from implementing a free Start
button in your window manager? Re-implementing a science-fiction story
in a fantasy world? Re-implementing some other country's political party
in your own?

There is no theft here.

>Very honest, very capital like.

And exactly what is the loss to Namesys?

Yes, Hans Reiser thought up good ideas for a file system and implemented
them. This was work.

Then someone else looks at his work and says that it could be implemented
somewhere else with some work. Thus Hans Reiser just gave himself and
his company a chance to get more fame from people who port it to other
operating systems, not to mention possible support and helpdesk income
from helping with this porting.

The only reason you think it's dishonest is the fact that Namesys would
not get money from this project?
Maybe they will, maybe they won't, but they get fame and that always
carries a bit.

Maybe one of the reasons Reiserfs is being ported here instead of NTFS
is the fact that it's free. It's just every bit as OK to re-implement NTFS.
Hans gave people this chance to (easier) re-implement Reiserfs, contributing
a lot to society, and no one should piss on the people who take advantage
of this contribution. How else would you expect this show to go on?

To the original poster:
Implement Reiser4 instead, it's far superior.

-- 
mjt

Re: implementing reiserfs in C++ for a new OS

[Dr. Giovanni A. Orlando]

Markus Törnqvist wrote:

For me is not honest.

At first why they send an email to reiserfs-list?. Because they need 
support and they need additional help, and they don't
have the sufficient background to run the changes.

If they have the background torun the changes, or ... an honest move is:

Re-write in C++, and release in GPL mode sharing with ReiserFS community.

I feel, the commandement "We will do" ...like "We will open fire. It is 
not our problem, if you die".

... Again, to get benefits, and it is the same reason because GNU/Linux 
is better than other OS, it is because
people share information.

There are no more time in the world, to realize a perfect OS, contact an 
army of sales managers and become rich.

The Microsoft story never will be repeated.

My comments on this matter finish here.

Regards,
Giovanni

>On Mon, Aug 09, 2004 at 04:52:24PM +0200, Dr. Giovanni A. Orlando wrote:
>
>  
>
>>Opps. There are something wrong here.
>>    
>>
>
>No, there is not.
>
>  
>
>>Seems that Capital OS, will stole the ReiserFS idea and re-implement.
>>    
>>
>
>No, Capital OS is _re-implementing_ a free (in every sense of the word)
>file system. Is that so very different from implementing a free Start
>button in your window manager? Re-implementing a science-fiction story
>in a fantasy world? Re-implementing some other country's political party
>in your own?
>
>There is no theft here.
>
>  
>
>>Very honest, very capital like.
>>    
>>
>
>And exactly what is the loss to Namesys?
>
>Yes, Hans Reiser thought up good ideas for a file system and implemented
>them. This was work.
>
>Then someone else looks at his work and says that it could be implemented
>somewhere else with some work. Thus Hans Reiser just gave himself and
>his company a chance to get more fame from people who port it to other
>operating systems, not to mention possible support and helpdesk income
>from helping with this porting.
>
>The only reason you think it's dishonest is the fact that Namesys would
>not get money from this project?
>Maybe they will, maybe they won't, but they get fame and that always
>carries a bit.
>
>Maybe one of the reasons Reiserfs is being ported here instead of NTFS
>is the fact that it's free. It's just every bit as OK to re-implement NTFS.
>Hans gave people this chance to (easier) re-implement Reiserfs, contributing
>a lot to society, and no one should piss on the people who take advantage
>of this contribution. How else would you expect this show to go on?
>
>To the original poster:
>Implement Reiser4 instead, it's far superior.
>
>  
>


-- 

-- 

--
Check FT Websites ... 
http://www.futuretg.com  - ftp://ftp.futuretg.com
http://www.FTLinuxCourse.com
	http://www.FTLinuxCourse.com/Certification
http://www.rpmparadaise.org
http://GNULinuxUtilities.com
http://www.YourPersonalOperatingSystem.com

--

Re: implementing reiserfs in C++ for a new OS

[mjt]

On Mon, Aug 09, 2004 at 05:22:52PM +0200, Dr. Giovanni A. Orlando wrote:
>
>At first why they send an email to reiserfs-list?. Because they need 
>support and they need additional help, and they don't
>have the sufficient background to run the changes.

They asked for help, is that wrong?
I think asking for help is always a good thing, things get done faster
if people help each other.

>Re-write in C++, and release in GPL mode sharing with ReiserFS community.

They said the code will be free. Or at least available.
But as for reimplementing ideas, they technically wouldn't have to GPL it.
(OK, someone confirm this, but an implementation of a GPL-implementation
of an idea is hardly a derived work, yes?)

As far as I'm concerned, anyone who re-implements something like Reiserfs,
and sells it closed-source, because they didn't technically 
derive the work from Reiserfs, should have their testicles collected.

Also it would be a good idea to donate money to Namesys as a sign of good
will, especially if there is a significant amount of money in Capital OS.

But NOTHING says these things really have ANY effect, it's just how the
Capital OS guys are viewed by the community.

Every asshole has as much a right to free speech as anyone else.
Even if they'd have to shut up to pass for sages.

>I feel, the commandement "We will do" ...like "We will open fire. It is 
>not our problem, if you die".

That would be a lot more correct if they actually stole something.

>... Again, to get benefits, and it is the same reason because GNU/Linux 
>is better than other OS, it is because
>people share information.

Yes, Mr. Reiser et al shared their information.

The door swings both ways here.

-- 
mjt

Re: implementing reiserfs in C++ for a new OS

[Chris Dukes]

First the mildly tangential point.
Of all the languages other than 'C' to implement this in,
WHY 'C++'?  
On Mon, Aug 09, 2004 at 06:35:56PM +0300, Markus Törnqvist wrote:
> >Re-write in C++, and release in GPL mode sharing with ReiserFS community.
> 
> They said the code will be free. Or at least available.
> But as for reimplementing ideas, they technically wouldn't have to GPL it.
> (OK, someone confirm this, but an implementation of a GPL-implementation
> of an idea is hardly a derived work, yes?)

This could be construed as reverse engineering and may be covered under the
DMCA in the US.  This would be utterly insane and stupid path to take, but 
without as much penalty as say refilling and transporting a non-refillable
propane cylinder.
> 
> As far as I'm concerned, anyone who re-implements something like Reiserfs,
> and sells it closed-source, because they didn't technically 
> derive the work from Reiserfs, should have their testicles collected.

So, how do you feel about the closed source licenses for Reiserfs
that namesys.com sells?

-- 
Chris Dukes
Been there, done that, got the slightly-charred t-shirt. -- Crowder

Re: implementing reiserfs in C++ for a new OS

[mjt]

On Mon, Aug 09, 2004 at 06:01:45PM +0100, Chris Dukes wrote:
>So, how do you feel about the closed source licenses for Reiserfs
>that namesys.com sells?

Namesys can double-lincense as they want.

Anyway, I'd like some final clarity on this point, as Hans has said
that all is GPL, totally, but the README doesn't confirm this yet.

Just for the record, also, I would buy whatever closed-source stuff
Namesys sells, if it works and if it doesn't cause me more trouble than
it's worth, by, say, not giving free updates or requiring a certain
kernel with certain configurations.

Because it's the right thing to do.
They have the right to sell it and they need the money. 

But still I think ransomware is the best way to go if in a pinch,
there will always be a next feature that could be held as ransomware,
after enough money has been gathered.

-- 
mjt

Re: implementing reiserfs in C++ for a new OS

[Ramachandra K]

Let me clarify that the nature of Capital project is academic research.

rgds
Ram

Re: implementing reiserfs in C++ for a new OS

[Quinn Harris]

ReiserFS is a copyrighted work distributed under a GPL license and possibly 
others as the copyright holder Hans Reiser sees fit.  Copyright law protects 
an expression of an idea, as in the actual code, not the idea itself.  In 
other words, copyright law provides no legal means to restrict someone else 
from implementing the ideas in ReiserFS.  Patents on the other hand provide 
such power.  But I strongly doubt Hans holds any patents that would restrict 
the reimplementation of ReiserFS.

I believe it is your right to reimplement ReiserFS and I expect at least US 
law would back that up.  Despite this, I would question the cost 
effectiveness of doing so.  Reimplementing ReiserFS or any other file system 
is far from trivial.  You should strongly consider negotiating an alternative 
license for ReiserFS for a reasonable fee.  In doing so, you could have an 
excellent functional file system for Capitol OS in a very short time at a 
cost that could be subsantially lower than what it would take you to 
reimplement it.  It would also be much easier to transform a fully functional 
C code base into a C++ equivalent.

Considering that you would like to start from a fully functional file system 
design, I would assume that the file system design isn't a core part of the 
research in this project.  What I would expect you need is a good file system 
with a C++ interface to the rest of your kernel.  It should be vastly easier 
to create that interface on the existing code base than reimplement the 
entire file system in C++.

- Quinn


On Monday 09 August 2004 11:26 pm, Ramachandra K wrote:
> Let me clarify that the nature of Capital project is academic research.
>
> rgds
> Ram

Re: implementing reiserfs in C++ for a new OS

[Hans Reiser]

Quinn Harris wrote:

>ReiserFS is a copyrighted work distributed under a GPL license and possibly 
>others as the copyright holder Hans Reiser sees fit.  Copyright law protects 
>an expression of an idea, as in the actual code, not the idea itself. 
>
Yes, but it protects the expression not just in every literal word, but 
in the sense of it.  Look at the lawsuits over ripping off movie script 
writers.

Taking the essence of our expression, and transforming it into C++ and 
Indian, will not avoid infringement of our copyright.  Our documentation 
is as protected as  the source code, probably more so, because we 
haven't GPL'd everything on the website.

I don't really understand why he doesn't license it from us or adhere to 
the GPL if it is going to be free as he describes. 

Re: implementing reiserfs in C++ for a new OS

[mjt]

On Tue, Aug 10, 2004 at 12:51:26PM -0700, Hans Reiser wrote:
>Yes, but it protects the expression not just in every literal word, but 
>in the sense of it.  Look at the lawsuits over ripping off movie script 
>writers.

That happens mostly in the so-called land of the free, where movie writers
engage in pissing contests over who has the biggest asteroid or volcano.

Of course it's possible these ideas came simultaneously, and you can't
lock an idea down because people have the same ideas, it's just that
those guys are greedy as well and thus start shouting about intellectual
property infringment just to cover their industrial espionage.

>Taking the essence of our expression, and transforming it into C++ and 
>Indian, will not avoid infringement of our copyright.  Our documentation 
>is as protected as  the source code, probably more so, because we 
>haven't GPL'd everything on the website.

I have nothing but respect for you, your team and your work, but this
is a bit wrong.

Sure the documentation is copyrighted, but it's the actual documentation,
not the idea behind it, that is copyrighted.

>I don't really understand why he doesn't license it from us or adhere to 
>the GPL if it is going to be free as he describes. 

Me neither.

The point is, they don't have to pay anything if they just re-implement
it without creating a derived work.

