There is no way, as far as I can tell, that the file download dialog box can be forced to open for files that the browser tries to display.

Finally, don't forget that the file you've linked to has to be uploaded as well.

Absolute and Relative URLs

An absolute address is a link that enables a file to be found from any part of the web. A relative addresses is a link that identifies the position of a file from the current page or file. It's like the difference between addressing an envelope and giving someone directions. The envelope will get there from wherever you post it, the directions are only valid from one specific place to another.

The address of this page on the brisay.com server is http://www.brisray.com/web/web1.htm This is the absolute address, the URL can be placed anywhere on the net and the document will still display.

Suppose I have a directory structure like this

     brisray.com
          index.html
          biog.htm
          bristol
               batbris.htm
               bemmy1.htm
          optill
               oface1.htm
               omag.htm
          web
               web1.htm
               server1.htm

bristol, optill and web are all folders containing their own files. The visitor is currently looking at web1.htm in the web folder.

Suppose I want to put a link from this page to server1.htm which is also in the web folder.

Using the absolute URL, the link would be <a href="http://www.brisray.com/web/server1.htm">Server 1</a>

Using a relative URL the link would be <a href="server1.htm">Server Information</a>

Suppose I want to put a link from this page to biog.htm which is the root of the webspace

Using the absolute URL, the link would be <a href="http://www.brisray.com/biog.htm">Biography</a>

Using a relative URL, the link would be <a href="../biog.htm">Biography</a>

The ../ simply means from this folder move to the parent. this is important, and will be used in other examples

Suppose I wanted to put a link from this page to batbris.htm in the bristol folder

Using the absolute URL, the link would be <a href="http://www.brisray.com/bristol/batbris.htm">At Bristol</a>

Using a relative URL, the link would be <a href="../bristol/batbris.htm">At Bristol</a>

Again the ../ means go up one folder, but the inclusion of the bristol folder means go up one folder then go in to the bristol folder.

At first sight it would seem simpler just to use absolute URLs, this this isn't the case. I write the pages on my computer before uploading them to the server. My computer contains a duplicate of the folder structure of the website. Using relative URLs means I can test all the internal links before I upload anything. One server I was using for a while went bankrupt and I had to find a new home for my website. If I had used absolute URLs for all my internal links I would have had to change every single one of them to match the new server name. As it was all I had to do was upload the folders and files from my computer to the new server - I didn't have to change any of the links. Another reason for using relative URLs is that I have copies of the site on several servers. Relative URLs mean that I don't have to change the links for any particular server.

Some people argue that relative URLs means that somewhere along the line the browser has to convert them into absolute URLs in order to retrieve the pages. I've never noticed any difference in speed between relative and absolute URLs. Other people argue that some search engine robots won't follow relative URLs to other pages. Again, I've never found this to be so.

Why don't my links work? - HTML

There could be a couple of reasons for this.

A) Was the linked file uploaded?

B) If it was, does the directory structure of the link and the actual directory structure match?

C) Is the casing of the link and the directory structure the same?

A couple of things you should know about web-servers. Most run UNIX not Microsoft software, this makes them behave slightly differently. When you specify the directory structure Microsoft uses "\" but also understands "/". Unix uses "/" but does not understand "\". For example, you may have a directory called "Images" inside a directory called "Web". To ensure you can link into the "Images" directory you should use the form "Images/Web" not "Images\Web" as you would do in DOS.

Unix is also case sensitive, what this means is that MyPic.gif is not the same as mypic.gif. You must ensure the name of the file and the link that refers to it are spelled the same, including the letter casing.

I find it easier to use "relative" linking rather than "absolute." This means rather than typing the whole of the path to a file I can use certain shortcuts. Suppose I had a page in http://members.lycos.co.uk/brisray/dad called dind.htm and I wanted to link to a file called qind.htm in the http://members.lycos.co.uk/brisray/qbasic directory. Rather than put

<a href="http://members.lycos.co.uk/brisray/qbasic/qind.htm"><font color="#0000FF" size="4">QBasic Index</a>

I'd use the ".." shortcut, which means "from this directory go up one," this means I could use

<a href="../qbasic/qind.htm"><font color="#0000FF" size="4"> QBasic Index</a>

There are several advantages to this. It saves typing, because the links are relative to where in the site you are, you can test the links on your own computer before uploading them. If you ever need to change servers, or have the servers changed for some reason (as Lycos / Tripod did in early 2002 when they changed the address from http://members.tripod.co.uk/  to http://members.lycos.co.uk/) you don't have to rewrite all the links for your site.

If you are linking to another site, it may be that that site is down, either temporarily or permanently.

How do I check all my links are working?

My site has several thousand internal and external links, far too many to check by hand. To automate the process of checking them I use Xenu's Link Sleuth. This software checks all the links on your site and reports on their validity.

"Hot Linking" - what is it?

