OT: Industrial Design and CAD

Design Intent...as hacked a phrased as it maybe, there is that possibliity of just doing arbitrary design with out taking anything into consideration. (ie just to get it out the door, get it done, what have you.) So is thinking of nothing, while desiging, a conscience choice and a design intent?

When I speak of it in context of CAD, design intent is not just some hackphrase. Sure you could have wanted that square to be extruded with the 5* taper. But what happens down the road when you start building off the fact that the base is 5* and you find out that it was supposed to 7*. You make that change and if not done properly, everythign else fails. Thus not maintaining your design intent. Knowing what you want to do and excuting it are two completely seperate, though interwoven as they may be, spereate. THis is kinf of leaning towards what Ed wrote about whatchign how designers change according to what is on the line.

Of course when you think of your product and you want it to look a certain way you put thought into it. How it will look and the likes. Getting that to translate and change properly so that it survives different iterations in the 3D modeling world is something else.

To be clear-- that was in no way a slam on anybody in particular or intended to denigrate anyone.

Arthur Y-S wrote:

supposed

"design intent" translated (from above) equates to "intent intent", isn't that a double negative? did i miss something in the translation? maybe i need a translation of the translation. ;)

"Intent" is a lousy synonym for "design". "Design" carries the implication of synthesizing a group circumsstances, factors, objects, etc. to achieve a goal or goals. The goal or goals would be "intent". Design intent in CAD refers to the collection of individual goals that add up to an overall design. I can have an intent that the draft angle be 2°, but that is only one intention within the over all design. Said another way, a design is a collection of intentions.

A thesaurus is a dangerous thing which dilutes the language. By equivocating the meanings of words, the distinctions between them blur, making it more difficult to convey specific meanings. So if you have a decent vocabulary and can use the specific differences between "synonyms", people don't understand you. So we have to use a whole sentence or paragraph to do what one word could do. Thus, a fabulously polular series of books for dummies, all of which coud have been less than one tenth the thickness and conveyed just as much information.

People are just too darn lazy to think about a text anymore, so we have dumbed-down help for our CAD tools that are next to useless. The writers are afraid to give a dense explanation of what the tools do and how to use them.

I feel slightly better now.

I actually agree with whoever originally dissented against this dumb phrase. Whenever I'm in a position where I'm tempted by some Dilbertian impulse to use the phrase "design intent", I instead substitute the phrase "Design for Change", which to me seems much clearer and to the point.

"Design Intent" makes it sound like the design doesn't speak for itself, so you have to explain your intention, and hope that makes a difference somehow. Look at the awkward explanation of "design intent" in the SW training manuals.

Anyway, you design something once, but you change it many times, so how it reacts to change is more important than how you originally build it.

matt

Dale Dunn wrote in news:Xns9591B4415E7C2daledunnatjamestoolc@65.24.7.150:

I've always figured it was because if there were concise, comprehensive descriptions in Help promise the moon advertising (not thinking about SW specifically) wouldn't stand a chance if it were necessary to defend it in court.

Sometimes ya just gotta let it out. 8~)

I've come to look at the semantics of 'design intent' this way: What do you want the design to do (yes, what is your intent for this design)? This is the question that I ask every time I sit down to help someone with a modeling problem, or when I do my own stuff. Because a 'design' can just as easily NOT fulfill its ultimate goals if you do not vigorously interrogate it.

For instance, there are a good number of ways to dimension an elongated (pellet shaped) slot. Is the critical dim from center to center, or from end to end? Is it located from centerline of the slot to an edge, the edge of the slot to the edge of the part, or centerline of the part to the centerline of the slot? Etc.

This design information might have nothing to do with change - if I said 'design for change' I could get hit with the statement that the part is not going to change. THAT is a very CAD-centric view, and I want to try to maintain a product-centric view (neat tie in with an earlier part of the thread, eh?). My questions relate to the function of the part as it will be inspected when it comes from the vendor (because, of course, we use design dimensions in the drawing) Put simply - as designed, will the part absolutely work if you make it this way?