They may be assholes if they don't. Maybe.

Also, I don't pretend to understand how stressful it must be being the
director of this project in this situation, but I don't see why this is
something to get an ulcer over. This is pretty much what freedom is about
and freedom's a damned good thing that a lot of people forgot about long ago.

But the bottom line is, it's not Namesys', or anyone's, loss if ReiserFS
gets reimplemented. Of course it can be seen as THEFT or something, but
it can also be seen as ADMIRATION. Either way, it's not out of your pocket.

On the contrary, it may be money donated out of good will or support requests
or whatever that comes in. Either way, you lose nothing.

There was a short discussion on the IRC channel about porting to Solaris.
I asked, out of academic interest, the philosophical question "Would you
pay for a Solaris port if it were made GPL?"

I got the obvious answer that the company would not pay for something they
could get for free, but the other side of the coin states that you can not
get it for free as it does not exist.

If it got implemented, it'd be foolish to do so under a license that prohibits
the created work from being relicensed.

Therefore, you have to pay for something that will ultimately be free.

Take enough examples like this, another company paying for a tailored
database system that goes GPL and the company who paid for Reiser4 for
Solaris gets that one for free. It evens out.

This may seem like a big digression, but the customer who pays for the Solaris
port doesn't really lose anything, he is willing to pay a sum of money
for a product, so he does it, and if other people get it free later he should
sleep well at night because he did a good thing, and by doing this he will
probably benefit from the open-source community reviewing the code constantly.

He just got the product he paid for and should all in all let it be.

Then the mandatory digression; Some time ago I bumped into the works of an
old economist, Bastiat[1] and as I've read some of his stuff, I find it
weird that he's pretty much saying what I've been saying for years.

Partly because of having seen pro-freedom stuff on the Namesys site, mind you.

But no one listened to him a 150 years ago, in fact, if he was as ignored
as they say (instead of him pissing a lot of people off), they must have
shrugged him off for saying the self-evident and then going on about
doing quite the reverse ;)

[1]
http://bastiat.org/
Well worth the reading time.

-- 
mjt

Re: implementing reiserfs in C++ for a new OS

[Hans Reiser]

Markus Törnqvist wrote:

>On Tue, Aug 10, 2004 at 12:51:26PM -0700, Hans Reiser wrote:
>  
>
>>Yes, but it protects the expression not just in every literal word, but 
>>in the sense of it.  Look at the lawsuits over ripping off movie script 
>>writers.
>>    
>>
>
>That happens mostly in the so-called land of the free, where movie writers
>engage in pissing contests over who has the biggest asteroid or volcano.
>
>Of course it's possible these ideas came simultaneously, and you can't
>lock an idea down because people have the same ideas, it's just that
>those guys are greedy as well and thus start shouting about intellectual
>property infringment just to cover their industrial espionage.
>  
>
No, if you walk into someone's office with a cute movie concept, you 
should not get ripped off, and you should not have to keep it secret 
until you have your own resources to produce the movie for fear that you 
will be.

>  
>
>>Taking the essence of our expression, and transforming it into C++ and 
>>Indian, will not avoid infringement of our copyright.  Our documentation 
>>is as protected as  the source code, probably more so, because we 
>>haven't GPL'd everything on the website.
>>    
>>
>
>I have nothing but respect for you, your team and your work, but this
>is a bit wrong.
>
>Sure the documentation is copyrighted, but it's the actual documentation,
>not the idea behind it, that is copyrighted.
>  
>
There is an idea, and then there is an expression of an idea.  
expression is not just the letters, or the words, it is somewhere 
between the exact characters and the idea.

>  
>
>>I don't really understand why he doesn't license it from us or adhere to 
>>the GPL if it is going to be free as he describes. 
>>    
>>
I think he is going to clarify it, he sent me a private email in which 
things make more sense, I'll let him say the rest.

Re: implementing reiserfs in C++ for a new OS

[Spam]

> Markus Törnqvist wrote:
>>On Tue, Aug 10, 2004 at 12:51:26PM -0700, Hans Reiser wrote:
>>  
>>
>>>Yes, but it protects the expression not just in every literal word, but
>>>in the sense of it.  Look at the lawsuits over ripping off movie script
>>>writers.
>>>    
>>>
<snip>
>>  
>>
>>>Taking the essence of our expression, and transforming it into C++ and
>>>Indian, will not avoid infringement of our copyright.  Our documentation
>>>is as protected as  the source code, probably more so, because we 
>>>haven't GPL'd everything on the website.
>>>    
>>>
>>
>>I have nothing but respect for you, your team and your work, but this
>>is a bit wrong.
>>
>>Sure the documentation is copyrighted, but it's the actual documentation,
>>not the idea behind it, that is copyrighted.
>>  
>>
> There is an idea, and then there is an expression of an idea.  
> expression is not just the letters, or the words, it is somewhere 
> between the exact characters and the idea.

  I  do  not  believe  that copyright protects ideas, only the written
  texts, graphics, and music.

  Take  as  an  example a inventor. If he has an idea that he wants to
  protect he would need to patent it to protect it. Otherwise only his
  work  can be protected with copyright. I am not sure that the actual
  invention can be copyrighted either.

  In  this  case  with  Hans,  Namesys and reiserfs/reiser4 the actual
  written  code  and  documentation  etc  is covered by copyright. The
  ideas  behind  the filesystems are not. They would only be protected
  by patents.

Re: implementing reiserfs in C++ for a new OS

[Hans Reiser]

Spam wrote:

>  
>
>>Markus Törnqvist wrote:
>>    
>>
>>>On Tue, Aug 10, 2004 at 12:51:26PM -0700, Hans Reiser wrote:
>>> 
>>>
>>>      
>>>
>>>>Yes, but it protects the expression not just in every literal word, but
>>>>in the sense of it.  Look at the lawsuits over ripping off movie script
>>>>writers.
>>>>   
>>>>
>>>>        
>>>>
><snip>
>  
>
>>> 
>>>
>>>      
>>>
>>>>Taking the essence of our expression, and transforming it into C++ and
>>>>Indian, will not avoid infringement of our copyright.  Our documentation
>>>>is as protected as  the source code, probably more so, because we 
>>>>haven't GPL'd everything on the website.
>>>>   
>>>>
>>>>        
>>>>
>>>I have nothing but respect for you, your team and your work, but this
>>>is a bit wrong.
>>>
>>>Sure the documentation is copyrighted, but it's the actual documentation,
>>>not the idea behind it, that is copyrighted.
>>> 
>>>
>>>      
>>>
>>There is an idea, and then there is an expression of an idea.  
>>expression is not just the letters, or the words, it is somewhere 
>>between the exact characters and the idea.
>>    
>>
>
>  I  do  not  believe  that copyright protects ideas, only the written
>  texts, graphics, and music.
>
>  Take  as  an  example a inventor. If he has an idea that he wants to
>  protect he would need to patent it to protect it. Otherwise only his
>  work  can be protected with copyright. I am not sure that the actual
>  invention can be copyrighted either.
>
>  In  this  case  with  Hans,  Namesys and reiserfs/reiser4 the actual
>  written  code  and  documentation  etc  is covered by copyright. The
>  ideas  behind  the filesystems are not. They would only be protected
>  by patents.
>
>
>
>
>  
>
If you translate a poem into french, it is still rightfully copyright 
protected by the original author.

But hey, these guys are just doing this work for academic purposes they 
say and not releasing it to the world, so I think they are not going to 
be violating the GPL, so it is all nothing to worry about.  Maybe this 
sort of thing makes for a good student assignment in their view, I don't 
know....

Hans

Re: implementing reiserfs in C++ for a new OS

[Nikita Danilov]

Hans Reiser writes:
 > Spam wrote:
 > 
 > >  
 > >
 > >>Markus Törnqvist wrote:
 > >>    
 > >>
 > >>>On Tue, Aug 10, 2004 at 12:51:26PM -0700, Hans Reiser wrote:
 > >>> 
 > >>>
 > >>>      
 > >>>
 > >>>>Yes, but it protects the expression not just in every literal word, but
 > >>>>in the sense of it.  Look at the lawsuits over ripping off movie script
 > >>>>writers.
 > >>>>   
 > >>>>
 > >>>>        
 > >>>>
 > ><snip>
 > >  
 > >
 > >>> 
 > >>>
 > >>>      
 > >>>
 > >>>>Taking the essence of our expression, and transforming it into C++ and
 > >>>>Indian, will not avoid infringement of our copyright.  Our documentation
 > >>>>is as protected as  the source code, probably more so, because we 
 > >>>>haven't GPL'd everything on the website.
 > >>>>   
 > >>>>
 > >>>>        
 > >>>>
 > >>>I have nothing but respect for you, your team and your work, but this
 > >>>is a bit wrong.
 > >>>
 > >>>Sure the documentation is copyrighted, but it's the actual documentation,
 > >>>not the idea behind it, that is copyrighted.
 > >>> 
 > >>>
 > >>>      
 > >>>
 > >>There is an idea, and then there is an expression of an idea.  
 > >>expression is not just the letters, or the words, it is somewhere 
 > >>between the exact characters and the idea.
 > >>    
 > >>
 > >
 > >  I  do  not  believe  that copyright protects ideas, only the written
 > >  texts, graphics, and music.
 > >
 > >  Take  as  an  example a inventor. If he has an idea that he wants to
 > >  protect he would need to patent it to protect it. Otherwise only his
 > >  work  can be protected with copyright. I am not sure that the actual
 > >  invention can be copyrighted either.
 > >
 > >  In  this  case  with  Hans,  Namesys and reiserfs/reiser4 the actual
 > >  written  code  and  documentation  etc  is covered by copyright. The
 > >  ideas  behind  the filesystems are not. They would only be protected
 > >  by patents.
 > >
 > >
 > >
 > >
 > >  
 > >
 > If you translate a poem into french, it is still rightfully copyright 
 > protected by the original author.

Copyright to result of translation belongs to the translator. At
least, according to Russian law.

 > 
 > But hey, these guys are just doing this work for academic purposes they 
 > say and not releasing it to the world, so I think they are not going to 
 > be violating the GPL, so it is all nothing to worry about.  Maybe this 
 > sort of thing makes for a good student assignment in their view, I don't 
 > know....
 > 
 > Hans
 > 

Nikita.

Re: implementing reiserfs in C++ for a new OS

[Hans Reiser]

Nikita Danilov wrote:

>
>Copyright to result of translation belongs to the translator. At
>least, according to Russian law.
>
>  
>
I may be completely wrong legally, esp. in Russia.  I personally think 
it should belong to the original author if the original copyright is 
still valid.

Otherwise harry potter can get translated and no need to pay rowling.

Hans

Re: implementing reiserfs in C++ for a new OS

[mjt]

On Wed, Aug 11, 2004 at 09:43:13AM -0700, Hans Reiser wrote:
>I may be completely wrong legally, esp. in Russia.  I personally think 
>it should belong to the original author if the original copyright is 
>still valid.

Of course it belongs to the original author.

But the translation is the translator's copyright.

