Mailing-List: contact cygwin-apps-help AT sourceware DOT cygnus DOT com; run by ezmlm Sender: cygwin-apps-owner AT sourceware DOT cygnus DOT com List-Subscribe: List-Archive: List-Post: List-Help: , Delivered-To: mailing list cygwin-apps AT sources DOT redhat DOT com Date: Sat, 31 Mar 2001 12:28:27 -0500 From: Christopher Faylor To: cygwin-apps AT cygwin DOT com Cc: dankamin AT cisco DOT com Subject: Re: those nonintuitive cygdrives... (Was: Making RXVT The Standard Terminal) Message-ID: <20010331122827.A2545@redhat.com> Reply-To: cygwin-apps AT cygwin DOT com Mail-Followup-To: cygwin-apps AT cygwin DOT com, dankamin AT cisco DOT com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.11i In-Reply-To: <011b01c0b95a$9ad63a60$1500040a@na.cisco.com>; from dankamin@cisco.com on Fri, Mar 30, 2001 at 12:47:11PM -0800 Since we still seem to be in "you should all do this because I know what I'm talking about" mode, it appears that Dan Kaminsky is not reading this mailing list. I have forwarded my last response, which didn't Cc him, back to cygwin-apps and Cc'ed him this time. Hopefully this will move this discussion into a more productive direction. Btw, Dan, I think you may have been missing the point that Egor was responding tongue-in-cheek. Or, maybe I was. cgf ----- Forwarded message from Christopher Faylor ----- From: Christopher Faylor To: cygwin-apps AT cygwin DOT com Subject: Re: Making RXVT The Standard Terminal Date: Fri, 30 Mar 2001 12:05:20 -0500 Reply-To: cygwin-apps AT cygwin DOT com Mail-Followup-To: cygwin-apps AT cygwin DOT com In-Reply-To: <00ca01c0b923$e2b3ec10$1500040a AT na DOT cisco DOT com>; from dankamin AT cisco DOT com on Fri, Mar 30, 2001 at 06:15:31AM -0800 On Fri, Mar 30, 2001 at 06:15:31AM -0800, Dan Kaminsky wrote: >> > /cygdrive is not "effectively undocumented"... > >In terms of: > >1) Crypto/Security >2) Legality >3) UI Design >4) Actual user (lack of) awareness > >It's not a listed resource when you ask for all available resources to be >shown, therefore it's effectively undocumented. > >> > Although, I guess it could be undocumented if you never actually read >> > the documentation. > >ls / > >doesn't show /cygdrive. Neither does mount. Therefore, it's >undocumented, by virtue of expectation. The semantic purpose of listing >a directory and a mount table is to request from the operating system a >documentation of what resources are available. The term "undocumented" has a very precise meaning. It doesn't mean "I looked at my screen and couldn't figure out what to do". If you want to make a point that it is not immediately obvious, I won't argue. However, much of cygwin's operation is not obvious from the start. What is a mount table anyway? Why can't I use backslashes in bash? My file has CRLF line endings. Why? FWIW, I'd be happy to make the cygdrive directory show up when you do a 'ls /'. Someday. Cygwin is a work in progress and this is very low on my personal list of things to do. >Clearly *something* needs to be done to make it more obvious how to access >system drives. I personally prefer /c, /d, etc. simply because it's a quick >mental translation from c:\ to /c. At minimum, a /cygdrive folder should >exist by default, perhaps just renamed to /drives. > >Ideally, the setup program itself would give the user the choice. This is a >*core need* for Cygwin to address; it's explicitly non-obvious how to access >resources outside the Cygwin psuedosandbox. I'll be interested to see your changes to setup.exe. Please send a patch and a ChangeLog entry here. Check out the "Contributing" link at the Cygwin web site for details. We don't need an assignment form for setup sources, so there is no need to fill one out unless you plan on contributing a patch to implement your idea about having /cygdrive show up when you type 'ls /', too. cgf ----- End forwarded message -----