"Hot linking" is the name given to linking to a specific file on someone else's website to provide content for your own. Suppose I found an image called "nicepic.gif" on the site at http://www.anyoldsite.com. Rather than ask permission from that site owner to take a copy of the image and use it on my site, I simply linked to it with 

<img src="htpp://www.anyoldsite.com/nicepic.gif">

This is hot linking, and there are a couple of reasons why you shouldn't do it. "Hot Linking" may be against either your or their hosts Terms and Conditions. This could result in yours, theirs or both websites being disabled without notice. The process increases the amount of bandwidth being used. Rather than simply opening the file on your host, the file has also got to be retrieved from anyoldsite's host. Another reason is that is anyoldsite goes down for any reason, the image you want to display goes with it.

Hot-Linking - How to stop it

If you're allowed to edit the .htaccess file (it should be in the root of your web space) then you can add the following lines of code to it

RewriteEngine on
RewriteCond %{HTTP_REFERER} !^$
RewriteCond %{HTTP_REFERER} !^http://(www\.)?domain.com/.*$ [NC]
RewriteRule \.(gif|jpg)$ - [F]

All you need do is edit "domain.com" to your webspace. For example, this site is hosted at http://members.lycos.co.uk/brisray, all I'd need do is replace "domain.com" with my site address. If you want more than gif's and jpg's protected you can add other file extensions such as (gif|jpg|mp3|mpg). If you want to you can even force the hot-linkers site to display a different image altogether, this is done by adding the following line

RewriteRule \.(gif|jpg)$ http://www.domainname.com/images/stolen.gif[R,L]

Here you obviously replace the http://domainname.com/images/stolen.gif  with the path and image name you want to be displayed.

.htaccess can do more besides. It can password protect directories, redirect URLs, disallow visitors from certain IP's and more. Here's some links to sites with information on these uses.

.htaccess

.htaccess Authentication Tutorial *

.htaccess Files

.htaccess Magic

.htaccess Modifications

.htaccess Tutorial

Baremetal .htacces Files

Comprehensive Guide to .htaccess

Configuring .htaccess

Configuring Your .htaccess file

Preventing Image Bandwidth Theft

What is the difference between URIs, URLs, URNs and URCs?

This can be difficult concept to understand, some people think the terms URI and URL are interchangeable. In many instances they can be, mainly because to most users the difference is academic. Properly speaking though, a URL is a URI but a URI need not necessarily be a URL.

A Uniform Resource Identifier (URI) is a set of standards that identify resources (files, articles, images and so on) on the Web. The standards include Uniform Resource Locator (URL), Uniform Resource Name (URN) and Uniform Resource Characteristics (URC).

Most people are familiar with URLs. This is the normal way of web addressing. An example would be http://www.brisray.com/web/web1.htm which is the URL of this page. It is made up of several parts the protocol, in this case http: examples of other protocols are ftp: and mailto: Next comes the server location in this example www.brisray.com, followed by the path /web and finally by the resource - in this case a HTML document named web1.htm.

URNs and URCs are not commonly seen on the Internet, but they both share certain characteristics. The both attempt to identify resources independently of where they exist. URCs also include descriptions of the resource to make locating that resource easier. Both depend on the use of a Resource Description Framework or Resolver to find the resource.

URNs - the form of a URN is urn:namespace-id:naming-authority:specific-string. Like URLs they are made up of different parts. urn: identify it as a URN. namespace is the registered owner of the URN. For example IETF or WEB3D. The naming authority is a string specific to the DNS. This value is usually the high level domain name for an organization. The format and interpretation of the specific-string is defined by the naming authority.

Examples of URNs are

urn:inet:dstc.edu.au:dstc/tr017
urn:x-dns-2:library.bigstate.edu:aj17-mcc
urn:duns:002372413:annual-report-1997

URCs - the simplest type of URC are lists that only consist of a document title and document location (URL), such as

Record: 210
Title: Table of Contents
URL: http://www.netscape.com/search/

URCs are basically searchable databases of the sort employed by Internet search engines and other organizations.

The sources for the above are :-

Basic URN Service

IANA Registry of URN Namespaces

Metadata deployment for publishing environments

Naming and Addressing: URIs, URLs

Web Stuff | Audio | Basics | Browser | Code Tips | Design | Fonts | Forms | Frames (Page 1), (Page 2) | Images (Page 1), (Page 2) | Links | Random | Redirection | Sound | Video | Search Engines | Lycos UK | Home Server (Page 1), (Page 2), (Page 3)

HomePage | Optical Illusions | War Stories | QBasic | Dads Navy Days | Bristol | Bristol, USA | Bristol, Canada | Terre Haute | Miscellany | Web Stuff | About Ray | Site Map | Site Search | Messages | Credits | Links | Web Rings

This page created 6th March 2004, last modified 28th December 2004

GoStats stats counter