So it's a shared situation, the translator doesn't take anything away
from the original author and does no wrong in translating unless he
publishes without the original author's consent.

-- 
mjt

Re: implementing reiserfs in C++ for a new OS

[Nikita Danilov]

Hans Reiser writes:
 > Nikita Danilov wrote:
 > 
 > >
 > >Copyright to result of translation belongs to the translator. At
 > >least, according to Russian law.
 > >
 > >  
 > >
 > I may be completely wrong legally, esp. in Russia.  I personally think 
 > it should belong to the original author if the original copyright is 
 > still valid.
 > 
 > Otherwise harry potter can get translated and no need to pay rowling.

`Otherwise' is non-sequitur. One pays royalties to get a permission to
produce and/or sell derived work. But copyright to derived work lies
with translator.
 
Otherwise, copyright to reiser4 (re-)implementation of allocate-on-flush
belongs to SGI.

 > 
 > Hans

Nikita.

Re: implementing reiserfs in C++ for a new OS

[Hans Reiser]

Nikita Danilov wrote:

>Hans Reiser writes:
> > Nikita Danilov wrote:
> > 
> > >
> > >Copyright to result of translation belongs to the translator. At
> > >least, according to Russian law.
> > >
> > >  
> > >
> > I may be completely wrong legally, esp. in Russia.  I personally think 
> > it should belong to the original author if the original copyright is 
> > still valid.
> > 
> > Otherwise harry potter can get translated and no need to pay rowling.
>
>`Otherwise' is non-sequitur. One pays royalties to get a permission to
>produce and/or sell derived work. But copyright to derived work lies
>with translator.
> 
>Otherwise, copyright to reiser4 (re-)implementation of allocate-on-flush
>belongs to SGI.
>  
>
I would feel more guilty about that if their team had not gone so far 
out of their way to encourage me to learn and use their most valuable 
innovation when I visited SGI.  I try to credit SGI on that generally, 
but folks are welcome to point out where an additional credit is needed 
(it is too easy to neglect such things).  I do think that I took their 
idea to the next level....

> > 
> > Hans
>
>Nikita.
>
>
>  
>

Re: implementing reiserfs in C++ for a new OS

[Dr. Giovanni A. Orlando]

Hans Reiser wrote:

> Nikita Danilov wrote:
>
>> Hans Reiser writes:
>> > Nikita Danilov wrote:
>> > > >
>> > >Copyright to result of translation belongs to the translator. At
>> > >least, according to Russian law.
>> > >
>> > >  > >
>> > I may be completely wrong legally, esp. in Russia.  I personally 
>> think > it should belong to the original author if the original 
>> copyright is > still valid.
>> > > Otherwise harry potter can get translated and no need to pay 
>> rowling.
>>
>> `Otherwise' is non-sequitur. One pays royalties to get a permission to
>> produce and/or sell derived work. But copyright to derived work lies
>> with translator.
>>
>> Otherwise, copyright to reiser4 (re-)implementation of allocate-on-flush
>> belongs to SGI.
>>  
>>
> I would feel more guilty about that if their team had not gone so far 
> out of their way to encourage me to learn and use their most valuable 
> innovation when I visited SGI.  I try to credit SGI on that generally, 
> but folks are welcome to point out where an additional credit is 
> needed (it is too easy to neglect such things).  I do think that I 
> took their idea to the next level....

There are no problems at all.

Both projects are GPL.

Thanks,
Giovanni

>
>> > > Hans
>>
>> Nikita.
>>
>>
>>  
>>
>


-- 

-- 

--
Check FT Websites ... 
http://www.futuretg.com  - ftp://ftp.futuretg.com
http://www.FTLinuxCourse.com
	http://www.FTLinuxCourse.com/Certification
http://www.rpmparadaise.org
http://GNULinuxUtilities.com
http://www.YourPersonalOperatingSystem.com

--

Re: implementing reiserfs in C++ for a new OS

[Nikita Danilov]

Hans Reiser writes:
 > Nikita Danilov wrote:
 > 
 > >Hans Reiser writes:
 > > > Nikita Danilov wrote:
 > > > 
 > > > >
 > > > >Copyright to result of translation belongs to the translator. At
 > > > >least, according to Russian law.
 > > > >
 > > > >  
 > > > >
 > > > I may be completely wrong legally, esp. in Russia.  I personally think 
 > > > it should belong to the original author if the original copyright is 
 > > > still valid.
 > > > 
 > > > Otherwise harry potter can get translated and no need to pay rowling.
 > >
 > >`Otherwise' is non-sequitur. One pays royalties to get a permission to
 > >produce and/or sell derived work. But copyright to derived work lies
 > >with translator.
 > > 
 > >Otherwise, copyright to reiser4 (re-)implementation of allocate-on-flush
 > >belongs to SGI.
 > >  
 > >
 > I would feel more guilty about that if their team had not gone so far 
 > out of their way to encourage me to learn and use their most valuable 
 > innovation when I visited SGI.  I try to credit SGI on that generally, 
 > but folks are welcome to point out where an additional credit is needed 
 > (it is too easy to neglect such things).  I do think that I took their 
 > idea to the next level....

That's orthogonal to the topic of this discussion. The question is:
did you ask SGI's permission to re-implement allocate on flush idea?
If no, why re-implementors of reiserfs in C++ should seek yours? I
guess, that they are pretty sure that they will took your ideas to the
next level... :-)

 > 
 > > > 
 > > > Hans
 > >

Nikita.

 > >
 > >
 > >  
 > >
 > 
 > 

Re: implementing reiserfs in C++ for a new OS

[Dr. Giovanni A. Orlando]

Nikita Danilov wrote:

>Hans Reiser writes:
> > Nikita Danilov wrote:
> > 
> > >Hans Reiser writes:
> > > > Nikita Danilov wrote:
> > > > 
> > > > >
> > > > >Copyright to result of translation belongs to the translator. At
> > > > >least, according to Russian law.
> > > > >
> > > > >  
> > > > >
> > > > I may be completely wrong legally, esp. in Russia.  I personally think 
> > > > it should belong to the original author if the original copyright is 
> > > > still valid.
> > > > 
> > > > Otherwise harry potter can get translated and no need to pay rowling.
> > >
> > >`Otherwise' is non-sequitur. One pays royalties to get a permission to
> > >produce and/or sell derived work. But copyright to derived work lies
> > >with translator.
> > > 
> > >Otherwise, copyright to reiser4 (re-)implementation of allocate-on-flush
> > >belongs to SGI.
> > >  
> > >
> > I would feel more guilty about that if their team had not gone so far 
> > out of their way to encourage me to learn and use their most valuable 
> > innovation when I visited SGI.  I try to credit SGI on that generally, 
> > but folks are welcome to point out where an additional credit is needed 
> > (it is too easy to neglect such things).  I do think that I took their 
> > idea to the next level....
>
>That's orthogonal to the topic of this discussion. The question is:
>did you ask SGI's permission to re-implement allocate on flush idea?
>If no, why re-implementors of reiserfs in C++ should seek yours? I
>guess, that they are pretty sure that they will took your ideas to the
>next level... :-)
>
>  
>
I don't think that a simply re-implementation of the Reiser code in C++,
will offer any benefit to anyone.

They are swim in a new sea (ReiserFS) and have no idea about what is.
C++ is not the language
generally used for command based utilities.

C++ is generally used to program with objects, for example the graphical
interface where buttons, menu
bar are objects, and also a library may be extrapoled to work with MacOS
X, MS Windows or GNU/Linux, using
the same code, like did Trolltech people.

The Linux kernel had been written in C, and therefore the reiserfs code
inside the kernel have no sense in another language.

Hans, instead invent the ReiserFS, and he is not a newbie in FileSystem
architecture.

Have no sense to request some code or any authorization to anyone
because but ReiserFS 3 and 4 are GPL, at the same level
than SGI XFS.

So, I suppose that Hans choose a particular routine inside XFS that is
used to solve a particular problem that probably the employees he hire
was not capable to solve.

Because the original works are available under GPL, this is just in the
sense and in the respect of the GPL license, Use the code, modify
and release.

Instead these guys, want to close the code, they will realize without to
understand really nothing about it.

This simply have no sense and their comment is a stupid comment.

They can do in C++ ... and? They can will rewrite the ReiserFS Utilities
in C++ and? Who will use? We? I don't thing.

The Capital comments are silly comments like a boy: "I will take the bus
to back home, and don't want to be disturbed".

For us, here nothing is changed. We are more happy to use ReiserFS 4.
What is important for us, is than will be 100% error free.

Otherwise I will contrib to solve the bugs following my expertise, also
taking some code from other GPL places. This is GNU/Linux!

Thanks,
Giovanni

> > 
> > > > 
> > > > Hans
> > >
>
>Nikita.
>
> > >
> > >
> > >  
> > >
> > 
> > 
>
>  
>


-- 

-- 

--
Check FT Websites ...
http://www.futuretg.com  - ftp://ftp.futuretg.com
http://www.FTLinuxCourse.com
	http://www.FTLinuxCourse.com/Certification
http://www.rpmparadaise.org
http://GNULinuxUtilities.com
http://www.YourPersonalOperatingSystem.com

--

Re: implementing reiserfs in C++ for a new OS

[Hans Reiser]

Dr. Giovanni A. Orlando wrote:

>
>
> Hans, instead invent the ReiserFS, and he is not a newbie in FileSystem
> architecture.

I used to be one.;-)

>
>
>
> So, I suppose that Hans choose a particular routine inside XFS that is
> used to solve a particular problem that probably the employees he hire
> was not capable to solve.

No, I never read the code, I just listened to them describe how they 
only allocate blocks at flush time and realized they were right.

Re: implementing reiserfs in C++ for a new OS

[Chris Dukes]

On Wed, Aug 11, 2004 at 09:43:13AM -0700, Hans Reiser wrote:
> Nikita Danilov wrote:
> 
> >
> >Copyright to result of translation belongs to the translator. At
> >least, according to Russian law.

Unless it was translated as a work for hire, then the contract is
generally that the copyright belongs to whoever hired the work.
> >
> > 
> >
> I may be completely wrong legally, esp. in Russia.  I personally think 
> it should belong to the original author if the original copyright is 
> still valid.

Hans,
Since you obviously have a stake in this intellectual property,
have you considered talking with your lawyer instead of giving us
your fantasies?
> 
> Otherwise harry potter can get translated and no need to pay rowling.

Not entirely.  Such a translation would be deemed a derivative work.
The copyright on the translated version belongs to the translator
(Unless the translator did it for pay).  However, distribution of that
derivative work may be controlled by copyright holder of the work
the translation was derived from (IE, rowling wouldn't be able to distribute
the translation, but may be able to prevent the distribution of the
translation).

Now, I've been deleting a lot of this crap but.
How do these folks intend to redo reiserfs in C++?  Is it a function by
function translation to C++, or do they intend to read it, develop a
language independent specification for reiserfs and implement from that 
(reverse engineer)?

