first consult the file ../README for a vague idea on the meaning of subdirectories.

Compatibility:

Should work with Ruby 1.6.6 and PureData 0.36, 0.37, 0.38

Release steps (OPERATE WITH CAUTION):

make vvtest # leak detection

ruby base/test.rb formats

cvs tag gridflow_0_7_7 .

mkdir gridflow

cvs -d :pserver:anonymous@cvs.gridflow.ca:/home/cvs/gridflow export -r gridflow_0_7_7 -d gridflow .

cd gridflow

rm -rf CVS CVSROOT

chmod -R go=u-w .

cd ..

tar cfzvv gridflow-0.7.7.tar.gz gridflow

rm -rf gridflow # caution!!!

scp gridflow*gz gridflow@artengine.ca:download

mv gridflow*gz /home/matju/net/GridFlow

ssh gridflow@artengine.ca : expand tarball into public_html/latest

download somewhere else and try to compile

update main page

jmax/pd/ruby-talk mlists : post release (note: prefix subject with [ANN] !!!!)

freshmeat.net : post release

RAA : post release

Pictures identified as "<help>" should be screenshots of the help files. They shouldn't be screenshots of the template.

A Help file is a visual summary of what an object does. It should be interactive and allow the user to understand what the object does. Sample inputs are provided and sample results can be seen.

All modifications to the documentation must be done in the XML. The HTML must be kept up-to-date by rerunning "make" in the "doc/" directory.

Help-wise, Formats are considered objects. Operators are to be documented as part of [@] and [@!].

Images grabbed directly from a jMax window should be encoded in PNG. This includes "icons" (object boxes, message boxes, etc). If they are in JPEG, only change to PNG when the image has to change anyway. Distinguish "icons" from "thumbnails", the latter of which are scaled-down grabs, which should be encoded in JPEG.

There are not help-thumbnails anymore in the reference manual because they were not helpful.

Large images should be kept out of the release. The release itself must be small enough not to discourage us from doing releases and keeping them all. It also should be downloadable on a bad modem. =) Let's keep it under 500k please.

in doc do: for z in images/*.* images/*/*.*; do fgrep -q $z *.html || echo "$z"; done

environment variables CVSEDITOR=pico CVS_RSH=ssh2