Of course, the slot is a really simple sample. It gets more complicated on complicated geometries, but the question is still the same - what do you intend this feature, face, geoemtry to do on the part so we can design it in SolidWorks to insure that intent is preserved and communicated to anyone who touches it, whether they are making changes (to your point) or just manufacturing it (to mine)

Yes, I used to hate hearing 'design intent', 'design intent', 'design intent', repeated by VAR reps. But then I stopped to think it through and found it to be such a useful phrase that I have embraced it. Heck, I will be working on my wifes resume this weekend and every time she asks for some formatting quirk I will ask her 'what do you want this to do/communicate/accomplish, etc - e.g., what is your design intent? so I can be sure that the formatting she is asking for actually achieves the result she REALLY wants. Because to her, the 'design' is how it looks, while in reality, looks are just part of a greater goal

So, I am going to continue to use the term 'design intent' in all of its theoretically redundant glory because, in my experience, it does mean something unique that 'Design' or 'design for change' would not capture.

-Ed

p.s. I hope this doesn't make me a dummy if we run into each other again. I still think you're smart, and look forward to our next meeting even if you will be looking down on me :).

Ed,

I could never look down on you, you're much taller than I am. ;o)

I understand what you're saying, and I agree. I just found a different term that means more to me.

matt

"Edward T Eaton" wrote in news: snipped-for-privacy@uni-berlin.de:

Another contention of mine is that I almost always use different schemes to dimension the sketch vs the dimensions I put on the drawing. The way it's dimensioned in the sketch only deals with how I want to control change, which almost never coincides with how the shop wants to see the dims. There are certain types of change that are impossible if it is dimensioned a certain way, or if you use sketch relations instead of dims.

Anyway, maybe it's just my quirky way of looking at it.

Can I take back that "dumb" word?

matt

"Edward T Eaton" wrote in news: snipped-for-privacy@uni-berlin.de:

That is a constant struggle for me in our office.

The policy is that the part drives the drawing, and the drawing of course drives manufacture. On simple parts its no big deal, but on complicated parts this policy can be tough when SolidWorks pushes back. I have to admit that I have to take heat with some of my 'reliable' modeling techniques that don't promote easy drawings, or when I attempt 'drawing friendly' design that risk burning extra time or risk design conflicts when the models go through predictable changes. We frequently struggle for better ways to accomplish both goals (reliably drive the drawing AND the model). It can be anywhere between a relatively simple head-scratcher and a miserable bitch of a problem, and to get back to Arthur's origination of this thread, it is clear that CAD isn't truly mature until we don't slam our foreheads OVER AND OVER AND OVER agian against this quandry.

Way, Way, Wayyyyyyy off topic, Weird thing about being kinda tall (99th percentile)... You never notice the height of the folks around you, because everyone is lower. You develop different metrics, and it only registers when you see yourself in a mirror next to the person you are standing with. In my brain, strength of personality dictates the relative heights of the folks around me . I once dated a girl who was 4'10" and didnt think about how wierd that was because she had such a powerful presence that she seemed taller than pretty much everyone else in a room. And my high school footbal coach, only an inch or two shorter than me, seemed very much smaller. Example you will appreciate - I have hung out with McElenay a couple of times (recommended, by the way) and have been told he is short, but could not verify because it never registered. God forbid my ever getting called into a lineup...

If I might inject my left handed two pence worth....

Easy drawing creation.... all ya gotta do is "show dimensions" ...... 3D vendors trying to entice flatlanders. Where does trying to realize the goal become counter productive? Very low on the ladder that goes from discretely modeling and assembling standard structural shapes to associative top down design of complex assemblies. (Y14.41, an emerging standard, isn't about showing dimensions that drive the model, it's about adding annotations in 3D space, hopefully negating the need for 2D representations.)