If they intend to do direct translation it's a derivative work.
Copyright belongs to them, right to distribute is limited to the terms
of the license they were under when they did the translation.

If they intend to reverse engineer, you could go after them through
the aid of DMCAish legislation on the books, but you'll also do a good
job of convincing everyone that you're an asshole.

You might also want to consider taking this offline and
1) Determine the specific license these fools wish to use for their work.
2) Determine their jurisdiction.
3) Determine the compensation they can provide namesys.

-- 
Chris Dukes
Been there, done that, got the slightly-charred t-shirt. -- Crowder

Re: implementing reiserfs in C++ for a new OS

[Nikita Danilov]

Chris Dukes writes:
 > On Wed, Aug 11, 2004 at 09:43:13AM -0700, Hans Reiser wrote:
 > > Nikita Danilov wrote:
 > > 
 > > >
 > > >Copyright to result of translation belongs to the translator. At
 > > >least, according to Russian law.
 > 
 > Unless it was translated as a work for hire, then the contract is
 > generally that the copyright belongs to whoever hired the work.

Not according to the Russian law, which I referred to. In this country
translation is copyrighted by translator no matter who pays
him. Contract specifying otherwise will be voided in a court.

Nikita.

Re: implementing reiserfs in C++ for a new OS

[mjt]

On Wed, Aug 11, 2004 at 06:16:19PM +0100, Chris Dukes wrote:

>> Otherwise harry potter can get translated and no need to pay rowling.
>
>Not entirely.  Such a translation would be deemed a derivative work.
>The copyright on the translated version belongs to the translator
>(Unless the translator did it for pay).  However, distribution of that

Unless the translator did it for pay AND signed a waiver that he loses
the copyright for the translation.

>derivative work may be controlled by copyright holder of the work
>the translation was derived from (IE, rowling wouldn't be able to distribute
>the translation, but may be able to prevent the distribution of the
>translation).

Yes.

>If they intend to do direct translation it's a derivative work.
>Copyright belongs to them, right to distribute is limited to the terms
>of the license they were under when they did the translation.

Are computer languages really seen as translations?
I'd like to hear a lawyer's opinion on this matter.

I see those "translations" as reimplementations.

>If they intend to reverse engineer, you could go after them through
>the aid of DMCAish legislation on the books, but you'll also do a good
>job of convincing everyone that you're an asshole.

Reverse-engineering open source is not that difficult, hardly should
be called reverse-engineering.

Finnish law grants you the right to reverse-engineer in order to see
how the computer program works and "compile/change/translate the
shape/form/style of the code" (damn, I'd hate to be a legal translator)

You can do this as long as it's for finding out how a program works.
There are some restraints on what you can do with the reverse-engineered
code, though, but they all revolve around using the code for compatibility
reasons.

So if the implementation of ReiserFS is compatible with Namesys' ReiserFS,
it'd be good in Finland.

Not that this has to be true everywhere else, but I think most of this
stuff is international.

-- 
mjt

Re: implementing reiserfs in C++ for a new OS

[Hans Reiser]

Chris Dukes wrote:

>On Wed, Aug 11, 2004 at 09:43:13AM -0700, Hans Reiser wrote:
>  
>
>>Nikita Danilov wrote:
>>
>>    
>>
>>>Copyright to result of translation belongs to the translator. At
>>>least, according to Russian law.
>>>      
>>>
>
>Unless it was translated as a work for hire, then the contract is
>generally that the copyright belongs to whoever hired the work.
>  
>
>>>
>>>      
>>>
>>I may be completely wrong legally, esp. in Russia.  I personally think 
>>it should belong to the original author if the original copyright is 
>>still valid.
>>    
>>
>
>Hans,
>Since you obviously have a stake in this intellectual property,
>have you considered talking with your lawyer instead of giving us
>your fantasies?
>  
>
I have no stake in harry potter.  The other guys are conforming to the 
GPL by not distributing and don't realize it, from what I understand.

Hans

Re: implementing reiserfs in C++ for a new OS

[Alex Zarochentsev]

On Wed, Aug 11, 2004 at 09:43:13AM -0700, Hans Reiser wrote:
> Nikita Danilov wrote:
> 
> >
> >Copyright to result of translation belongs to the translator. At
> >least, according to Russian law.
> >
> > 
> >
> I may be completely wrong legally, esp. in Russia.  I personally think 
> it should belong to the original author if the original copyright is 
> still valid.
> 
> Otherwise harry potter can get translated and no need to pay rowling.

one needs to pay rowling and pay translator.

> Hans

-- 
Alex.

Re: implementing reiserfs in C++ for a new OS

[Hans Reiser]

Alex Zarochentsev wrote:

>On Wed, Aug 11, 2004 at 09:43:13AM -0700, Hans Reiser wrote:
>  
>
>>Nikita Danilov wrote:
>>
>>    
>>
>>>Copyright to result of translation belongs to the translator. At
>>>least, according to Russian law.
>>>
>>>
>>>
>>>      
>>>
>>I may be completely wrong legally, esp. in Russia.  I personally think 
>>it should belong to the original author if the original copyright is 
>>still valid.
>>
>>Otherwise harry potter can get translated and no need to pay rowling.
>>    
>>
>
>one needs to pay rowling and pay translator.
>
>  
>
>>Hans
>>    
>>
>
>  
>
Yes, I think zam is right.

Re: implementing reiserfs in C++ for a new OS

[Spam]

<snip>

> If  you  translate  a  poem  into  french,  it  is  still rightfully
> copyright protected by the original author.

  Indeed.  But  I  could write another poem which is based on the same
  ideas.  Another  example  would  be  an  idea about a cool 3D GUI it
  doesn't  mean  no  one else can implement that idea. I would have to
  patent  it (in those countries that allow software patents) in order
  to protect the idea.

  Perhaps the line where copyright is applicable or not is very faint.

> But  hey,  these guys are just doing this work for academic purposes
> they  say and not releasing it to the world, so I think they are not
> going  to be violating the GPL, so it is all nothing to worry about.
> Maybe  this  sort  of  thing  makes for a good student assignment in
> their view, I don't know....

  It  would probably be a good student assignment. Reiser4 is a modern
  and very advanced file system with many interesting aspects. I think
  this  discussion  for most of us is on the theoretical level on what
  is acceptable or not?

> Hans

Re: implementing reiserfs in C++ for a new OS

[mjt]

On Wed, Aug 11, 2004 at 10:21:38PM +0200, Spam wrote:
>  Indeed.  But  I  could write another poem which is based on the same
>  ideas.  Another  example  would  be  an  idea about a cool 3D GUI it
>  doesn't  mean  no  one else can implement that idea. I would have to
>  patent  it (in those countries that allow software patents) in order
>  to protect the idea.

That's why patents are a bad thing.

Originally they're of course about giving the inventor some leeway to
do something with his invention, but they can be abused, and are abused,
so badly that they do more harm than good.

http://www.base.com/software-patents/disputes.html

That for example, Google got me many more hits, but everyone cand find
the data themselves :)

-- 
mjt

Re: implementing reiserfs in C++ for a new OS

[mjt]

On Wed, Aug 11, 2004 at 06:54:16AM +0200, Spam wrote:
>
>  Take  as  an  example a inventor. If he has an idea that he wants to
>  protect he would need to patent it to protect it. Otherwise only his
>  work  can be protected with copyright. I am not sure that the actual
>  invention can be copyrighted either.

Who was the guy who invented the telephone separately of Bell
at the same time?
The reason Bell got his product pushed to the market was that he
could patent it first.

That's about it, the idea is not protected and the only way to protect
it is by patents, but hey, where have we seen a patent that did more
good than keep it at a neutral level?

The way patents are dealt with is that everyone goes out and patents
everything they can and then get money from the court because they
"own the idea of a seesaw swing"...

>  In  this  case  with  Hans,  Namesys and reiserfs/reiser4 the actual
>  written  code  and  documentation  etc  is covered by copyright. The
>  ideas  behind  the filesystems are not. They would only be protected
>  by patents.

Exactly.

The only patent that makes any sort of sense to me is, even theoretically,
is the medical patent. Medicine can take decades to implement so those
guys will want sovereignity over their work.

But there is the hidden aspect here as well: They would benefit from other
people having access to the sama data, the same research, the same idea,
and get it done faster and cheaper. Cut down the need for bootleg medicine.

This is not so far fetched from this case either imo.

Hans mentioned the cute movie script. Studios stealing scripts is not
really the same thing. The movie does not exist yet, only the script,
and what the studio is really doing is depriving the writer of his
deserved compensation for his work. That's not reimplementing an
existing idea.

As for the poem translated into French, of course the original author
has the copyright to his work, but the translator has the copyright
to his translation. The idea is untouched. What if the Greeks of old
had patented, nay, copyrighted as some people think ideas should be
copyrighted, the common structures of storytelling?

Or maybe if we couldn't use the material from old mythology, because
they are the intellectual property of someone else?
What about those guys who lived 50 years after the Greeks and took
ideas from them into their own mythos, before the copyrights would
have expired?

-- 
mjt

Re: implementing reiserfs in C++ for a new OS

[Hans Reiser]

Markus Törnqvist wrote:

>
>
>As for the poem translated into French, of course the original author
>has the copyright to his work, but the translator has the copyright
>to his translation.
>
The translator is infringing, if he does it without permission on a 
copyrighted work (IANAL).

> The idea is untouched. What if the Greeks of old
>had patented, nay, copyrighted as some people think ideas should be
>copyrighted, the common structures of storytelling?
>
>Or maybe if we couldn't use the material from old mythology, because
>they are the intellectual property of someone else?
>What about those guys who lived 50 years after the Greeks and took
>ideas from them into their own mythos, before the copyrights would
>have expired?
>
>  
>

Re: implementing reiserfs in C++ for a new OS

[mjt]

On Wed, Aug 11, 2004 at 12:51:06AM -0700, Hans Reiser wrote:

>The translator is infringing, if he does it without permission on a 
>copyrighted work (IANAL).

Of course only if he publishes his translation without the consent
of the original author.

Finnish law has a threshold for what is considered a work, in the
sense of a work of art, in copyright law. It's also a bit diffuse
about what it defines as a work, but if the original being translated
falls under this threshold, no copyrights are dealt.

Naturally anything discussed in this thread go over that threshold :)

-- 
mjt

Re: implementing reiserfs in C++ for a new OS

[Dr. Giovanni A. Orlando]

Markus Törnqvist wrote:

>On Wed, Aug 11, 2004 at 06:54:16AM +0200, Spam wrote:
>  
>
>> Take  as  an  example a inventor. If he has an idea that he wants to
>> protect he would need to patent it to protect it. Otherwise only his
>> work  can be protected with copyright. I am not sure that the actual
>> invention can be copyrighted either.
>>    
>>
>
>Who was the guy who invented the telephone separately of Bell
>at the same time?
>  
>
Was an italian, and he did before Bell. His name is Antonio Meucci.

Meucci open the copyright file 5 years before Bell, but don't have the money
to renew the copyright.

