Web site logical path: [www.psy.gla.ac.uk] [~steve] [equator] [this page]

Notes on the Mackintosh house project

These notes are complementary to Matthew's central description: about the things I would like added to it, not substituted.

First: I would like to explore massive user input. Museums, like lectures etc., take no notice of users at all in the sense that the labels etc. are fixed by the curator, and at best are only indirectly related to estimates of average visitor interests. Constructivism in education theory points out that people only learn by relating new stuff to what is already in their minds: their prior interests, conceptions, misconceptions. Applied here, as a start, we could apply it first to a selection of information tailored to individual visitors; and secondly to recommended "tours": subsets of objects or positions within the house. Most museums (rightly) limit the amount of printed and other information to a drastically small subset of that available. We could acquire a few books' worth and thus have many times more than is normally presented. The visitor then is invited to type in things they are interested in or know, and standard IR engines used to select out a subset of the stored items on that basis.

So simply applying exisiting technology in this context could be a significant winner, just because it allows user input and so interactivity and a personal adaptation (by simple selection) of the material to each user.

Secondly, we should promote "pre-visits" by visitors: who could be using the journey to the house to interact with the WWW and decide what they want to see. If the pre-visit includes extensive VR or a web tour that mirrors the structure of the museum itself, then we can research the comparison between digital world routes and material world routes. Real routes are typically only loosely guided by plans, but the things that catch the eye and cause detours will be different in the digital and material worlds. In fact there are potentially two contrasts to study here: digital vs. material (behaviour, ...), and planned vs. actual routes and behaviour.

Pre-visits are only what many tourists do at one level higher: reading the guidebook for a town and selecting out what they want to see. Such planning is typical for tourists: selecting out a small subset from the many things briefly mentioned in a guide as an outline plan (though contingencies may then lead you to see/visit others as well). Pre-visits are also like the educational technique for promoting learning by getting the learner to prepare themselves (load up the topic into their minds in advance, so that the activity is more effective at making connections between prior knowledge and what is in front of them). And thirdly, pre-visits are simply a good entertainment strategy to heighten enjoyment by anticipation.

Thirdly, technically, adaptation in a strong sense may not be necessary for a while. There's important mileage first in:

• Selecting text for each user individually: fixed "document" collection, individual retrieval/selection.
• Selecting routes / object subsets. Think of creating temporary personal hyper-link sets on an unchanging object / node set. It is in fact already revolutionary to discard the (hitherto technically necessary) tyranny of the editor and open up many different aspects of a city or museum.
  
• Storing user routes/selections/behaviour as possible future data for other users.

Web site logical path: [www.psy.gla.ac.uk] [~steve] [equator] [this page]
[Top of this page]