Vol. IX, No. 3

   
November, 1996

Review of AutoSurf - Plus Some Ruminations about File Formats for Scholars

Harrison Eiteljorg, II


A new product for review here is AutoSurf, an add-on program for AutoCAD. AutoSurf provides routines to simplify the creation of complex surfaces in an AutoCAD model. In particular, it makes it very easy to make surfaces for enclosed areas marked by a boundary or such bounded areas with interior holes. The program makes it possible to model virtually any surface that can be described.

Just as I was becoming more familiar with the possibilities of AutoSurf, I decided to see how an AutoCAD drawing, modified with AutoSurf features, would appear when re-opened in AutoCAD. My eagerness to experiment further vanished. The entities created in AutoSurf may or may not be visible in AutoCAD (depending on the state of certain variables when the model was saved), but they are not recognized by AutoCAD. Thus, interior surfacing lines may show on screen, but the surfaces themselves are completely ignored. As a result, hidden line drawings cannot be correctly generated.

In the case of AutoSurf, an add-on program is more than an added set of drawing aids; it is also an added set of drawing entities that can be included in the data file. Unfortunately, AutoCAD alone, that is, without AutoSurf, cannot understand those entities; so the net gain for scholars using the program is questionable. I am reluctant to use AutoSurf for models that will be made available to others, since they would need AutoSurf in order to use them. On the other hand, AutoSurf offers possibilities for making more sophisticated models for rendering; so I may use it for that purpose.

The problem with AutoSurf and its unique entities, when compared with AutoCAD, is symptomatic of the difficulty faced with many current software packages. We often need to have more complex data to express our information, but every change to file structure to accommodate more complex data has costs. The obvious cost is for the required new software; the hidden cost is for extra training. Every such move also has the potential for making old files less useful and for making communication more difficult, since files may not be useful to those scholars who have not also upgraded their software.

So when do we change; when do we embrace new software that enables us to express more complex ideas? Every time something better appears? On a schedule? Only when significant improvements are made? I have no simple answer, but it does seem to me that scholars may be well served by taking a unified approach to this problem. Since our needs are to communicate what we know - more and more often by sharing data files - perhaps we need to find ways to agree about standard file types in a variety of areas. For instance, we might agree that users of AutoCAD will not include AutoSurf entities in their models for the time being. Or we might agree that users of word processors will not use certain fonts that create problems with non-standard characters.

CAD is a particularly difficult example, because the necessary complexity in CAD models has been slower to arrive. But CAD is not unique; all file types evolve and we who use computers are sometimes victims of that evolutionary process. I would very much like to hear from readers about their views of this problem.


Next Article: New AIA Publication Series

Table of Contents for the Nov, 1996 issue of the CSA Newsletter (Vol. 9, no. 3)

Master Index Table of Contents for all CSA Newsletter issues on the Web

Return to CSA Home Page