This is the story in Italian language:

http://www.stpauls.it/gio97/1797gi/1797gi5c.htm

>The reason Bell got his product pushed to the market was that he
>could patent it first.
>
>  
>
False.

>That's about it, the idea is not protected and the only way to protect
>it is by patents, but hey, where have we seen a patent that did more
>good than keep it at a neutral level?
>
>The way patents are dealt with is that everyone goes out and patents
>everything they can and then get money from the court because they
>"own the idea of a seesaw swing"...
>
>  
>
>> In  this  case  with  Hans,  Namesys and reiserfs/reiser4 the actual
>> written  code  and  documentation  etc  is covered by copyright. The
>> ideas  behind  the filesystems are not.
>>
Also this is false. Also the idea behind is copyright. This is the 
invention and the copyright not the stupid code, in C or C++ ;-)

I want to tell you more. In the year 2000, at Comdex. I introduce the 
FTKernelConfigurator idea, my idea.

I introduce to granroth and other people from KDE. Well, after some time 
appears magically the kernel configurator
inside the Kcontrol.

I suppose they stole my idea, and for me is logical. I don't yet find 
the time, to write the code for FTKernelConfiguator, but I will do.

I know in the past one of the high copyright attorney in America, was a 
causal meeting. I contact him to protect my FTLinuxCourse.

I meet him before Comdex 2000. He comment to me a great true: "Don't 
comment, or they will take".

Easy.

I will finish here with some comments:

    * At first, the truth is ethrnal. It is not important how bastard 
you are and how tricks you apply. The truth, is ethernal, is one, and never
        can be overwrite. You can only hide the truth, temporarely.

    * At second, because here all is clear, public. I expect the minimal 
respect that goes to the owner and inventor. It is a question
       of class and necessary morality. The Stallman GNU Public License 
warrants this morality to the owner.

    * The third and last point is that you can run in false mode, also 
for years, and stole the software, the onwer will re-create
       and will re-invent almost in ethernal mode. The movie 'Tron' 
offers a nice underline mode to these facts.

I expect NEVER to see here again words like 'asshole' or other offensive 
words.

Thanks,
Giovanni.

PS. Poor Capital people :-)

>> They would only be protected
>> by patents.
>>    
>>
>
>Exactly.
>
>The only patent that makes any sort of sense to me is, even theoretically,
>is the medical patent. Medicine can take decades to implement so those
>guys will want sovereignity over their work.
>
>But there is the hidden aspect here as well: They would benefit from other
>people having access to the sama data, the same research, the same idea,
>and get it done faster and cheaper. Cut down the need for bootleg medicine.
>
>This is not so far fetched from this case either imo.
>
>Hans mentioned the cute movie script. Studios stealing scripts is not
>really the same thing. The movie does not exist yet, only the script,
>and what the studio is really doing is depriving the writer of his
>deserved compensation for his work. That's not reimplementing an
>existing idea.
>
>As for the poem translated into French, of course the original author
>has the copyright to his work, but the translator has the copyright
>to his translation. The idea is untouched. What if the Greeks of old
>had patented, nay, copyrighted as some people think ideas should be
>copyrighted, the common structures of storytelling?
>
>Or maybe if we couldn't use the material from old mythology, because
>they are the intellectual property of someone else?
>What about those guys who lived 50 years after the Greeks and took
>ideas from them into their own mythos, before the copyrights would
>have expired?
>
>  
>


-- 

-- 

--
Check FT Websites ... 
http://www.futuretg.com  - ftp://ftp.futuretg.com
http://www.FTLinuxCourse.com
	http://www.FTLinuxCourse.com/Certification
http://www.rpmparadaise.org
http://GNULinuxUtilities.com
http://www.YourPersonalOperatingSystem.com

--

Re: implementing reiserfs in C++ for a new OS

[Hans Reiser]

Dr. Giovanni A. Orlando wrote:

>
>
>    * At second, because here all is clear, public. I expect the 
> minimal respect that goes to the owner and inventor. It is a question
>       of class and necessary morality. The Stallman GNU Public License 
> warrants this morality to the owner.
>
>
Ah, you missed the Debian-legal vs. me flame fest....  probably for the 
best.... ;-)

The GPL needs better language protecting credits.

Re: implementing reiserfs in C++ for a new OS

[mjt]

On Wed, Aug 11, 2004 at 10:10:16AM +0200, Dr. Giovanni A. Orlando wrote:
>
>>The reason Bell got his product pushed to the market was that he
>>could patent it first.
>False.

Why then?

Meucci did not have enough money to push it?
So Meucci had bad luck, that's life, unfortunately.

>Also this is false. Also the idea behind is copyright. This is the 
>invention and the copyright not the stupid code, in C or C++ ;-)

No it is not, at least not in Finland, and I'm pretty sure it's the
same in all Europe and/or other countries.

http://www.cs.tut.fi/~jkorpela/tekoik/tekoik.html#mita

Unfortunately that's in Finnish...

"Tekijänoikeus suojaa teoksen kirjallista tai muuta asua, ei asiasisältöä."

Copyright protects the literary or other presentation of a work, not the
factual content.

"Teoksessa esitetyn tiedon tai mielipiteen voi tekijänoikeuden estämättä
esittää muualla omin sanoin."

The data or opinion in a work can be represented elsewhere in own words
without copyright preventing it.

"Tekijänoikeuslaki ei tällöin aseta edes vaatimusta lähteen ilmoittamisesta."

In this case copyright law does not even force you to give out the source.

Also http://www.cs.tut.fi/~jkorpela/tekoik/tekoik.html#moroik

"Tekijänoikeuteen sisältyy myös ns. moraalisia oikeuksia."

Copyright includes also so-called moral rights.

"Niihin sisältyy, että teoskappaleissa tai teosta esitettäessä on tekijä
ilmoitettava (isyysoikeus) ja että teosta ei saa muuttaa "tekijän kirjallista
tai taiteellista arvoa tahi omalaatuisuutta loukkaavalla tavalla"
(respektioikeus) (3 §)."

They include that when presenting items of the work (I wonder if that means
more like copies of the work..) or the work the original maker must be
noted (paternityright) and the work may not be changed 'in a way that insults
the makers literary or artistic value or uniqueness' (respect right)

>I want to tell you more. In the year 2000, at Comdex. I introduce the 
>FTKernelConfigurator idea, my idea.
>I introduce to granroth and other people from KDE. Well, after some time 
>appears magically the kernel configurator
>inside the Kcontrol.
>I suppose they stole my idea, and for me is logical. I don't yet find 
>the time, to write the code for FTKernelConfiguator, but I will do.

Well, there is a certain orifice in the human body that you did not
want to hear on this list that may describe the guys in this case.

But they did not really do anything wrong. If you had really forbidden
them from doing this, they might not have done it, but it's always
reasonable doubt that they invented the idea themselves.

I have no idea what FTKernelConfigurator is supposed to do, but I gather
it's an interface for configuring the kernel?
Surely this is not an idea to be held sovereignly for oneself.

>I meet him before Comdex 2000. He comment to me a great true: "Don't 
>comment, or they will take".

Yeah, if you wan't to keep a secret, keep it secret.

>Easy.

Indeed.

>   * At first, the truth is ethrnal. It is not important how bastard 
>you are and how tricks you apply. The truth, is ethernal, is one, and never
>       can be overwrite. You can only hide the truth, temporarely.

Yes, but I fail to see how this applies exactly.
No one is trying to lie here.

If someone stole your idea, or implemented it before you, it's not
a lie.

It's also hard to steal an idea that you gave them first.

>   * At second, because here all is clear, public. I expect the minimal 
>respect that goes to the owner and inventor. It is a question
>      of class and necessary morality. The Stallman GNU Public License 
>warrants this morality to the owner.

I'm, of course, not against respect here.
But the respect should not be ordered by the law, it could easily
lead into the patent mess that reigns in the USA.

Respect should come from respecting someone else.

>   * The third and last point is that you can run in false mode, also 
>for years, and stole the software, the onwer will re-create
>      and will re-invent almost in ethernal mode. The movie 'Tron' 
>offers a nice underline mode to these facts.

I must confess I haven't seen Tron...

Sure, it's called competition.

Also most of the time if your secret leaks out, and it's an important
secret, it leaks out when you have developed it far enough that it's
stupid to compete with you.

It is also true that often secrets come in the way of advancement,
co-operation and collaboration are the way to go.

-- 
mjt

Re: implementing reiserfs in C++ for a new OS

[Dr. Giovanni A. Orlando]

Markus Törnqvist wrote:

>On Wed, Aug 11, 2004 at 10:10:16AM +0200, Dr. Giovanni A. Orlando wrote:
>  
>
>>>The reason Bell got his product pushed to the market was that he
>>>could patent it first.
>>>      
>>>
>>False.
>>    
>>
>
>Why then?
>
>Meucci did not have enough money to push it?
>So Meucci had bad luck, that's life, unfortunately.
>  
>
Meucci open the file for copyright before Bell, but seems it is 
necessary to renew each year.

Therefore, he don't renew.

The word "inventor" in the US, means generally "rich". This have another 
sense outside America, or
for immigrant people.

>>Also this is false. Also the idea behind is copyright. This is the 
>>invention and the copyright not the stupid code, in C or C++ ;-)
>>    
>>
>
>No it is not, at least not in Finland, and I'm pretty sure it's the
>same in all Europe and/or other countries.
>
>http://www.cs.tut.fi/~jkorpela/tekoik/tekoik.html#mita
>
>Unfortunately that's in Finnish...
>
>"Tekijänoikeus suojaa teoksen kirjallista tai muuta asua, ei asiasisältöä."
>
>Copyright protects the literary or other presentation of a work, not the
>factual content.
>
>  
>
This is bad. The code or documentation explains the ideas, and must be 
copyright in the same mode
are the ideas itself.

>"Teoksessa esitetyn tiedon tai mielipiteen voi tekijänoikeuden estämättä
>esittää muualla omin sanoin."
>
>The data or opinion in a work can be represented elsewhere in own words
>without copyright preventing it.
>
>"Tekijänoikeuslaki ei tällöin aseta edes vaatimusta lähteen ilmoittamisesta."
>
>In this case copyright law does not even force you to give out the source.
>
>Also http://www.cs.tut.fi/~jkorpela/tekoik/tekoik.html#moroik
>
>"Tekijänoikeuteen sisältyy myös ns. moraalisia oikeuksia."
>
>Copyright includes also so-called moral rights.
>
>"Niihin sisältyy, että teoskappaleissa tai teosta esitettäessä on tekijä
>ilmoitettava (isyysoikeus) ja että teosta ei saa muuttaa "tekijän kirjallista
>tai taiteellista arvoa tahi omalaatuisuutta loukkaavalla tavalla"
>(respektioikeus) (3 §)."
>
>They include that when presenting items of the work (I wonder if that means
>more like copies of the work..) or the work the original maker must be
>noted (paternityright) and the work may not be changed 'in a way that insults
>the makers literary or artistic value or uniqueness' (respect right)
>
>  
>
>>I want to tell you more. In the year 2000, at Comdex. I introduce the 
>>FTKernelConfigurator idea, my idea.
>>I introduce to granroth and other people from KDE. Well, after some time 
>>appears magically the kernel configurator
>>inside the Kcontrol.
>>I suppose they stole my idea, and for me is logical. I don't yet find 
>>the time, to write the code for FTKernelConfiguator, but I will do.
>>    
>>
>
>But they did not really do anything wrong. If you had really forbidden
>them from doing this, they might not have done it, but it's always
>reasonable doubt that they invented the idea themselves.
>
>I have no idea what FTKernelConfigurator is supposed to do, but I gather
>it's an interface for configuring the kernel?
>  
>
Yes. Actually today it is also possible to configure the kernel 
graphically, but my idea is better.

