Spud thoughts

[Wookey]

On Sat 07 Oct, Olly Betts wrote:

> * Labelled levels of the survey prefix hierarchy - e.g. area, system, cave,
>   survey book.

Definately a useful concept that will do much to improve useabilty and help
make graphical project-managemnet tools feasible. However note that 'survey
book' does not generally correspond to a hierarchical level. It needs to be
implemneted as an attribute than can be applied to all sorts of bits of data
potentially widely scattered throughout the dataset.

An individual surveyor's book can easily have data in it from Wales,
Yorkshire, Austria and Turkey in it, for example.


> * Concept of relocatable stations - many survey stations are inherently
>   ephemeral and trying to link another survey to one is clearly a mistake
>   (or the station wasn't ephemeral in which case its status is wrong).

Good idea. Any reason why this shouldn't just be another attribute?

> * Polygonal meshes for terrain or scanned chambers.
> 
> Any other possibilities? 

The only one that springs to mind is interesting new survey devices which
might produce a very large set of very short 'leg' segments, with occaisional
'proper stations' along the path (this would be the output from an inertial
device). The 'refinadable stations' idea might be sufficient to deal with
this for of data.

Thinking about what happens for novel new surveying tools might be a good way
of getting back to the fundamentals, or it may just be paralysing :). e.g the
Wakulla-II survey device produced a stream of positions. You can draw legs
between them (for the route it took) but the stations position the legs
rather than the other way round. It also produces a spiral set of wall
positions. This does not fit too well into the model of 'cross sections
attached to legs', although it could be made to fit with a bit of effort.

For these sorts of novel devices we need concepts a bit like 'internal path'
which is something that travels through the cave between the walls (and
current survey centrelines are an example of one) and 'walls' which can be
represented in all sorts of ways (measurements, X-sects, skinned tubes, dot
clouds). Deciding what is part of the dataset and what is just an artifect of
the representation/display (eg skinned tubes could be either) could be
interesting. I certainly think that the plug-in idea, where these sorts of
things are encapsulated as much as possible, is a very good idea, but it's
going to be the interfaces between them where things get sticky...

Anyway, overall Ol's list looks good to me.

Wookey
-- 
Aleph One Ltd, Bottisham, CAMBRIDGE, CB5 9BA, UK  Tel (00 44) 1223 811679
work: http://www.aleph1.co.uk/     play: http://www.chaos.org.uk/~wookey/