Dirsum is relatively self explanatory. It is currently just a command line tool. I intend to do more with it in the future like making it a gui program and having an option to look at blocks instead of bytes. The project dirsum is a command line tool to assist a user in sorting out which directories in a filesystem contain the most bytes. It will sort all of the subdirectories under a common directory according to the total bytes in all files recusively through its subdirectories. It gives a more complete and desctiptive information than using existing command line tools like du. The code is c++ so hopefully it will operate quickly.. Unfortunately to get the information on each line I wanted I needed more than 80 characters wide. So use of an xterm is quite preferable. By default dirsum will perform it's function on the present working directory. If one argument is put on the command line it will try to open the text as a directory and if it is, perform it's function on that directory. By default dirsum will not recurse into new partitions. To recurse into new partitions use the -x option. First dirsum will sum the bytes of the files in the directory in question. Next dirsum will create a list of directories under the directory performed on. The list will include the following header (with example first line). /root has 7712584 Bytes in 44 Files, with 27 Subdirectories which is 74.28 percent of total Subdirs This Directory 1 Deeper 2 - 45 Deeper Totals Name Bytes Files Dirs Bytes Files Dirs Bytes Files Files Bytes % fontwork 1.5G 27 1 106.7M 26 53 1.6G 15.81 For this example there are a 44 files in root which posses 74% of the bytes in files under root. The rest are in the list of subdirectories. Each of the first level subdirectories will be listed with the appropriate count of directories under it and files and bytes in those files 1 level and a sum of the files and direcories deeper than that. That's all there is to it. Dirk