> If the original intention of the uppcasing is to draw attention to for > example reading important things such as the README file, having 8 such > files (all "polluting" the root directory) is unlikely to help.
DEVELOPMENT was split off from README so that the Installation instructions on the website would make sense (being generated from this file) and TROUBLESHOOTING was added so that it could also be included in the website. Following your comments I have: * Merged DEVELOPMENT back into README * Moved TROUBLESHOOTING to http://wiki.apache.org/couchdb/Troubleshooting * Updated the Wiki to point to the new page plus README/NEWS in Subversion As for the reasoning behind these files, I concede that the last two were a little much, but as for the rest, they have been part of UNIX lore for since I was knee-high to a grasshopper. ;) http://en.wikipedia.org/wiki/Readme >> /www/incubator.apache.org/couchdb >> >> Are you suggesting that this should be a Subversion checkout? > > Yes. Okay, I have changed the following directory to be a Subversion check out: /www/incubator.apache.org/couchdb It concerns me that this exposes the .svn directories to the web so I did a quick check to see what the other directories were doing. Of 75 directories, 48 of them were also using direct check outs, so I figure it's standard: abdera, activemq, agila, audit, cayenne, depot, doap_converter, ftpserver, graffito, guides, harmony, images, imperius, incubation, ip-clearance, ivy, jdo, jspwiki, learn, log4php, lokahi, lucene.net, lucene4c, nmaven, ofbiz, official, old-activemq, old-servicemix, openejb, openjpa, pig, projects, qpid, river, roller, servicemix, shindig, stdcxx, style, tika, triplesoup, tsik, tuscany, uima, wicket, xap xmlbeanscxx Of course, the .svn directories are available via HTTP: http://incubator.apache.org/depot/.svn/ Perhaps it might be an idea to block this in Apache? > And build.sh now relies on sitting next to a trunk checkout to pull in > those other files. I'd prefer it remained self-contained, at least to > the extent that you don't need to checkout both trunk and site to > regenerate the site. I have reverted my changes to the site build system. Of course, we have no way of including the changelog or installation instructions on the website, which is a great shame. Can you think of any other way round this? > And I would've really preferred having this discussion before you checked in > the changes. I'm sorry that you feel this way. One of the first things I read when joining CouchDB was: It's OK to be aggressive with the edits! All edits are versioned and monitored and it's easier to ask forgiveness than permission, so have at it. http://www.couchdbwiki.com/index.php?title=Main_Page&oldid=3270 This certainly has set the tone for the last 6 months and it surprises me that you feel such a trivial change to the website should have been discussed first. Thanks, -- Noah Slater <http://bytesexual.org/>