I need time that I have no actually.

>Surely this is not an idea to be held sovereignly for oneself.
>
>  
>
>>I meet him before Comdex 2000. He comment to me a great true: "Don't 
>>comment, or they will take".
>>    
>>
>
>Yeah, if you wan't to keep a secret, keep it secret.
>
>  
>
>>Easy.
>>    
>>
>
>Indeed.
>
>  
>
>>  * At first, the truth is ethrnal. It is not important how bastard 
>>you are and how tricks you apply. The truth, is ethernal, is one, and never
>>      can be overwrite. You can only hide the truth, temporarely.
>>    
>>
>
>Yes, but I fail to see how this applies exactly.
>No one is trying to lie here.
>
>If someone stole your idea, or implemented it before you, it's not
>a lie.
>
>It's also hard to steal an idea that you gave them first.
>
>  
>
If you comment to KDE people something that does not exist, and after 
your comment exist something similar
you need simply do a small sum: 2+2.

Be stoled by someone that re-offer it under GPL your idea, cause some 
bad humor.

>>  * At second, because here all is clear, public. I expect the minimal 
>>respect that goes to the owner and inventor. It is a question
>>     of class and necessary morality. The Stallman GNU Public License 
>>warrants this morality to the owner.
>>    
>>
>
>I'm, of course, not against respect here.
>But the respect should not be ordered by the law, it could easily
>lead into the patent mess that reigns in the USA.
>
>Respect should come from respecting someone else.
>  
>
In 1987 or so, Chinese and people for Taiwan and Japan go around the 
world and comment:
"We will copy your product and do it better". Today, Taiwan and Japan is 
probably the best
place for electronic components worldwide.

Different companies here in Italy, about manufacturing suffering some 
fraud from Chinese companies
that re-produce the hardware for car testing and stole the clients. Do 
will someone open a case on China
for copyright contacting chinese attorneys?. Very complex and expensive.

I see the entire software industry going to disaster because some young 
boy release a similar program
at no cost. No one may compete with free. In any this is a complex 
global problem.

>>  * The third and last point is that you can run in false mode, also 
>>for years, and stole the software, the onwer will re-create
>>     and will re-invent almost in ethernal mode. The movie 'Tron' 
>>offers a nice underline mode to these facts.
>>    
>>
>
>I must confess I haven't seen Tron...
>  
>
You don't know Tron, because probably you are too young.

Tron was made by Walt Disney (Buena ventura). It is dated 1982.

Check at Amazon.com:
http://www.amazon.com/exec/obidos/tg/detail/-/B00005OCMR/qid=1092217246/sr=8-1/ref=pd_ka_1/102-0873987-4980163?v=glance&s=dvd&n=507846

Jeff Bridges, a young programmer in a small company was stoled by its 
game program. A friend of him write a program
that inspect information, and also some time ... the owner of the 
company was fired because he claim to be the author
of the game and Jeff Bridge then get the 'expected' money and respect.

I have this movie and like it. Also it is copyright :-)

Thanks,
Giovanni

>Sure, it's called competition.
>
>Also most of the time if your secret leaks out, and it's an important
>secret, it leaks out when you have developed it far enough that it's
>stupid to compete with you.
>
>It is also true that often secrets come in the way of advancement,
>co-operation and collaboration are the way to go.
>
>  
>


-- 

-- 

--
Check FT Websites ... 
http://www.futuretg.com  - ftp://ftp.futuretg.com
http://www.FTLinuxCourse.com
	http://www.FTLinuxCourse.com/Certification
http://www.rpmparadaise.org
http://GNULinuxUtilities.com
http://www.YourPersonalOperatingSystem.com

--

Re: implementing reiserfs in C++ for a new OS

[Rudy L. Zijlstra]

Dr. Giovanni A. Orlando wrote:

> Markus Törnqvist wrote:
>
>> On Wed, Aug 11, 2004 at 10:10:16AM +0200, Dr. Giovanni A. Orlando wrote:
>>  
>>
>>>> The reason Bell got his product pushed to the market was that he
>>>> could patent it first.
>>>>     
>>>
>>> False.
>>>   
>>
>>
>> Why then?
>>
>> Meucci did not have enough money to push it?
>> So Meucci had bad luck, that's life, unfortunately.
>>  
>>
> Meucci open the file for copyright before Bell, but seems it is 
> necessary to renew each year.
>
> Therefore, he don't renew.
>
> The word "inventor" in the US, means generally "rich". This have 
> another sense outside America, or
> for immigrant people.
>
>
Aren't the two of you confusing copyright and patents? In my shortened 
laymen terms, a copyright is about the expression, a patent about the 
idea (and the expressions of it).

Cheers,

Rudy

Re: implementing reiserfs in C++ for a new OS

[Nikita Danilov]

Dr. Giovanni A. Orlando writes:
 > Markus Törnqvist wrote:
 > 
 > >On Wed, Aug 11, 2004 at 10:10:16AM +0200, Dr. Giovanni A. Orlando wrote:
 > >  
 > >
 > >>>The reason Bell got his product pushed to the market was that he
 > >>>could patent it first.
 > >>>      
 > >>>
 > >>False.
 > >>    
 > >>
 > >
 > >Why then?
 > >
 > >Meucci did not have enough money to push it?
 > >So Meucci had bad luck, that's life, unfortunately.
 > >  
 > >
 > Meucci open the file for copyright before Bell, but seems it is 
 > necessary to renew each year.
 > 
 > Therefore, he don't renew.
 > 
 > The word "inventor" in the US, means generally "rich". This have another 

Ha-ha. Whenever one applies for a job in the US, one signs a contract,
and contract usually specifies that everything (tangible and
intangible) produced by employee is owned by employer. As vast
majority of `inventions' in our times are done within corporation
research laboratories, `inventors' themselves get no special rights
for their inventions.

Nikita.

Re: implementing reiserfs in C++ for a new OS

[Hans Reiser]

Nikita Danilov wrote:

>Dr. Giovanni A. Orlando writes:
> > Markus TЖrnqvist wrote:
> > 
> > >On Wed, Aug 11, 2004 at 10:10:16AM +0200, Dr. Giovanni A. Orlando wrote:
> > >  
> > >
> > >>>The reason Bell got his product pushed to the market was that he
> > >>>could patent it first.
> > >>>      
> > >>>
> > >>False.
> > >>    
> > >>
> > >
> > >Why then?
> > >
> > >Meucci did not have enough money to push it?
> > >So Meucci had bad luck, that's life, unfortunately.
> > >  
> > >
> > Meucci open the file for copyright before Bell, but seems it is 
> > necessary to renew each year.
> > 
> > Therefore, he don't renew.
> > 
> > The word "inventor" in the US, means generally "rich". This have another 
>
>Ha-ha. Whenever one applies for a job in the US, one signs a contract,
>and contract usually specifies that everything (tangible and
>intangible) produced by employee is owned by employer. As vast
>majority of `inventions' in our times are done within corporation
>research laboratories, `inventors' themselves get no special rights
>for their inventions.
>
>Nikita.
>
>
>
>
>  
>
If it was not intended sarcastically, then I should say that in the US 
inventors are usually thought of as poor geniuses working in basements, 
and in reality most tend to work for companies.  Some however really do 
work in basements.  I did.;-)

One of the strengths of US culture is that it is generally accepted that 
you can work in a basement and invent something and become rich.  If you 
attempt it, you are considered more credible in the US, though still not 
very credible.... it is considered a respectable madness.

I need to conform on the rich description.....

Hans

Re: implementing reiserfs in C++ for a new OS

[mjt]

On Wed, Aug 11, 2004 at 11:44:59AM +0200, Dr. Giovanni A. Orlando wrote:

>Meucci open the file for copyright before Bell, but seems it is 
>necessary to renew each year.

Copyrights are not filed, patents are.

Are you sure you're not mistaken in words here?

>Therefore, he don't renew.

How stupid of him.

>The word "inventor" in the US, means generally "rich". This have another 
>sense outside America, or
>for immigrant people.

I have never heard of a rich inventor, except the chance to get money
by producing the invention.

>>"Tekijänoikeus suojaa teoksen kirjallista tai muuta asua, ei asiasisältöä."
>>Copyright protects the literary or other presentation of a work, not the
>>factual content.
>This is bad. The code or documentation explains the ideas, and must be 
>copyright in the same mode are the ideas itself.

The idea is not copyrighted or patented or protected in any way, but the
code is. There is a difference there.
Of course the idea can be patented in some degenerate country like the US
and it seems the EU is making the same mistake, but technically the idea
can not be protected. And should not be.

>Yes. Actually today it is also possible to configure the kernel 
>graphically, but my idea is better.
>I need time that I have no actually.

I know how you're feeling.

I have also many projects for which I haven't had time, so I've given
them away to see if anyone else would do it and I would help and give
ideas. All in freedom and free software.

Maybe you should talk to the KDE guys about your idea that they would
implement it more like you would, both of you would win.


>If you comment to KDE people something that does not exist, and after 
>your comment exist something similar
>you need simply do a small sum: 2+2.
>Be stoled by someone that re-offer it under GPL your idea, cause some 
>bad humor.

You gave it to them, whose fault is that?

>>Respect should come from respecting someone else.
>In 1987 or so, Chinese and people for Taiwan and Japan go around the 
>world and comment:
>"We will copy your product and do it better". Today, Taiwan and Japan is 
>probably the best
>place for electronic components worldwide.

Sure. And we let them do it, whose fault is that?

>Different companies here in Italy, about manufacturing suffering some 
>fraud from Chinese companies
>that re-produce the hardware for car testing and stole the clients. Do 

Fraud?

If they do it better, where's the fraud?
I mean, what's stopping Italy from doing it better than China?
Berlusconi? C'mon, you could still compete with China, it's not like the
world's biggest army is holding you in siege.

>will someone open a case on China
>for copyright contacting chinese attorneys?. Very complex and expensive.

I don't know how the car is patented, I'm sure it is, but still everyone
has the right to set up a car shop and build the things.

This is one of the points of free trade. Everyone's equal and free,
so if someone pushes out better products, you respond by making an even
better product.

I fail to see the point in whining about something as glorious as this.

>I see the entire software industry going to disaster because some young 
>boy release a similar program
>at no cost. No one may compete with free. In any this is a complex 
>global problem.

