Glossary

I'm putting this together for you now, but my brain is already so fried that it may bear little resmblance at all to what i really meant to send. I realise that the humble glossary entry may seem a poor candidate for our list of priorities right now, however:

• Making a glossary is similar to indexing, but rather easier, and may give us get going in that direction, as well as being a source of helpful feedback.
• Many of our entity pages will be little more than overgrown glossary entries, in fact glossing entities would be a fine way to start accumulating for our InfoBank.

Altogether, a glossary has the combined benefits of being useful work towards other things, tractable to produce, and something we can get up on the site fairly quickly once we have enough entries, which will be a useful confidence building exercise for all involved (and will help to maintain enthusiasm for collecting data, some of which is clearly not going to be used for quite a while). So ...

The Glossary will largely consist of an alphabetical list of technical concepts and terms with explanations. It is important to focus on the concepts rather than just the words (albeit less so than when generating indices) although the words in the glossary should always be readily identifiable with those seen in the text.

Where an editorial decision is made to use part or all of an article for a glossary entry, this should usually be separated from the process of writing the copy.

There will be three main components to our Glossary then:

1. The list itself, which will look pretty much like a nested DL with some extra links.

2. Links to items in the list from occurrences of the glossed terms in the rest of the site. As a class of links, these should be clearly identified wherever possible, so that browsers know what to expect, e.g. we might use a tiny icon, or perhaps an easily recognisable icon in the margin beside the link.

   This idea of Link types and explainers, will also be vital to us elsewhere, however, in this case it could be possible to approach an ideal situation by using onMouseOver to display immediately in the status bar, not only the word "Gloss" to explain the type of the link, but also and abbreviated (50 to 65 chars should be ok), i think definition (we could call this an epigraph, to go with epitome).

3. Some articles will contan text to which we will wish to refer --- these may be divided up as either passages within an article (e.g. a formal definition) or as references to whole articles. The former should be marked up using <DIV> and <SPAN> elements as suggested below.

The glossary term:

This is intended to looks as much like the phrase in question as possible, but also to give a unique identifier for this entry (a similar for may also be used as an ID/NAME in mark-up). An example is the easiest way to explain how these work: if "mutual" appears a sub-entry of "fund", then the term for this entry is "fund!mutual" (although we may need to alter the "!" for use as SGML ID?).

An (optional) alphabetic sort key ---

this is just in case the glossary term is weird or summat.

The Visual text for the entry

"mutual" in the example given above (i.e. what to print)

An epigraph

(mentioned above)

A possible list of sub-entries:

A nested level of entries is good --- more than 3 levels in total would be excessive.

A possible list of grouped entries

Finally, a list of EXPLANATIONS.