From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754836Ab0IBNPo (ORCPT ); Thu, 2 Sep 2010 09:15:44 -0400 Received: from borg.medozas.de ([188.40.89.202]:51923 "EHLO borg.medozas.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753478Ab0IBNPn (ORCPT ); Thu, 2 Sep 2010 09:15:43 -0400 Date: Thu, 2 Sep 2010 15:15:37 +0200 (CEST) From: Jan Engelhardt To: Miklos Szeredi cc: Neil Brown , linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, vaurora@redhat.com, viro@zeniv.linux.org.uk, jblunck@suse.de, hch@infradead.org Subject: Re: [PATCH 0/5] hybrid union filesystem prototype In-Reply-To: Message-ID: References: <20100826183340.027591901@szeredi.hu> <20100827170551.19616048@notabene> <20100827213502.31af4a4c@notabene> <20100829144207.4fbf2713@notabene> User-Agent: Alpine 2.01 (LNX 1266 2009-07-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=UTF-8 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Monday 2010-08-30 12:18, Miklos Szeredi wrote: > >> My comment about set-theory unions being commutative set me thinking. I >> really don't think "union" is the right name for this thing. There is >> nothing about it which really fits that proper definition of a union. > >We could call it overlayfs. People learn new names quickly :) There is a much larger issue that you should be very well aware about — "The name wanted to be a clever acronym for "Filesystem in USErspace", but it turned out to be an unfortunate choice. The author has since vowed never to name a project after a common term, not even anything found more than a handful of times on Google." overlayfs already exists. Right next to fuse on sourceforge... Oh and I what I like to see is support for multiple readonly branches :)