Add the following lines to your .emacs file:

(setq 
      backup-by-copying nil 
      backup-by-copying-when-linked t
      backup-by-copying-when-mismatch t)

The group CVS repository is in /u/ml/groupcode/cvsroot. If you start projects there, please be sure to change the default group to lunar by executing 'chgrp -R lunar dir' on every dir you create. All CVS documentation is located here.

For more information, visit our Mini CVS tutorial page.

There are two things to keep in your mind when you put something in the /u/ml directory.

• Have all the directories in /u/ml/software be u+rwX and g+rwX, but o-rwX. This would allow group members to make modifications to the installations e.g. making minor (documented) changes to Makefiles, gunzipping files unused for a long time, etc. At the same time, it would not provide such access to the general public.
  It can be achieved for directory "foo" and all its subdirectories by using the command:
  
  chmod -R ug+rwX,o-rwX foo

• Have all the directories be owned by "lunar".
  It can be achieved for directory "foo" and all its subdirectories by using the command:
  
  chgrp -R lunar foo

• To fix all files in a directory to a group we can use 'chgrp -R lunar *'.

  We can also set the setgid bit for /u/ml and its subdirectories. Then every new file and subdirectories created inside will belong to lunar by default. To set the setgid bit, use 'find -type d | xargs chmod g+s'.

There is a way to check CS emails from web using UT's webmail facility. This is probably useful from Internet cafes which don't let you telnet etc..

1. Go to https://webmail.utexas.edu/
2. Give your UT EID and password
3. In the new page, under "Custom Login" give your CS login and password, write "imap.cs.utexas.edu" for server and "Inbox" for Folder.

I get the following URL exception error while running weka.gui.explorer.Explorer in VNC (or from an X-shell running on home computers). How do I fix it?

goya.cs.utexas.edu 101> java weka.gui.explorer.Explorer

Warning: No configuration property found in
weka/gui/GenericObjectEditor.props
for weka.experiment.InstanceQuery
Warning: Can't find public property editor for property "options" 
  (class "[Ljava.lang.String;"). 
Skipping.
Setting numeric precision to: 10
Setting axis colour to: green
Setting background colour to: black
Setting attribute bar colour to: black
Displaying attribute bars
Uncaught error fetching image:
java.lang.NullPointerException
      at sun.awt.image.URLImageSource.getConnection(URLImageSource.java:99)
      at sun.awt.image.URLImageSource.getDecoder(URLImageSource.java:108)
      at sun.awt.image.InputStreamImageSource.doFetch(InputStreamImageSource.java:251)
      at sun.awt.image.ImageFetcher.fetchloop(ImageFetcher.java:168)
      at sun.awt.image.ImageFetcher.run(ImageFetcher.java:136)
goya.cs.utexas.edu 102> exit

• You might want to change the CLASSPATH. Be sure to incude the current directory, denoted by a period (.) in your classpath, as in: -classpath .:classes. Click here for more information on this problem.

For those of you who are tired of seeing Java flush the annoying "Couldn't get file lock" exception when running Weka, the workaround is to run Java with this command-line option:

-Djava.util.prefs.syncInterval=2000000

(e.g. java -Xmx500m -Djava.util.prefs.syncInterval=2000000 
weka.gui.experiment.Experimenter)

Please use "dvips -Ppdf" when you convert your .dvi file to .ps file.

Instructions from AAAI

dvips -Ppdf -G0 -tletter paper
ps2pdf -dPDFSETTINGS=/printer -dCompatibilityLevel=1.4 -dMaxSubsetPct=0 \
  -dSubsetFonts=false -dEmbedAllFonts=true -sPAPERSIZE=letter paper.ps

Old Instructions

 
dvips -f -Ppdf -G0 < file.dvi > file.ps
ps2pdf13 -dPDFSETTINGS=/printer file.ps

To check the fonts of your pdf file, you can do one of the following:

• Open the pdf file with acroread and choose File, Document Properties, Fonts tab. Embedded fonts will be listed as "Embedded." Fonts that are not embedded will not be marked in any way.
• Use the command

  pdffonts file.pdf

gs -dNOPAUSE -sDEVICE=pdfwrite -sOUTPUTFILE=combined.pdf -dEmbedAllFonts=true \
  -dSubsetFonts=false -dBATCH a.pdf b.pdf

Since you may find creating figures easier using powerpoint or you want to use existing slides in your document, follow these steps.

• Save the powerpoint slides into a pdf (print -> save as PDF)
• Open the pdf using gimp
• Select to import the page you want
• Use Tools->Selection Tools to perform any cropping (NOTE: Use Image->Crop to Selection instead of Layer->Crop to Selection)
• Save the cropped figure as an eps file