Differences between revisions 1 and 4 (spanning 3 versions)
Revision 1 as of 2005-03-08 02:28:52
Size: 339
Editor: anonymous
Comment:
Revision 4 as of 2005-03-08 02:59:28
Size: 802
Editor: anonymous
Comment:
Deletions are marked like this. Additions are marked like this.
Line 6: Line 6:
 * ftp.it.debian.org: http://blog.bofh.it/id_82
 * ftp.se.debian.org: http://err.no/personal/blog/tech/Debian/2005-03-04-18-13_ftp.se_download_numbers
 * ftp.uk.debian.org: http://blog.einval.com/2005/03/07#uk-mirror-200503
 * [http://blog.bofh.it/id_82 ftp.it.debian.org]
 * [http://err.no/personal/blog/tech/Debian/2005-03-04-18-13_ftp.se_download_numbers ftp.se.debian.org]
 * [http://blog.einval.com/2005/03/07#uk-mirror-200503 ftp.uk.debian.org]

These numbers are unrepresentative in several ways. Use with caution. For example:

 * Lots of people have private mirrors
 * Web caches distort download statistics effectively
 * Architecture popularity varies between countries
 * There is no statistics for CD distribution
 * It's unlikely that people do network installs of s390
 * It's possible that one downloads once, then install multiple times
 * Most embedded people are likely to have local mirrors

Debian architecture usage

FTP download numbers from mirror logs:

These numbers are unrepresentative in several ways. Use with caution. For example:

  • Lots of people have private mirrors
  • Web caches distort download statistics effectively
  • Architecture popularity varies between countries
  • There is no statistics for CD distribution
  • It's unlikely that people do network installs of s390
  • It's possible that one downloads once, then install multiple times
  • Most embedded people are likely to have local mirrors