ďobo-CADzone, true men don't kill coyotes and use AutoCAD…

Toto je článok (resp. skôr séria článkov), ktorými mal pravdepodobne blog dobo/cadzone začínať. Vzhľadom na pomerne živelný spôsob, s akým som začal postovať články, je najvyšší čas trochu zabrzdiť a poskytnúť ucelenejší rozklad situácie. Seriál CAD a archeológia začína, meine Herrschaften…

This item should be the real beginning of dobo/cadzone. I’ve started a little bit unrestrained so there is more than time to get a more compact analysis. CAD and archaeology – here we go…

The aim of this post (and following ones) is to define position in archaeology/CAD relationship, possibilities in computer aided drafting support and to define ambitiosly the standards for CAD aplication in terrain survey. I really appreciate any reaction in comments bellow. Examples given here are all demostrated on Autodesk application – firstly we have the licence for these solutions, secondly – from a bunch of widely known apps as Autodesk – Solidworks – Catia – Microstation it is just Autodesk which put down it’s foot in a direction that is extremely usable in archaeology (specialised
releases that will be mentioned).

CAD and GIS – a demarcation line

In computer archaeology development there are two main turning points – the first one is formalized and statistical processing of huge amount of terrain data (70. – 80.s) and the second one is connected with a GIS technology which had broken into archaeology in 80s. and blossomed in 90.s. At the very end of 90.s, GIS became the standard surveyor tool and made the way for CADs also (on technological, not philosophical basis).
While GIS is a preminently analytical tool, there is a strong dependence on underlaying CAD data – (cadaster, government land register). In whole Slovak and Czech geodesy it should be appropriate to wear T-shirt with the lettering “I have Microstation therefore I am”.
In terrain archaeology there is a prevailing opinion of CADs appropriate for primary data collection; for analysis and results there is a GIS. In CAD family there are AutoCAD and Microstation like first violin, GIS went from GRASS GIS, Idrisi to today’s “big brother” under ESRI leadership. It is generally valid argument, that CADs are quick and rapid during drawing, and GIS have a bunch of analytical/statistical modules. However there is strong indication of mutual integration last years – CADs are creeping into analytical/topographic solutions (Autodesk Map, Civil, Topobase a MapGuide), on the other side the base drawing and editing in ArcGIS doesn’t resemble pulling the teeth anymore².
Convergence between CAD and GIS is so expresive that it will be probably without any sense to use software according this demarcation line and software brand loyalty will become more important.

Roots

CAD was born (according to tradition) in an electrical and engineering industry, but this is only partially true. These industry branches of course improved technical drawings automatization, but “true CAD” originates from “the model” idea, which as we will see, quite differs from “drawing” or “plan”. The very first calling for the models came from a plastic industry (injection molds). Nevertheless, the rise of Autodesk is a typical american success story about young ingenious guys in father’s garage – everything has been started by californian native John Walker (you know what to drink in Autodesk community). Anyway, since then it is 25 years and everything has changed – computers are hundred times quicker and the engineers today do the models, not the drawings…

In comparison with technical industry, the relationship between CAD and archaeology is different (more difficult):

1. In the industry (engeneering, architecture, industrial design, geodesy,BIM) there are standards and workflow progress defined very clearly. Migration from papers to PC was only formal.
2. Industrial modeling (in most cases) is relying on geometric primitives (e.g. cube, cylinder, line) and their mutual combination. It is not an archaeology case. The most documented entities are the archaeological structures (”deposit”, “mound”, “river flood-plain”) that have irregular shapes.

