There is a file index.html placed in each directory.
This file gets served to the browser if a default request to that
directory using http protocol is made. For example, http://ncep.emc.ncep.noaa.gov/sst_anlysis
will display index.html. If this file is not there then a
directory listing is displayed. To avoid that keep a file index.html
and make it your top level document.
One purpose of this "inhouse" page is to be able to mirror it on the outside
server with least amount of work. To do that .html documents
should be portable and therefore should be independent of directory structure.
Use only relative URLs and not absolute ones. If needed, you can
retrace back your path to the top level directory (i.e. CMB home page)
using ../.. (like you would do in unix).
Do not give write permission to anyone except yourself
Try not to use symbolic links.
You must use a Jump Page to let the user know you are leaving the NOAA site under the following conditions:
-- For organizational links (see definition below) to governmental
agencies outside NOAA
-- To go to private orgs/businesses
You do not need a jump page:
-- To webpages within NOAA
-- For data links to governmental agencies outside NOAA (data links
are links to graphics or ftp or ascii data; see below)
Data link: hyperlinks to other US federal government agencies for the
purpose of displaying data stored in the other agency's web site which is
directly relevant to the NWS web site display.
Organizational link: links to organizations with related missions (e.g.
FAA, EPA, FEMA, USGS, US Army Corps of Engineers) provided so our
customers can contact these organizations for information or services they
No need to learn "html" to create pages any more. One can use "Composer"
if you are using "netscape communicator" or you can use sgi utility called
'webmagic" (most likely in the directory /usr/sbin/webmagic).
In communicator you might have to check the composer preference setting
and make sure no extra paths to links created by you get added.