Date: Fri, 11 Feb 2000 06:41:55 -0600 From: Richard Wackerbarth <rkw@dataplex.net> To: "Matthew D. Fuller" <fullermd@futuresouth.com> Cc: freebsd-ports@FreeBSD.ORG Subject: Re: /usr/ports/ too big? Message-ID: <00021020293801.00825@localhost.localdomain> In-Reply-To: <20000210200836.C13279@futuresouth.com> References: <200002102358.PAA03763@mina.sr.hp.com> <00021020011700.00825@localhost.localdomain> <20000210200836.C13279@futuresouth.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 10 Feb 2000, Matthew D. Fuller wrote: > I don't get it. > Really, cvsweb at the least gives you MORE 'recent' history than you > could get locally, since you always have up to an hour (or more) lag > between a change and your access via CVSup. What detail does a local CVS > repo supply that the above don't? 1) My local repository matches the code that I am compiling. 2) I don't have to be connected to the net to view it. > > > > I have one copy, you have one copy, he has one copy ... Replicated around > > the world. > By the same reasoning, we shouldn't have separate copies of binaries on > everyone's system, replicated around the world. I venture that most > people don't ever use a lot of the binaries on their systems. When was > the last time you used ctm? Or dtmfdecode? enigma? etc. I don't know about you, but I do use ctm (I was the ctm-meister for a year or so until it got too big to handle on my limited resources) As for the others, I routinely "strip" things from my installations. I'm certainly less concerned about having the current version of some component. The real waste is that I could care less about its history. If my needs change, I'll copy from a convenient repo. > > The problem is "packaging". You have the choice "all or nothing". I would > > prefer to be able to keep (and eventually use transparently) a combination of > > up-to-date history on HD and ancient history on archive CD's or the net. > > Oh, so you want a version of CVS that only keeps 'recent' revisions > around; sorta an auto-expiration of revisions once they reach a certain > age. Good luck writing it, or finding one; at the least, it goes against > the whole purpose of a revision control system. I disagree. There is nothing that dictates that the entire history of a file has to be stored in a single file. A cvs file consists of two parts. The first is a directory that allows conversion of a tag into the rcs revision target. The second is an rcs archive that allows the generation of the desired contents. Just as it is possible to store a subtree on a different physical device, there is no technical reason which would prevent the "directory" part of the cvs file to refer the request to an alternate source. -- Richard Wackerbarth rkw@Dataplex.NET To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-ports" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?00021020293801.00825>