Of course you can compete with free, fight free with free, make a better
free solution.

It's like a*x=a*y => x=y

>>I must confess I haven't seen Tron...
>You don't know Tron, because probably you are too young.
>Tron was made by Walt Disney (Buena ventura). It is dated 1982.

I know such a movie exists, but not much more. I'm also born in 1982 ;)

>Jeff Bridges, a young programmer in a small company was stoled by its 
>game program. A friend of him write a program
>that inspect information, and also some time ... the owner of the 
>company was fired because he claim to be the author
>of the game and Jeff Bridge then get the 'expected' money and respect.

Yeah, I should see it.

-- 
mjt

Re: implementing reiserfs in C++ for a new OS

[Christophe Saout]

[-- Attachment #1: Type: text/plain, Size: 115 bytes --]

Am Mittwoch, den 11.08.2004, 13:54 +0300 schrieb Markus Törnqvist:

> It's like a*x=a*y => x=y

Or a=0 :-)


[-- Attachment #2: Dies ist ein digital signierter Nachrichtenteil --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: implementing reiserfs in C++ for a new OS

[Dr. Giovanni A. Orlando]

Markus Törnqvist wrote:

>On Wed, Aug 11, 2004 at 11:44:59AM +0200, Dr. Giovanni A. Orlando wrote:
>
>  
>
>>Meucci open the file for copyright before Bell, but seems it is 
>>necessary to renew each year.
>>    
>>
>
>Copyrights are not filed, patents are.
>  
>
Yes. is the patent, not the copyright.

He don't have the money to pay the renew the patent.

>Are you sure you're not mistaken in words here?
>
>  
>
>>Therefore, he don't renew.
>>    
>>
>
>How stupid of him.
>
>  
>
>>The word "inventor" in the US, means generally "rich". This have another 
>>sense outside America, or
>>for immigrant people.
>>    
>>
>
>I have never heard of a rich inventor, except the chance to get money
>by producing the invention.
>  
>
Well. If an inventor have no money is a BIG problem, however actually
in the GPL times, it is very easy to release a software also with no money.

>>>"Tekijänoikeus suojaa teoksen kirjallista tai muuta asua, ei asiasisältöä."
>>>Copyright protects the literary or other presentation of a work, not the
>>>factual content.
>>>      
>>>
>>This is bad. The code or documentation explains the ideas, and must be 
>>copyright in the same mode are the ideas itself.
>>    
>>
>
>The idea is not copyrighted or patented or protected in any way, but the
>code is.
>
This is bad and incorrect for me.

> There is a difference there.
>Of course the idea can be patented in some degenerate country like the US
>and it seems the EU is making the same mistake, but technically the idea
>can not be protected. And should not be.
>  
>
The US is not a degeneated country. Do you stay in the US sometime?

>  
>
>>Yes. Actually today it is also possible to configure the kernel 
>>graphically, but my idea is better.
>>I need time that I have no actually.
>>    
>>
>
>I know how you're feeling.
>
>I have also many projects for which I haven't had time, so I've given
>them away to see if anyone else would do it and I would help and give
>ideas. All in freedom and free software.
>
>Maybe you should talk to the KDE guys about your idea that they would
>implement it more like you would, both of you would win.
>
>
>  
>
>>If you comment to KDE people something that does not exist, and after 
>>your comment exist something similar
>>you need simply do a small sum: 2+2.
>>Be stoled by someone that re-offer it under GPL your idea, cause some 
>>bad humor.
>>    
>>
>
>You gave it to them, whose fault is that?
>
>  
>
I comment what we will do. They release a version. However, actually is 
not important.

>>>Respect should come from respecting someone else.
>>>      
>>>
>>In 1987 or so, Chinese and people for Taiwan and Japan go around the 
>>world and comment:
>>"We will copy your product and do it better". Today, Taiwan and Japan is 
>>probably the best
>>place for electronic components worldwide.
>>    
>>
>
>Sure. And we let them do it, whose fault is that?
>
>  
>
>>Different companies here in Italy, about manufacturing suffering some 
>>fraud from Chinese companies
>>that re-produce the hardware for car testing and stole the clients. Do 
>>    
>>
>
>Fraud?
>
>If they do it better, where's the fraud?
>  
>
No they copy 100% the same machine for car test.

>I mean, what's stopping Italy from doing it better than China?
>Berlusconi? C'mon, you could still compete with China, it's not like the
>world's biggest army is holding you in siege.
>
>  
>
>>will someone open a case on China
>>for copyright contacting chinese attorneys?. Very complex and expensive.
>>    
>>
>
>I don't know how the car is patented, I'm sure it is, but still everyone
>has the right to set up a car shop and build the things.
>  
>

>This is one of the points of free trade. Everyone's equal and free,
>so if someone pushes out better products, you respond by making an even
>better product.
>
>I fail to see the point in whining about something as glorious as this.
>
>  
>
>>I see the entire software industry going to disaster because some young 
>>boy release a similar program
>>at no cost. No one may compete with free. In any this is a complex 
>>global problem.
>>    
>>
>
>Of course you can compete with free, fight free with free, make a better
>free solution.
>
>It's like a*x=a*y => x=y
>
>  
>
>>>I must confess I haven't seen Tron...
>>>      
>>>
>>You don't know Tron, because probably you are too young.
>>Tron was made by Walt Disney (Buena ventura). It is dated 1982.
>>    
>>
>
>I know such a movie exists, but not much more. I'm also born in 1982 ;)
>
>  
>
>>Jeff Bridges, a young programmer in a small company was stoled by its 
>>game program. A friend of him write a program
>>that inspect information, and also some time ... the owner of the 
>>company was fired because he claim to be the author
>>of the game and Jeff Bridge then get the 'expected' money and respect.
>>    
>>
>
>Yeah, I should see it.
>
>  
>
OK.

Thanks,
Giovanni


-- 

-- 

--
Check FT Websites ... 
http://www.futuretg.com  - ftp://ftp.futuretg.com
http://www.FTLinuxCourse.com
	http://www.FTLinuxCourse.com/Certification
http://www.rpmparadaise.org
http://GNULinuxUtilities.com
http://www.YourPersonalOperatingSystem.com

--

Re: implementing reiserfs in C++ for a new OS

[mjt]

On Wed, Aug 11, 2004 at 02:12:50PM +0200, Dr. Giovanni A. Orlando wrote:

>Yes. is the patent, not the copyright.
>He don't have the money to pay the renew the patent.

That sucks then.

Maybe he should have had someone market and promote it better, maybe
he would not have needed a patent at all, then.

Or he would have gotten money for a patent through marketing.

Remember, Philips pushed the CD out with cheap prices, everyone could
use CDs cheaply, so there was no real competition. It was the superior
solution for cheap.

It did not become common because of forcing it on customers, nor did
it become common like the VHS tape because JVC bought a big market
presence.

>Well. If an inventor have no money is a BIG problem, however actually
>in the GPL times, it is very easy to release a software also with no money.

Or release the specs for some other invention so cheap that you make
much money through huge sales.

>>The idea is not copyrighted or patented or protected in any way, but the
>>code is.
>This is bad and incorrect for me.

I have heard much better arguments for why it's good that ideas can NOT
be protected, not one good argument for why they should be.

I'd like to hear that one good argument.

People have the same ideas independently, person A and person B.
Person A patents it and person B can't compete with person A.
Person A makes a bad implementation and no one can do anything about it,
no one can do it better.

Your idea about protecting ideas starts to sound like Microsoft, doesn't
it?

Only that Microsoft and maybe some pantent-holders just abused their big
money to suppress competition.

>>Of course the idea can be patented in some degenerate country like the US
>>and it seems the EU is making the same mistake, but technically the idea
>>can not be protected. And should not be.
>The US is not a degeneated country. Do you stay in the US sometime?

I've never been there.

But look at the Digital Millennium Copyright Act and others.
USA was supposed to be the land of the free, now it's the land of the
rich and powerful, where some people are more free than others and some
people are more equal than others.

Every honest individual who makes an invention may get sued over patents
because someone else can claim the most stupid patent ever.

There's a term Götterdämmerung Capitalism which is such abuse of a fine
thing, and that's what they're into across the Atlantic.

The point is not in culminating a lot of money and spending it stupidly
or not at all, it's about how money is moved smartly into good causes.

Maybe it's unfair to generalize the States, I know there's a lot of
good and smart people there, but the country is ran by a monkey, among
other flaws.

Maybe some libertarian-like state could be found, where it would be good
to live.

>>You gave it to them, whose fault is that?
>I comment what we will do. They release a version. However, actually is 
>not important.

But this is a chance for co-operation, you can ask the KDE guys to
do as you say if your system is better.

They will get a better system and you will see your brainchild become
a used application.

Everyone wins.

>>Fraud?
>>If they do it better, where's the fraud?
>No they copy 100% the same machine for car test.

So? Everyone else copies them too.

Copy something back from China.

It should not be seen as a bad thing but as a good thing.

-- 
mjt

Re: implementing reiserfs in C++ for a new OS

[Dr. Giovanni A. Orlando]

Markus Törnqvist wrote:

>On Wed, Aug 11, 2004 at 02:12:50PM +0200, Dr. Giovanni A. Orlando wrote:
>
>  
>
>>Yes. is the patent, not the copyright.
>>He don't have the money to pay the renew the patent.
>>    
>>
>
>That sucks then.
>
>Maybe he should have had someone market and promote it better, maybe
>he would not have needed a patent at all, then.
>  
>
He have the original patent and is recognized like the official inventor
of the phone by the US court

Read it in the file http://www.stpauls.it/gio97/1797gi/1797gi5c.htm

>Or he would have gotten money for a patent through marketing.
>
>Remember, Philips pushed the CD out with cheap prices, everyone could
>use CDs cheaply, so there was no real competition. It was the superior
>solution for cheap.
>
>It did not become common because of forcing it on customers, nor did
>it become common like the VHS tape because JVC bought a big market
>presence.
>
>  
>
>>Well. If an inventor have no money is a BIG problem, however actually
>>in the GPL times, it is very easy to release a software also with no money.
>>    
>>
>
>Or release the specs for some other invention so cheap that you make
>much money through huge sales.
>
>  
>
>>>The idea is not copyrighted or patented or protected in any way, but the
>>>code is.
>>>      
>>>
>>This is bad and incorrect for me.
>>    
>>
>
>I have heard much better arguments for why it's good that ideas can NOT
>be protected, not one good argument for why they should be.
>
>I'd like to hear that one good argument.
>
>People have the same ideas independently, person A and person B.
>Person A patents it and person B can't compete with person A.
>Person A makes a bad implementation and no one can do anything about it,
>no one can do it better.
>
>Your idea about protecting ideas starts to sound like Microsoft, doesn't
>it?
>
>  
>
No.

