From COMP.TEXT.SGML 1996-02-29
I'm intresting in flexibility, structure and how hypertext should be.
Jorn Barger wrote:
We've talked about:
- Latency: if it takes too long for a definition to be retrieved, it's probably better to put it inline, or at least on the same page
That should be up to the user to decide. If he is intrested, he wait for it to load. His browser should be abel to download all links of special types. Maby all definitions. Retrival time should never be considered in the konstruction of the document.
- Markup: it's inefficient to mark every occurence of a defined term with a definition link
With HTML today, you could use automatic replacement of special words to include the link. But i think what you should not have to mark it, other than in special cases. If the document included a description of the definitions used (a list in the head of the links to special words (or terms or list of terms) defined or a link to this list), the browser could display this terms any way i wishes. Maby by putting an astrix after the term. The special cases mentioned abow is then the the term to define is stated in an unusual way that woudn't be recognized by the browser. This could either be a reference to the original spelling of the term, or it could be the link to the document. The first alternative is very much better.
- Page layout: highlighting defined terms can be distracting
This is up to the browser. You shoud be abel to change how links are displayd. Maby compleatly hide links (not the words) of some sorts. (Like links to definitions you already followed.
- Disorientation: a popup window imposes a cost in disorientation
This shoud be up to the browser.
A intresting hypertext function is the enlargement of links. (I don't know what it's called.) If you select this link, the text will be substituted with the definition. In this way you could incorporate a definition in the document in a way what a reader is used to from papers. The definition could be put in parenthesis or the linked text could be changed to be more explanatory. A browser could handel this in several ways: It could uses special markers to indicate what portions is enlargements and what portions that could be enlarged. If the document is properly typed in regard to the content and difficulty, the browser could automaticly show some levels of enlargements to fit the users skill level in the subject. Of course should you be abel to tell the browser to give you a popup window instead of an enlargement.
Popup windows can be displayd in diffrent forms. Where they showes up and how they looks is significant. I think that the browser should use popup windows by default if the document only contain a few words. For those documents, the popup window should be minimal. It shoud not have button-rows, or anyting at all, just like the Windows help popup windows. It shoud be easy to get rid off and also easy to move around.
It shoud not hide the text that gives it its context. But it should also appear near to the selected term. This means that the window shoud apear directly abow, below, before or after the term, in a way that hide as littel as possible of the context. If it is the last line of a paragraph, put the window below. If it is the first line, put it abow. If the paragraph is big, look at the sentence. If the computer or the browser programmer is to lazy to do this, put it anywhere!
An alternative to the selective display positioning is an static behaviour, that makes it easier to directly beginning reading its content. To accomplishing this it could always appear in the center of the screen or at the lower right of the term (in a way that makes the beginning of the popup window content to be a direct continuation of the present line). I dont know wich alternative to choose.
- Custom/Generic definitions: a universal dictionary will be less useful than a customized glossary for a given document
This should be flexible! I suggest a several-layer solution:
Many sorts of links would be marked up as usual (like in HTML). The repeatedly used terms, or the terms that has explanatios uniquely for the context, should be gathered in a reference document, like described under the Markup topic. If a term mentioned in the definition list also is a real link, the selection of the term shoud invoke the real link. But the precense of a definition could be displayed anyway (as explained) and the definition link could be followed by a special action (right mouse button).
If a word is not a definiton and not a real link, the browser shoud use a dictionary of some sort. The first choise will be to use a context-related dicionary. The document shoud be typed in regard to its context. (i.e. The topic of the document shoud be stated in a formal way.) If the document doesn't give any context, the browser could generate the information by analysing the content (using word count or a better technique). This context should be used for selecting the dictionary to use (among many other things). The browser could have a list of diffrent dictionaries, lexicons, resources, etc., for diffrent occasions and contexts. The document could suggest a special dictionary or resource to use and the browser or user could choose to ignore it, follow the suggestion instead of his own alternative, or automaticly add the dictionary to the list.
And there should be a standard dictionary for thoose contexts that hasn't any special dictionarys, and fore words that isn't included in the special dictionary. This dictionary could automaticly be used by the special dictionary, or could be one of the browsers choise. Any of the dictionarys could either be downloaded and used as a local program, or anywhere.
When a reader is made aware that a definition is available, it's not a simple question whether it will be worth visiting.
... you never feel confident that you're not missing something by skipping the definition!
This is diffrent from person to person. Some reads all the foot notes, some don't. It depends on how much time you have and what it is you want to read about.