"Morris Dovey" wrote in message
...
|
| Well, I still have my T-square and drawing table; but I
| really am more productive with the CAD tools.

As are many designers. If you have good CAD tools and you know how to use
them, by all means use them. I looked at your sketches and they're quite
appropriate for the tool, and presumably for the task. Most importantly,
you get things like offset lines for free. Those are tedious to draw. And
if you have patterns that can be described algorithmically, you can get more
done quickly if you're allowed to do that. You mention that your tool is
scriptable. AutoCAD is too. The systems I've helped build are consummately
scriptable, having been built upon a full-featured procedural programming
languages.

For example, to create a radial pattern of holes, such as for a flange, you
could say:

outerFlangeBoundary := circle( centerPt, flangeOuterRadius );
innerFlangeBoundary := circleConcentric( outerFlangeBoundary,
-flangeLipWidth );
draw( outerFlangeBoundary, innerFlangeBoundary );
prototypeHole := simpleBore( holeDiam, normalDir( outerFlangeBoundary ),
HOLE_THROUGH );
holePattern := radialPattern( centerPt,
flangeOuterRadius - (flangeLipWidth / 2),
numHolesNeeded,
prototypeHole );

Or you could use the GUI. But the textual representation allows you to
capture the geometric relationships so that there's a basis for generating
the portions of the drawing that are defined as dependent upon the other
parts according to arbitrarily complex geometric and algebraic constraints.
In plain English, the above would read, "Drill a set of evenly spaced holes
halfway between the inner and outer flange boundaries."

These tools are incredibly powerful and enable modern engineering. But they
have daunting learning curves for newcomers. If you're going to run a
professional shop, regardless of the material you work in, you need that
kind of flexibility and power to stay competitive. But if you're just
putting cabinets in the bathroom and you don't already know how to design
this way, you're better off with pencil and paper.

Most people already know how to use a pencil, and it doesn't take much to
realize that you can draw a little dimension arrow that says, "5/8 inch" and
that keeps track of how big something has to be.

| I don't think I could have managed the design with pencil
| and paper only.

Perhaps, but see below.

| /I/ think the end result will be better than bland.

Definitely. It's not at all bland. I like it.

| These two sketches represent about an hour of effort.

Well, yes and no. It took you that long to produce the drawing, but you
first had to learn the system in order to make it do that. That's the
essence of my point. If you merely want to make one inlay, learning a
software system is overkill. If you plan on making many, many overlays,
learning a design system is a wise investment. Most people already know how
to use a pencil and a straightedge, so if that's all they need then that's
all they should use. The question is whether or not you're happy with your
tools. You shouldn't dive into CAD simply because it's there. You should
dive into it if it's going to get you what you want.

| I guarantee that I couldn't have managed them in
| an hour (or even five) with just pencil and paper.

It would probably take me half an hour to do the tool path with traditional
instruments. I can already see the construction lines, etc., that I'd have
to draw. The double-lined version would take longer. But that's taking
into account seeing what it's supposed to look like. I certainly could do
it in far less than five hours. But then again I was trained in the
traditional methods of geometric construction on paper, so perhaps when I
recommend pencil and paper it's with that knowledge in mind. It would be
fair to say I'm adept in it because I've invested time to learn *that* mode
of drawing too.

| Hmm. The tool set isn't bad at this stage; and it's getting
| better all the time. Less expensive too.

I agree. Especially with various parametric curves and interpolating
techniques you can leap ahead of many pencil techniques. We were just
talking about old shipwrights and splines and ducks. These days
interpolating through points or drawing freeform curves of appropriate
"bendiness" is drag-and-drop. A while back we had a discussion about
scribing free form contours such as in making boat bulkheads. It's much
easier to instruct a CAD system to extract the appropriate curve from a
freeform surface.

CAD has distinct advantages over paper design. I'm just wondering how much
of that helps the novice designer who just wants to make a chair or two.
I've seen motivated people bog down in the limitations of their tools.

| Re-use of design elements isn't necessarily bad

In the functional sense, it's absolutely essential. If your design library
already contains a bracket or a caster assembly or an inlay pattern --
complete with tool paths and assembly steps -- you should use it. It will
save you time and money.

But where aesthetics are important, such as in residential and furniture
design, re-use isn't always good. It leads to the "bland" look if you're
not careful. Sometimes re-use is good if you want a product line -- a
dining room table and its associated chairs, for example -- to have some
sort of dinstinctive look and feel. You might duplicate the table's center
inlay on the back of each chair. But that doesn't always translate to a cut
and paste in a CAD system.

| I think that needs to be considered on a case by case basis.

I agree. Some people enjoy the design aspect more than the production
aspect. Some people would rather spend time making sawdust than pushing
mice or pencils. Some people need everything spelled out for them in order
to have confidence to actually do it. Some people don't need to know just
how big a pilot hole is going to be or how deeply a screw will penetrate.

You can underdesign, and you can overdesign. Again, my advice to the novice
is not to design any farther than you think you need to go. And so don't
get wrapped up in the mechanism of expressing the design. If you don't need
a full layout plus isometric, don't worry about making it. If you make a
design from a dimensioned sketch and it looks like a Picasso, spend a bit
more time on paper.

| Me me me! I do! (-: I don't know about his chairs; but I get
| quite a buzz from his approach to design.

Don't get me wrong; I admire Wright immensely. The point is that someone so
well known for design can merit legitimate criticism. Wright's chairs are
unquestionable works of art. But they don't fit the human body. Many of
his early chairs have straight vertical backs. To sit in one is
excruciating. I'm only 5'4" (basically no taller than Wright himself) so I
feel quite at home in Wright rooms. My good friend is 6'4" and can't speak
about some Wright ceilings without using profanity.

Grumble all you want, but time has proven that Wright's ideas about varying
the ceiling height are right on the mark. His ideas about using native
materials are right on the mark. Many of his ideas, which he fought hard
for during his career, are now commonplace in architecture.

And there can be no question that Usonic design and CAD would go hand in
hand. Wright wanted design reuse and spent a large portion of his career
advocating it and practicing it. He never got it quite right, but the
notion of "parameterizing" a design in order to hold down its cost -- a
feature of most modern CAD systems -- would have appealed greatly to him.

But if the hobbyist merely wants to create an attractive picnic table or
jewelry box as a gift, or a writing table to fit a custom space in his home,
then the overhead of a parameterized design may be too much work.

| I will argue (forever) that learning to understand the materials
| and the tools is a necessary step...

Yes.

| This may be where we differ most. I prefer to have all design
| issues resolved before I start making sawdust.

As do I. My point in bringing up design on the fly is that it was
surprisingly effective in that one case. It's wasteful and dangerous as a
general method of design. My clock could have just as easily turned out as
a pile of odd maple burl scraps. I took that risk. Not everyone should,
nor certainly on every project. But there's a certain freedom in "drawing"
with a bandsaw blade on wood.

The true path, I believe, lies somewhere between extremes. What I want to
instill in the minds of design novices is neither recklessness nor anal
retentativity. I believe that some people looking for a CAD system are
really looking for guidance on how to approach the problem of design. I
would want to warn them away from CAD if their thought is that it's a magic
bullet. The best thing I believe I can offer people is confidence in their
innate design ability, regardless of the tools used. I want people to look
for design expertise where it really lies, not in some shrink-wrapped
package.

--Jay