I've tried to keep the structure of BeebMaster as simple as possible, but in case you get lost, the information below on how things work may help. It's also a good reminder for me in case I forget.

I think of the BeebMaster layout as a hierarchical structure, a bit like an ADFS disc in fact. There is the home page, which is like the Root, linked to each of the main sections on the home page. Most of the main sections have sub-sections and some have further sections beneath until the lowest level of individual items is reached.

Clicking on the BeebMaster logo at the top of each page will always return you to my home page. Logos for main sections can also be clicked to go back to the main page for that section.

Each picture set has an index page with small images of all the pictures in the set. Each of these small pictures can be clicked to give a full-size image with further information. In many of the early picture sets, I put "Next" and "Previous" arrows at the bottom of the page to aid progress through the set.

Later I introduced a link at the bottom of some of the index pages to explicitly refer to the parent page in the hierarchy. This is to eliminate the possibility of going round in an endless loop when using the Next/Previous arrows.

In November 2009 I decided I would try to make it simpler to view the pictures in a set by introducing a link on the large pictures to the next picture in the set, with the final picture linking back to the index page. This means that you don't have to scroll down for the Previous and Next arrows to continue, or return to the index page each time to see the next picture.

As from May 2023, there are 3 main navigation methods in use - I call them Basic, Map and Domesday. Basic is as described above, Map is the same except that the top-left corner of the picture will take you to the picture on its own in the browser window without the HTML page surrounding it, so the image can be viewed full size. It's called Map because the links are done using the HTML map function. Domesday navigation is an enhancement of the system I introduced within my Domesday section some years earlier allowing the picture to take you forwards or backwards within the set, if you click in the left part or the right part. This is broadly how navigation works when using the Domesday Discs. In order to maintain being able to see the full-size picture, I have retained this feature by allowing clicking within the middle part of the picture.

Domesday Navigation sets have a green border round the picture in the individual pages to distinguish them. I haven't retrospectively changed the navigation on earlier pages, so there are still some sets without any navigation function at all, and some with Next and Prev buttons at the bottom.

Many of my pages have references in the text to other items on my website and as far as possible I have put in links to the appropriate page so that you can see the item in question with a single click rather than having to search for it later.

At the very bottom of each page I used to have a the "back" option. This has exactly the same effect as pressing the Back button in your internet browser, i.e. you will always be taken to the previous page visited. Originally the "back" link was an explicit link to the referring page, but as the site developed and the number of links to other pages within the text increased, there was a possibility that clicking on the "back" link would take you to a page not previously visited, so I replaced all these explicit references with a relative link back to the previous page viewed. After that I erased all (almost all) of the back links as visitors seem to prefer using their browser buttons to shuffle about.

On the technical side, I write all the HTML myself. From 2008 onwards, this was done using GEdit in Ubuntu, and from 2020 onwards using Kate, At this time, I didn't have "templates" as such, but when I am creating a new page, I would select one of the existing pages to use as a base and save a copy and then work on it. This means that I don't have to start from scratch every time and it also helps to ensure that all the pages are uniform in terms of HTML header information and layout.

For the first time in December 2009, I decided to try to collect together some of the more commonly used bits of HTML, such as the page headers, table styles, etc. into a single "library" document which I can use to cut and paste into new pages to help to ensure uniformity in the page layout.

In 2019, I automated the process of generating the HTML for the picture sets and index pages using a custom-written HTML Generator in BBC BASIC, originally in ARM BASIC but later converted - with hardly any changes needed - to 8-bit BBC BASIC meaning I can run it on any Acorn on my Econet. Accompanying the HTML Generator is my own custom text file format for the captions used in the picture sets.

The HTML pages are stored in different directories within my webspace according to which section they fall into. Picture files are held separately in a sub-directory to the relevant section, usually called "Pics".

I make sure that links to other pages and to image files within the HTML code are always given in relative terms which means I can create all the webpages off-line and have a working local copy for updating and testing. I can then upload the finished pages to my webspace, which has the same directory structure, when they are ready.

Occasionally I need to do a blanket update to many - or sometimes all - my pages of HTML, either to fix some broken code or to apply a change to all pages. I used to use Regexxer, but then I found KFileReplace which I think is easier to use. Unfortunately this was discontinued, so for the last few years, I have been using the "search in files" option in Kate.

In 2011 I started using gFTP to upload files to my website. Previously I had used the FTP facility in Ubuntu's Dolphin for uploading files, but I always found logging in to be a little bit unreliable as sometimes it took a few attempts to get my online storage to display. gFTP is similar to a thing I used to use in Windows called CuteFTP, basically you get a local storage pane and an online storage pane and left and right arrows in between to choose which direction the up- or downloading takes place on selected files.

I was pretty good, but it couldn't cope with my December 2011 blanket update when I tried to copy everything in one go. It didn't seem to understand how to cope with directories, and I didn't want to have to go to the trouble of entering each directory individually, selecting all files and copying them across, so I abandoned it for that particular operation. I went back to Dolphin, although I wasn't able to distinguish between HTML and non-HTML objects, so I had to re-upload the entire site!

I now use Filezilla which is quite happy with recursive uploads. All I have to do is drop my local master directory over the remote one and it updates and replaces as necessary. It usually takes about 10 minutes, and will skip any file that hasn't been modified.

The only difference between my locally stored copy of BeebMaster and the online version is that there are additional items in the online directory "Downloads". This has two purposes. The first is for files I make available for people to download for their own use, such as a couple of disc images and utility routines etc. The second purpose is as a storage area for some files which aren't part of the main website as such, but which I want to send to people or post on a forum, such as a screenshot of something going wrong or an exciting discovery.

The Downloads directory isn't directly accessible by navigating around BeebMaster, but I have set the permissions so that anyone can obtain a directory listing of its contents and access any of the files if they are needed.

I try my best to keep any broken links to a minimum, but they do crop up from time to time. In the past I have had to rely on the kindness of visitors in reporting them to me, but in December 2011, I installed a utility called KLinkStatus which will check all links on a website. It can operate on filing systems as well as online, so I can check the local copy of my website for problems before uploading any changes. KLinkStatus has now been discontinued, but from about February 2015 I started been using Linkchecker GUI to the same effect. The GUI version has now been discontinued so I have been using the command line version since about 2020.

In June 2023, I completed the transition I had started many years before of the renaming of the picture directory within each sub-section's HTML directory. Originally these were all called "GenFiles", and most were renamed in a previous HTML sweep a long time ago, but the only remaining one up to this point was the Dr Who section which still had its picture directory called "WhoFiles". Anticipating many updates in the second half of 2023 to my Dr Who pages, I applied this change in June to ensure HTMLGen works on this section as it does with the others.

Updated 1st June 2023