Moving pictures

First of all we will define the relationship between archaeology and pictures, which is throughout positive, how has L Gal mentioned. I won’t envisage the 3D obssession, that broke into archaeology last ten years; I’d like to cast my eyes to ditinction between bitmap and vector graphics to throw some light on a CAD importance.
Bitmap graphics (in our cases mainly photography) is most truthful presentation of terrain reality until now. The possible restrictions are connected only with the technical limitations of camera (eg. a sensor resolution or memory requirements); photography authentically shows real substance and position of documented entity. The others advantages of photography are quickness, low-cost and availability. Decided disadvantage, on the other side, is it’s comlete absence of interpretative dimension with clearly defined boundaries of the archaeoentities (fig. 1).
Vector graphics (in this concept CAD entities) is an interpreted delineation of archaeological reality with some abstraction degree (fig. 2, 3), corresponding with the idea of reality transfer into points, lines, polygons, surfaces etc.

A realistic bitmap presentation of situation during excavation (fig. 1). There is no posibility to work meaningfully with that without stratigraphic decomposition. After vector graphic completion we can speak about an ideal documentation (”realistic and interpretative presentation together”, fig.2), which is reducible to the vector presentation only (with some degree of abstraction, fig.3).

In context of vectorization and in counterpoint with bitmap graphic I will pay an attention to two things – degree of abstraction (niveau of detail respectively) and vector like an object.

Cobble stone pavement case; or degree of abstraction

During the vectorisation process (assuming that photography is a basis) there appears a question how closely (in detail) the situation should be drawn. This question is present like a “zoom” in technical drawing, but have no relevance in archaeological CAD vectorisation.

Fig. 4 – 6. Archaeoentity – construction (wall). There is a possibility to draw it like a single line (highest degree of abstraction, fig. 4), polygon (besides position and direction there is a construction width also, fig.5), and finally like a polygon with a stone texture (additional info about composition, fig.6). Examples of various degrees of abstraction.

Naturally there is no way to apriori determine the depth of abstraction for whole archaeological documentation generally. But I think that it’s important to recognize that we have dealings with this question also and we should designate the degree of abstraction at least within the frame of particular project (excavation?).

And if You cut it too fat…

… it could fall out this way – fig.7 (above) shows vectorised cobble stone pavement, fig.8 (bellow) only hatch with the label. Although the complete vectorisation represents the whole composition of pavement, it took approximately 4 hours of work. It is questionable if this result is adequate in comparison with acquired information and therefore to use only abstract wildcart presentation.

All is only database, or vector like an object

Very important fact about vectorised entities is their “object” nature. While bitmap is always 1 object as a whole (in its entirety) regardless of how many archaeological phenomena contains (editable subentities are therefore pixels), during CAD vectorisation there is a need for observing a rules of object decomposition (”stratigraphic entities”, “research area polygons” etc.). We must divide the situation according to arbitrary key and CADs entities should correspond to subjects of that particular set. 1 “archaeological object” should be 1 CAD object. (editable subentities are line or polygon vertices).
I am not Autodesk developer .>), and I couldn’t see under proprietary software surface, but it sounds reasonably, that it’s aplication kernel works with predefined object classes and methods aplying on them. With this model You can understand the Autodesk application as a big drawing database. There are two implication of this:

• It is reasonable to keep modeling on the level 1 archaeoentity = 1 model entity
• Assuming that every database could be connected (regardless if it is relational or object one) to another database (very courageous statement .>)), there is no restraint to connect vector data to, say, external database.

This is all for the introduction. In the next article I will take a closer look to topography rules and will also choose the appropriate software fot CAD in archaeology.

1. 1 jedna zo základných rozlišovacích vlastností ovšem asi zostane zachovaná na veky vekúce – a tou je prístup k 3D. Zatiaľčo u CADov je vykresľovací engine natívny 3D, t.j. koordinátny systém vždy disponuje osami x, y, z, v GISoch je natívne 2D a výška je suplovaná atribútom. [↩]
2. 2 of course there is one strong distinctive feature between CADs and GIS, which I think will stay forever – attitude to 3D . While 3D is native for CADs drawing engine – x, y, z axis, it is substituted by attribute in GIS environment, which is always 2D. [↩]