OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

dita-help message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: RE: [dita-help] Layering in Help Systems

Thanks for your thoughts, Matthew.

You raised a particularly interesting point about auto-linking of glossary
terms, and the argument for only implementing an expansion link for the
*first* occurrence of a specific term. I don't think that would present too
much of a problem in a processing routine, because XSL-T (or probably more
correctly XPath) has a facility for picking out the first instance of an


-----Original Message-----
From: Matthew Ellison [mailto:matthew@ellisonconsulting.com] 
Sent: Tuesday, 6 January 2009 9:31 PM
To: dita-help@lists.oasis-open.org
Subject: RE: [dita-help] Layering in Help Systems

Greetings Tony, and Happy 2009 to you too!

Thanks for starting the new year with such an interesting question.  I
believe that information layering is a very important technique in user
assistance since our aim is to make the best possible use of the (usually
quite restricted) screen real estate that we have available.  Also, hiding
information and making it available only "on request" is a great way of
keeping Help topics concise and more inviting, especially for novice users.

I think it is possible to create a fairly rigid set of rules for when
information layering should be applied.  In response to your example: I
would say that, having established a convention for presenting footnotes,
this convention should be applied consistently throughout.  The only
argument I can think of against using a popup in a specific instance would
be if the amount of information displayed was too large -- but then, in that
case it probably wouldn't have been appropriate to mark it up as a footnote
in the DITA source.

Taking your other example of glossary terms, there is an argument for only
implementing an expansion link for the *first* occurrence of a specific
glossary term within any given topic.  All of the Help Authoring Tools that
support glossary term linking provide this as an option.  They also
typically enable "ad hoc" glossary term linking, but I don't see any good
reason for this.

Other examples of information types that might be implemented as layers are:

-- Images: these take up valuable space within a user assistance window and
are often omitted altogether -- however, it might be a good idea to make
them "user selectable".  Arguably it is less possible or desirable to apply
this as a rigid rule since there are likely to be specific images (perhaps
icons and buttons) that are critical to the effectiveness of the user
assistance and should not be layered

-- Tips: If layered, then they should have a meaningful and explicit hotspot
or trigger (not simply the word "Tip" since this provides no clue as to the
nature or relevance of the information)

-- Examples

-- Task sub-steps

-- Task main steps in cases where multiple tasks are chunked into a single
topic -- this rule could be difficult to implement since it would not make
sense to layer task steps in the more usual case of a task topic consisting
of a single set of steps.

Having said the above, my primary aim when applying information layers
within user assistance is to ensure that the entire topic (beginning and
end, but not necessarily all the detail in between) can be displayed to the
user without scrolling being required.  If the topic consisted *only* of a
short example, then it would not make sense (from a usability perspective)
to layer it.  On the other hand, if a topic consisted of six headings, then
it might make sense to layer the content of those six sections irrespective
of the nature of the information.  This suggests that, purely from the
perspective of usability, a Help-centric "layering" element might be useful.
However, on balance I would not be in favour of this due to the lack of
structure and predictability that this would introduce.

Best regards,


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]