You may want to first take a quick look at the colophon for a bit of background on how this set of web pages came to be, and my methodology for "content generation."
You shouldn't need to change URIs just because you changed which web application software you are using.
Things often included that should be left out include:
File name suffixes
Script names and other software mechanisms
One of the few things worth keeping is the creation date, as that shouldn't ever change.
I typically use Lynx , which can run happily on almost any kind of computer hardware known to mankind. Some Web pages are claimed to be Lynx Enhanced. The "Lynx Enhanced" quality has the unusual advantage (over "Netscape Enhanced" and "IE Enhanced," for instance) that these enhancements are non-exclusionary. Specifically, "Lynx Enhanced" indicates that:
The web page will be presented in a reasonably attractive fashion by all web browsers. (Unless you've specifically configured your browser to make text look ugly.)
The author of the page intends that reader be able to read the material regardless of platform or hardware.
The web page contains informative text
This final point is perhaps the most important. If there isn't any useful information on your web page, then I don't terribly much care to read it, and really don't care if your use of platform-specific tagging prevents me from reading that nothing.
Some people argue that their web page requires Netscape or Internet Explorer for proper viewing. I have seen very few such pages where this proved to be:
My primary heuristic is that if I can't read the page at all using Lynx, then it is fairly unlikely that I would get much more out of using a graphical browser. I am seldom disappointed when I make this assumption.
I use a tool called insert_html_templates which I wrote in the Perl language to rewrite "template" information into my web pages by reprocessing comment lines that I create in a (slightly) special form. This includes:
A common header described via the comment:
<!-- <CB: <head> -->
<!-- <CB: </head> -->
The "header" allows me to put some common information such as a link to my home page and other "style" information at the top of every one of my web pages.
A common footer for each web page:
<!-- <CB: <foot> -->
<!-- <CB: </foot> -->
This includes, at this point, a web search, and some contact information.
<!-- <CB: <indexdirectory> -->
<!-- <CB: </indexdirectory> -->
This gets rebuilt into a list of links to each of the web pages in the current directory, and allows my bookmark page to have a regularly-updated complete list of all of my web pages. None of them get lost.
Google search forms
This makes it easy for users to launch from one of my web pages to the search engine. The form can be initialized with some "default parameters" that are relevant to the web page the user is browsing.
<!-- <CB: <yahoo +term1 term2> -->
<!-- <CB: </yahoo> -->
The equivalent is available for AltaVista and Excite.
Direct link to AltaVista/Yahoo search
Here, the link is created directly; the user has no option to modify the query until they get the initial search engine results.
For AltaVista, we get Excite search for +Edwin +Hoogerbeets
With Google, we get Google search for +x +y
Thus, when composing a web page, I enter a couple of simple comments that are later expanded into a much longer and more complex result (that I do not wish to memorize).
A further benefit is that if access methods or my desired header or footer changes, I don't have to go in and manually edit all of my web pages. I need only change one file, namely the Perl script, head to my public_html directory, and type: % insert_html_templates *.html
All of my web pages are cleaned up in one fell swoop. Old text that got inserted between the comment lines gets thrown away and is replaced by a new set of text.
There are quite a number of more formalized templating systems that involve embedding code in some 'language' into the web page that is evaluated when pages are requested from the web server.
JSP - Java Server Pages
PHP is often used in a "templated" manner; that's really where PHP came from...
A templating system for Python .
Someday it may actually prove useful to use them.
Unfortunately, there is very little support in web browsers to manipulate them usefully. For instance, there is no scheme implemented to represent frame state in a fashion that allows frames to be bookmarked consistently most of the web browsers that support frames. Lynx is, surprisingly, a happy exception to this rule.
Mainsoft Corporation Providers, to Microsoft, of Unix/Win32 portability tools as used with Internet Exploder...
A web design contest with the crucial design constraint that the total size of the site must be no more than 5120 bytes. This ensures a considerable degree of minimalism.
If this was useful, let others know by an Affero rating