>Only that Microsoft and maybe some pantent-holders just abused their big
>money to suppress competition.
>
>  
>
>>>Of course the idea can be patented in some degenerate country like the US
>>>and it seems the EU is making the same mistake, but technically the idea
>>>can not be protected. And should not be.
>>>      
>>>
>>The US is not a degeneated country. Do you stay in the US sometime?
>>    
>>
>
>I've never been there.
>
>  
>
Therefore, you cannot comment about the US. Actually, the US is not the 
'normal' US. I hope things
back normal and 'for business' soon. I suppose with the new president.

>But look at the Digital Millennium Copyright Act and others.
>USA was supposed to be the land of the free, now it's the land of the
>rich and powerful, where some people are more free than others and some
>people are more equal than others.
>
>Every honest individual who makes an invention may get sued over patents
>because someone else can claim the most stupid patent ever.
>  
>
Yes. Unfortunately this is possible. Microsoft wins a cause a get 4 
million US$
for a small company owner.

In infringements patents, 1 million US$ is something like one dollar, 
the minimal quantity of money to speak about.

>There's a term Götterdämmerung Capitalism which is such abuse of a fine
>thing, and that's what they're into across the Atlantic.
>
>The point is not in culminating a lot of money and spending it stupidly
>or not at all, it's about how money is moved smartly into good causes.
>
>Maybe it's unfair to generalize the States, I know there's a lot of
>good and smart people there, but the country is ran by a monkey, among
>other flaws.
>
>Maybe some libertarian-like state could be found, where it would be good
>to live.
>
>  
>
>>>You gave it to them, whose fault is that?
>>>      
>>>
>>I comment what we will do. They release a version. However, actually is 
>>not important.
>>    
>>
>
>But this is a chance for co-operation, you can ask the KDE guys to
>do as you say if your system is better.
>
>They will get a better system and you will see your brainchild become
>a used application.
>
>Everyone wins.
>  
>
I prefer to release it like I comment and I design, when possible. We 
announce time ago
that almost all our problems are GPL and also commercial.

Basically, we charge for the manual and additional not free software, 
like Trolltech does.

Then, if someone like, and want to contrib. Great.

Also, have no sense to release a program to configure the kernel and 
offer only in commercial form. Must be
also available for free like the Kernel is.

>  
>
>>>Fraud?
>>>If they do it better, where's the fraud?
>>>      
>>>
>>No they copy 100% the same machine for car test.
>>    
>>
>
>So? Everyone else copies them too.
>
>Copy something back from China.
>
>It should not be seen as a bad thing but as a good thing.
>
>  
>
China and Hong Kong is the first place in the world where people 
re-produce 'everything' in almost illegal mode: shoes, pants, etc.

Italy have the fourth place. Also on the streets here, you can found 
Fendi, that is a bad reproduction.

Thanks for now.

Regards,
Giovanni


-- 

-- 

--
Check FT Websites ... 
http://www.futuretg.com  - ftp://ftp.futuretg.com
http://www.FTLinuxCourse.com
	http://www.FTLinuxCourse.com/Certification
http://www.rpmparadaise.org
http://GNULinuxUtilities.com
http://www.YourPersonalOperatingSystem.com

--

Re: implementing reiserfs in C++ for a new OS

[mjt]

On Wed, Aug 11, 2004 at 03:04:47PM +0200, Dr. Giovanni A. Orlando wrote:

>>People have the same ideas independently, person A and person B.
>>Person A patents it and person B can't compete with person A.
>>Person A makes a bad implementation and no one can do anything about it,
>>no one can do it better.
>>Your idea about protecting ideas starts to sound like Microsoft, doesn't
>>it?
>No.

I think it does.

It closes down on competition, and competition breeds evolution, and
evolution leads to prosperity.

>Therefore, you cannot comment about the US. Actually, the US is not the 
>'normal' US. I hope things
>back normal and 'for business' soon. I suppose with the new president.

The American Dream is not the America of today.

People moved there, built their towns, everyone was free, everyone
contributed to society, everything was ok.

Then things went wrong.

Sure the "Normal" US, the American Dream US, with freedom and strong
economics and equality would be just about the best thing ever.

That's not how things are now, ergo, they are degenerated.

>>Every honest individual who makes an invention may get sued over patents
>>because someone else can claim the most stupid patent ever.
>Yes. Unfortunately this is possible. Microsoft wins a cause a get 4 
>million US$ for a small company owner.
>In infringements patents, 1 million US$ is something like one dollar, 
>the minimal quantity of money to speak about.

Unfortunately a mille may be a lot of money for the small company.

And still people think that it's a good thing that someone patents
an idea even if it hurts small companies from advancing and competing.

Look at how well they split Microsoft into smaller companies. About as
effective as chopping a tree down with your finger. If the counter-plea
ever gets finished, I'm sure Microsoft will have won.

That's not the "normal" US, right?

>Basically, we charge for the manual and additional not free software, 
>like Trolltech does.
>Then, if someone like, and want to contrib. Great.

That is your right and I wish you the best of luck.

>Also, have no sense to release a program to configure the kernel and 
>offer only in commercial form. Must be
>also available for free like the Kernel is.

Sure, it can be free and a part of some package that contains the
manual and additional non-free software.

That's a good business plan, yes?
Circulate as much free software as possible and get paid only for what
really costs, ie. printing manuals. About keeping software closed forever
I will not comment any more, it depends on so many things. I would release
any closed software once I have made enough money off of it.

>China and Hong Kong is the first place in the world where people 
>re-produce 'everything' in almost illegal mode: shoes, pants, etc.

And Western companies set up factories there and pay the locals as much
as four times the salary they'd get in a local factory. Even if it's very
little by our standards.

Of course western companies should support also the local industries and
not just "abuse" them but they still pay relatively good money.

If they don't, they are criminal :)

What about bootled medicine in countries where people can't afford medicine
because of patents?

Almost the same thing.

-- 
mjt

Re: implementing reiserfs in C++ for a new OS

[Tom Vier]

On Wed, Aug 11, 2004 at 03:39:09PM +0300, Markus T?rnqvist wrote:
> But look at the Digital Millennium Copyright Act and others.
> USA was supposed to be the land of the free, now it's the land of the
> rich and powerful, where some people are more free than others and some
> people are more equal than others.

i hate the dmca as much as anyone, but wealth (money or resources) means
more "freedom" everywhere. it means the freedom to take more risks, make more
investments (thereby allowing you to pay people to make things to patent,
for example). but this is a little offtopic. 8)

on topic: hans, they have every right to reimpliment reiserfs3. but, it
MUST be done in a clean fashion, like ibm's bios was reimplimented, unless
they want get a license from you (either the gpl, or another).

-- 
Tom Vier <tmv@comcast.net>
DSA Key ID 0x15741ECE

Re: implementing reiserfs in C++ for a new OS

[mjt]

On Wed, Aug 11, 2004 at 07:22:07PM -0400, Tom Vier wrote:
>i hate the dmca as much as anyone, but wealth (money or resources) means
>more "freedom" everywhere. it means the freedom to take more risks, make more
>investments (thereby allowing you to pay people to make things to patent,
>for example). but this is a little offtopic. 8)

But without patents you wouldn't have to be rich to have something patented,
as the system would not exist. Nor would you gain anything by running to
the patent office, first in line, to have your everyday idea patented for
your own profit.

You are also right about the investments, but that's only one type of
freedom. I just wish the guys who made it rich in stocks or other types
of investment would use their money smartly and not just collect a heap
of it to sit on.

-- 
mjt

Re: implementing reiserfs in C++ for a new OS

[David Greaves]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ramachandra K wrote:

|Let me clarify that the nature of Capital project is academic research.
|
|rgds
|Ram

As an enquiry - why not use the GPL? Your messages imply that you are
concerned about using it and want to avoid that.

A lot of people are sensitive about the morals around GPL code.
Although I have no standing here I'd be interested to hear and it may
help calm those who worry.

Your messages to the list can be interpreted in various ways - and
some interpretations are not in keeping with the philosophical basis
of the GPL - maybe you could clarify :)

David
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFBGpf/8LvjTle4P1gRAoePAJ9h8eEvQbuTUASm67Qok/sf/u+PewCfd8lI
eyDtXXCd8faMORDXH1tmO1I=
=T1qi
-----END PGP SIGNATURE-----

Re: implementing reiserfs in C++ for a new OS

[Jonathan Briggs]

[-- Attachment #1: Type: text/plain, Size: 981 bytes --]

On Mon, 2004-08-09 at 08:52, Dr. Giovanni A. Orlando wrote:
> Ramachandra K wrote:
> 
> >Hans Reiser <reiser@namesys.com> wrote:
> >
> >  
> >
> >>Is it a free OS, or do you need to buy a license from us in
> >>addition to  the GPL?
> >>
> >>    
> >>
> >
> >ReiserFS would be implemented FROM SCRATCH for Capital
> >in C++ with the help of existing literature/documentation etc.
> >By this I mean that no existing code would be used, that is
> >ported or modified in any way. This should mean that GPL 
> >would not be applicable to the code written for my OS, Capital.
> >  
> >
> Opps. There are something wrong here.
> 
> Seems that Capital OS, will stole the ReiserFS idea and re-implement.
> 
> Very honest, very capital like.
> 
> Thanks,
> Giovanni

To be fair, it is the same as OSS has done to Microsoft with NTFS and
FAT.  I would not expect much help from Hans Reiser for such a project,
though.
-- 
Jonathan Briggs
jbriggs@esoft.com

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

RE: implementing reiserfs in C++ for a new OS

[David Dabbs]

> 
> I am planning to implement ReiserFS on my operating system named Capital
> (www.mitpune.com/research/capital1.html). To give a brief background -
> Capital is a 32-bit Object Oriented operating system for the Intel
> i386 range of microprocessors. Capital's features include
> multithreading,virtual memory and a VFS interface.
>
> 
> rgds
> Ram

Ram,

Based upon your project description, you might be better served by talking
to the folks in IBM's K42 OS project. K42's filesystem, kfs, seems similar
in philosophy to Reiser4, though to learn more you'll need to get on their
list and read the code. The short kfs overview isn't meaty enough. Their
team has tackled most of the work of providing a Linux API 'personality' for
the OS kernel.  The project is implemented in C++ and was built from the
ground up to run (and scale) on modern, cache-coherent multiprocessor
machines. See the K42 OS Home Page: http://www.research.ibm.com/K42/

Also of interest to file system designers will be (more at the home page)

KFS: Exploring Flexibility in File System Design 
http://www.research.ibm.com/K42/papers/wsoBrazil2004.pdf
Dilma Da Silva, Livio Soares, Orran Krieger, "KFS: Exploring Flexibility in
File System Design", to appear in the Brazilian Workshop on Operating
Systems (WSO'2004), Salvador, Brazil, August 2004.

Meta-data Snapshotting: A simple mechanism for File System Consistency 
http://www.research.ibm.com/K42/papers/snapi03.pdf
Livio Soares, Orran Krieger, Dilma Da Silva, "Meta-data Snapshotting: A
simple mechanism for File System Consistency", SNAPI'03 (International
Workshop on Storage Network Architecture and Parallel I/O).


Good luck,

David