It still won't be mature when donning a sensory cap results in some sort of physical prototype. Arthur's "what is needed to get the job done "right"", I think deserves an affirmative response. CAD (broad term) is "there" in the sense that compromising, fluffing details, settling for approximations because of system limitations is pretty much a thing of the past. All vendors are, of course, not capable of all things and whether napkin sketches are, or are not, more efficient is a much harder question to answer. 8~)

On this we are in 100% agreement. I see no reason to categorically design my models according to the constraints of 2d drawing (even worse, SW 2d drawing). Incentive to do this decreases more as the 2d tools mature to make drawings even easier. For example, I really like the new autodimension tools in 05. They are never right, but they give me a heck of a head start, particualrly where ordinate dimensions are what the machinist needs. I'm certainly not going to try to model a whole part with ordinates, especially when it's easy to put them in later. Basically, the manufacturing design (drawings) does not dictate all of the functional design (models).

Unfortunately, I recently found out that the way I draw, I rely too much on habit. Most of my parts are machined in-house on a mill or a lathe, in processes that I understand pretty well. Well, on this particular project, it made sense to use cast polyurethane parts. I dimensioned them according to habit, and those drawings were almost completely useless for conveying to anybody how to inspect the parts, or even what was critical to function. They were ok for the machining we needed to do to them, but useless to the vendor making them for us. They did not convey what we really needed the part to be. A lesson learned. It would have been better for me to use the model's dimensions.

Interesting. Without really thinking about this, I had already developed the habit of trying to get my hands on a user's manual rather than relying on the advertising to get a feel for what a product could really do.

I too try to resist buzzword compliance. I just so happens that the phrase "design intent" immediately made sense to me as a focus for using parametric modelers. There was not even a hint of redundancy in my mind.

I guess there is no way to predict how a certain phrase will be received when thrown out into the wild. I once read an article about companies who consult on "identity" projects, like naming cars. These are the people responsible for names like "Camry". They make no sense, but somehow they capture the sense of what the marketing department is trying to sell. Well one of these companies (10,000 Monkeys, I think) was making a presentation that included the fabricated word "jamcracker". Some women in the meeting found the word offensive, and even asked that the word be taken down in the meeting. How was anyone supposed to predict that "jamcracker" was offensive? It means nothing, but to some people, it sounds offensive. I don't mean that "design intent" pushes some emoptional button. I just think it's interesting that a phrase will be received differently by different people. It's more subtle than the famous Chevy Nova blunder in Mexico.

By the way, the word "jamcracker" is owned by that company.

I think of a "design" as what a product looks like and "design intent" as how you want the product to behave and/or what you want it to accomplish.

Example...

DESIGN: This television has a rectangular DESIGN.

DESIGN INTENT: The DESIGN INTENT of this television is to give the viewer a more panoramic view by means of a rectangular shape.

or

The rectangular DESIGN of this television is INTENDED to give a more panoramic view.

Obviously the word 'design' can mean several things, so I'm just giving an opinion.

Mike Wilson

But.... you have to get your hands on the product because, most of the time, documentation is so shallow. And, how long does it take to pick up a completely new program and learn enough about it to really probe the depths of it's capabilities? Literally years if you happen to be someone that's making the transition from 2D to 3D? (Is there a clue there?)

iDunno. It's always seemed to me that superb documentation should be (1*) a great marketing tool and (2**) of negligible cost in the grand scheme of things. So is some "irrational" reason (conspiracy theory #?) behind the lack of it? I'm sure there is profit potential; training, tech support, etc., that's supported by lack of documentation.

• One of the, rumor has it, hard to learn CAD programs has some excellent Help and supporting documentation. Learning the program is, in reality, not much harder than learning the more "intuitive" programs when that documentation is referenced. Formal training might be more efficient, but it's not an absolute necessity.

** A really inexpensive surfacing program has about the best, most comprehensive on-line Help I've seen in any design software.

Another thought / possibility: Some vendors actually develop almost every aspect of their product and others rely more on integrating licensed or purchased modules. It will be a lot harder for the latter to create really good documentation as they don't know any more (well, not a lot more) about the "nuts and bolts" than some users (?).

Don't know; just airing some thoughts.... er, brain farts?

So, by extension, we can safely assume the honda element was done in china on turbocrap ?

Look closely and you will discover that it is very carefully modeled with curvature continuous surfaces (plastic and metal parts). the result is no undesirable highlights along the intersection of the radii and the faces. Subtle, but essential for the success of such a minimal product.