Almost all the files mentioned in this FAQ are available by standard
Internet FTP.  If you don't know how to use FTP, please read the article
"Anonymous FTP FAQ List", which you can get by sending e-mail to
mail-server@rtfm.mit.edu with the single line "send faqs/ftp-list/faq"
in the body.  (See also "[17] Where are FAQ lists archived?")  This section
gives some quick reminders which are not meant as a substitute for reading
the FTP FAQ.

If you use a WWW browser such as Netscape or Lynx, it will do FTP for you.
To retrieve a file described here as "site.name:/path/to/file", tell the
browser to open the URL "ftp://site.name/path/to/file".  (If you are reading
this FAQ in the WWW FAQ archive, the file names should appear as links that
you can just click on.)  Depending on your browser, you may have to shift-
click or take some other special action to instruct the browser to save the
file to disk, rather than trying to display the file to you.

If you do not have direct access to FTP, you can use an "ftpmail" server to
obtain files by e-mail.  See the FTP FAQ for details.

Many of the pointers given here refer to popular central archive sites,
such as ftp.simtel.net for DOS software or sumex-aim.stanford.edu for Mac.
These sites are often overloaded, and are likely to refuse your connection
request when they are busy.  You can try again at a less popular time of
day, or you can look for a "mirror site".  Most central archive sites have
groups of mirror sites that keep copies of their files.  Find out the name
of the mirror site closest to you, and visit that site instead; it's good
net citizenship and you'll get faster response.  Check the FAQs for the
newsgroups specific to your system type to find lists of mirror sites.
(The archive site may list some mirror sites in its connection-refused error
message.  Unfortunately, some FTP programs won't show you the whole message.
WWW browsers are often bad about this.)

If you are able to reach the archive site, but the file you want doesn't
exist, most likely it's been replaced by a newer version.  Get a directory
listing of the directory that's supposed to contain the file, and look for
a file with a similar name but a higher version number.  In a WWW browser,
you can get a directory listing by removing the file name, that is opening
the URL consisting of everything up to and including the last slash.  (If
you find an out-of-date reference in a *current* version of the JPEG FAQ,
I'd appreciate hearing about it by e-mail.)

Practically all of the files listed here are compressed archive files.
This means you need to retrieve them in binary mode.  (WWW browsers do this
automatically, but many older FTP programs must be told to use binary mode.)
Once you've got the archive file, you'll need a decompressor/dearchiver
to extract the program and documentation files inside it.  Check the FAQs
for your system type to find out where to get dearchiver programs.