Hi!

On Fri, Jun 21, 2002 at 05:26:12PM +0800, Stas Bekman wrote:
> We have discussed this issue before. My opinion is that we don't want 
> any official mirrors. The reason: Search the web for the mod_perl guide. 
> You will find dozens of "mirrors" which are very very outdated and 
> include erroneous information. Making official mirrors, meaning that you 
> have to monitor those mirrors and make sure that they get updated.
> 
> There are also technical problems with mirrors, e.g. the search won't 
> work, certain hidden objects aren't linked (e.g. 404.html), etc.

see below.

> And I'm not aware of a bandwidth problem at apache.org. The only 
> drawback of not having the mirrors is a slower usage experience for 
> those behind slow links.

..and causing traffic over half of the world. But maybe that's a good
thing, or else even more backbone operators will go bankrupt
(KPNQuest...)  :-)


> If we do maintain a list of mirrors and add a warning so users verify 
> that they read the latest copy (how?) they will still go back to the 
> master site. The guide used to have a release every so often, so it was 
> easy to update, the new model supports release often approach, so it 
> really has to be a mirror that monitors changes very often.

I thought that the current site is very well suited for mirroring, as
the building can be totally automated (as it is on perl.apache.org):

% cvs update
% bin/build

If SWISH is installed on the mirror, searching should be possible to...

I wasn't thinking of a simple mirror of the HTML tree (dst_html) ...

> So I say that for the moment if you want to have a mirror, go for it, 
> but for now there is only one source of docs people should know about 
> and that's perl.apache.org. After we release the site, if there will be 
> people whom we can trust that they will keep their mirrors in sync we 
> may do something about it.

OK. My main concern was that I use the docs locally (faster), but the
search won't work. So I thought I just get the search running on my
box, and than, that maybe other people would like to use my local
copy, and so on ...

OK, I could use grep for finding stuff, but now that we have such a
wonderfull search engine (Thanks, Bill!), it just seems awkward ..


-- 
#!/usr/bin/perl                                http://domm.zsi.at
for(ref(bless[],just'another'perl'hacker)){s-:+-$"-g&&print$_.$/}

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to