From mboxrd@z Thu Jan 1 00:00:00 1970 From: Hans Reiser Subject: Re: file as a directory Date: Mon, 20 Dec 2004 09:21:35 -0800 Message-ID: <41C70A1F.8000609@namesys.com> References: <200412180152.iBI1q98X007507@laptop11.inf.utfsm.cl> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: list-help: list-unsubscribe: list-post: Errors-To: flx@namesys.com In-Reply-To: <200412180152.iBI1q98X007507@laptop11.inf.utfsm.cl> List-Id: Content-Type: text/plain; charset="us-ascii"; format="flowed" To: Horst von Brand Cc: David Masover , Peter Foldiak , reiserfs-list@namesys.com, linux-kernel@vger.kernel.org, Nate Diller Horst von Brand wrote: > >>>Can you point me to any such literature? I'm just curious. >>> >>> > > > >>Look in every field of cs except filesystems and kernels.;-) >> >> > >Right. Smart people are found elsewhere only. > > There is a huge pile of innovations by the database field that filesystems people do not yet make use of. The merits of interdisciplinary study are understood by many. > > >> Databases, >>garbage collectors, etc. >> >> > >Everthing stuff that works on the assumption that what they are working on >fits in RAM (or can overflow into swap space in a pinch), and that RAM is >fast (and even so they are infuriatingly slow). And disks are usually a few >thousand times larger than RAM (more stuff to shuffle around) and a million >times slower... > > > >> Specific reference, no, I didn't collect them, >>sorry, but alexander the befs driver guy knows more than I about this. >> >> > >Furious handwaving doesn't make it true. > > Ok, go talk to the befs driver guy, and you'll find out he has already done work on it. If his work does not satisfy you, sorry, I do detailed research just before coding, not to satisfy some guy on lkml.