Subject: Re: Thank you Tony (was: XSL-list doomed) From: Mike Brown <mike@xxxxxxxx> Date: Mon, 30 Oct 2000 15:26:35 -0400 (EST) |
Paul Tchistopolskii wrote: > How can I be 'fair' or 'unfair' with technical issues ? > > The archive of XSL-list is broken since summer > and nobody bothers to fix it. Is it 'fair' or 'not fair' - > I don't understand. It is the *fact*. There are 2 popular pieces of software that people use to provide mailing list interfaces on web sites: Hypermail and MHonArc. Both work the same way -- for each message a separate HTML file is created. I have used both of these systems and they both reach a point of critical mass on a large archive where you cannot run them at all, unless you have an infinite amount of memory. They are also extremely wasteful, tripling the size of an archive in terms of disk space, and using up thousands and thousands of inodes which makes doing even a directory listing a major chore. Add on top of this the nearly impossible task of adding a search engine, which cannot afford to simply grep through all those files every time someone wants to search but instead needs to build and use an optimized index... it becomes a huge pain to set up and maintain, and a monumental resource hog. You have to ask, "is it really worth it, just so people can see pretty HTML in their web browsers?" I have had to discontinue web-based archiving for a dozen lists because they are too big to be worth the effort, using the software that is out there. The only solution for large lists right now is to break up the archive into completely separate sections by month. It's still wasteful of resources, and then you also have to write a custom set of scripts to handle the automatic archiving under such a system. Rebuilding a large archive can also take *days* of one's time. It is not so simple as you would like to believe. It was this archive nightmare experience shared by me and a friend that led to the formation of the EyeBrowse project at http://eyebrowse.tigris.org/ ... it still has yet to reach a point of stability but it is getting there. In the mean time, I suggest that if nothing else, providing access to the raw mbox files is better than nothing.. we could set up our own archives or just download the files and search through them with our own tools. - Mike ____________________________________________________________________ Mike J. Brown, software engineer at My XML/XSL resources: webb.net in Denver, Colorado, USA http://www.skew.org/xml/ XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
Re: Thank you Tony (was: XSL-list d, Alex Albu | Thread | Re: Thank you Tony (was: XSL-list d, Warren Hedley |
Which version of IE, David Schach | Date | Re: Detail tables handled in XSLT, , Steve Muench |
Month |