[Oberon] filesystem with directories ?

p.provoost at rn.rabobank.nl p.provoost at rn.rabobank.nl
Mon May 1 08:36:53 CEST 2006


Maybe I should elaborate on the reason I asked about Bluebottles
filesystem ...

The main reason for a journaled filesystem with directories is the added
reliability with respect to FAT and EXT2. Especially FAT is notorious
for its vulnerability to crashes. FAT32 is just as bad IMHO.

Reiserfs is more stable, as is EXT3, but EXT3 is quite slow.

And of course the directories are a must when the number of files
increases. In the days that Oberon fitted on a floppy disk, one
root-direcotry was fine. But now that it is about 100 MB with hundreds
of files, I think the time has come to use a more sophisticated file
system.




p.provoost at rn.rabobank.nl wrote:
> > > ..snip...
> > >
> > > 4. Is there going to be a filesystem with directories (possibly a
> > > journaled file system) for Bluebottle in the forseeable future ?
> > >
Ulrike Glavitsch wrote:
> > ...snip...
> > Unfortunately, there will not be a file system with directories (and

> > other features) in the near future.
> > The attribute file system handed in by a master student last fall
will 
> > have to be re-engineered in order to be used by the BB community.

Chris Glur wrote
> Is it UNFORTUNATE that there will not be a file system with
directories ?
> Is it perhaps appropriate to give the theoretical background
[philosophy]
> why the existing file system was chosen ?


Cheers,

Paul Provoost
The Netherlands


------------------------------

Message: 4
Date: Sat, 29 Apr 2006 05:33:43 +0200 (SAST)
From: easlab at absamail.co.za
Subject: [Oberon] Re: Anybody want to contact Negroponte and suggest
	he look	at Oberon?
To: oberon at lists.inf.ethz.ch
Message-ID: <20060429033343.C60BB36834E at mmi01.absamail.co.za>

Douglas G. Danforth wrote:
> Cadef writes "According to a story on CNet News.com, Nicholas
Negroponte 
> says that Linux has gotten too fat, and will have to be slimmed down 
> before it will be practical for the $100 laptop project. From the 
> article: 'Suddenly it's like a very fat person [who] uses most of the 
> energy to move the fat. And Linux is no exception. Linux has gotten
fat, 
> too.'"

Yes each newer version of Linux is noticebly worse !
One needs to have the policy against bloat-kreep written 
up-front - unless you're selling hardware.
Perhaps it needs legislation and a special police force.

OTOH one needs  to explain the theoretical basis of IF/WHY oberon 
claims to be more compact.
Is it from the terrible, near-asm, syntax/style eg.
		Texts.OpenScanner(S, Oberon.Par.text, Oberon.Par.pos);
		Texts.Scan(S);
		IF S.class IN {Texts.Name, Texts.String} THEN
			Texts.WriteString(W, S.s);
			F := Files.New(S.s);
			Texts.Scan(S);
			IF (S.class = Texts.Char) & ((S.c = "@") OR (S.c
= "^")) THEN...
instead of the more maintainable:
   	IF FirstArg = StartHere OR SelectedOne THEN...;
even though [as I recently raised on c.l.o.] there's massive repeating
of code between modules, instead of using 'libraries' ?
This apparently because each student must write his OWN Ascii2Hex
procedure.
 
Jan Verhoeven wrote:
> Right you are. Linux is too fat for the OLPC computer like Windows.
> 
> There is an alternative, although it is not 'ready' yet: Oberon
> 
How can you write Oberon is not ready YET ?
How many decades old is it ?
N-O is over the hill, and going down fast.

> This lean operating system still fits on a single floppy disk. We have
> it on sourceforge at the moment.
=============
Bernhard Treutwein wrote:
> finally I found a little bit time to look again into
> the archives of NO. Everything what is necessary to
> build the Oberon0 diskette appears to be in subdirectory
> build, i.e.:
> ftp://ftp.inf.ethz.ch/pub/ETHOberon/Native/Update/Alpha/Build/
> 
> Especially there is also (what I had not found before)
> the CvsClient.Mod ...
> 
Good, thanks.
Is there any 'documentation' for CvsClient.Mod, or is
it another of many potential jewels except for the 'manual' ?

> I'm not shure, if the hen & egg problem (isn't that a nice
> problem for easter) can be resolved with the native linux
> oberon ... but in any case that does not resolve the problem
> of getting used to the NO user interface, which really needs
> some time to adapt to.

This quote applies:
"Most Web 2.0 start-ups fail when users don't find
enough utility to deal with the early bugs."
My circumstances when I first tried NO, were that I really needed
to succeed.  Later it would have been usefull, but not essential
to mouse-by-left-hand. There I didn't succeed.
==============
I've been following the MIT/OLPC computer; also re. the simputer,
which fail[ed]/ing for non technical reasons - a long psyco/social
story.
That's why I've repeatedly raised the N-O for ARM question.

Whereas the simputer has an excellent 'design philosophy',
OLPC computer seems to make no sense:
* the ability to make a useable/reliable windup power generator
  to drive the note-book at less cost than the rest, would be more
  of an acheivment that making the rest/computer.
  Yes, the power supply is the starting point for 3rd world rural use,
  but the technology is unrelated to 'computing'. 
* There's no need to follow the latest bloated linux.
  I often use mulinux: a single diskette linux [with extra addons for 
  mc, ..etc].  With his transparent, educational layout, I never fail to

 be fascinated how the Italian designer simplified this package.
 It's like reading a Wirth text: beyond technology into 'art'.
    
It seems that Negroponte is trying to talk to the kewl-gee-wizz-yuppies
and the 3rd world NGOs at the same time.   You can't do it.
The cultural divide is too much.  Reading his texts is like seeing the 
Hollywood 'reality jungle adventures' where they emerge with perfect
makeup and clean white shirts after living in the jungle.
I'd love to be proven wrong, since hi-tech can be very cost effective
for the 3rd world as the cell-phone has proven.

BTW a recent tour of Zimbabwe explained to me how N.Korea works.

== Chris Glur.



------------------------------

--
Oberon at lists.inf.ethz.ch mailing list for ETH Oberon and related systems
https://lists.inf.ethz.ch/mailman/listinfo/oberon


End of Oberon Digest, Vol 31, Issue 18
**************************************





================================================
De informatie opgenomen in dit bericht kan vertrouwelijk zijn en 
is uitsluitend bestemd voor de geadresseerde. Indien u dit bericht 
onterecht ontvangt, wordt u verzocht de inhoud niet te gebruiken en 
de afzender direct te informeren door het bericht te retourneren. 
================================================
The information contained in this message may be confidential 
and is intended to be exclusively for the addressee. Should you 
receive this message unintentionally, please do not use the contents 
herein and notify the sender immediately by return e-mail.



More information about the Oberon mailing list