Every Project’s Content Rollercoaster (come draw one at @RTCBCS…)
Its true that nothing can derail me like a tangent conversation about process mapping and evaluating where pain points are in a workflow… I mean, that’s what this job is all about, right? Tuesday morning, I’m in a groove hammering on some documents, and I get an awesome email from Randall and Parley: “If you were to plot 2 lines/curves independently for Geometry vs Data across a horizontal project phase axis and vertical fidelity axis what would that look like in your mind?”
Its a topic that’s crossed my mind, but never in a straightforward question that made me “map it out.” And that got me thinking to a lot of the problems we have with workflow in our industry, as it relates to Content and content management (not management like “I cant find my stuff,” but management like “your stuff doesn’t talk to my stuff.” Originally, mine looked like this:
Having spent a lot of time helping teams start intense models, generally AFTER a design team has already done some presentations, renderings, concept models, and fly-throughs, always started (for me) with a process like this:
1. Get their stuff
2. Find out how much of the stuff is *real* and *decided*
3. Stuff that isn’t real, break it in to groups (This is where mine drops off mid-project):
3a. Its real, but this probably is not accurate yet
3b. There will be something, but it wont be THIS. This just got thrown in to look good
3c. This isn’t there at all, but it made the image better
3d. This wont really be empty later. Oops.
4. Take the real stuff, and get the team mobilized on building the BIM
5. Realize how empty it is, now that all the stuff from number 3 is in the trash.
6. Spend DD/CD timeframes with the teams, researching what 3a/b/c/d need to actually be, so we can get them done CORRECTLY.
I decided I needed a third graph, for Geometry Quantity versus Quality/fidelity. As I see it, a whole lot of geometry DOES happen very quickly, but maybe not always well.
But in our industry confabs, don’t we always show this cute progression of the chair: It starts at blocky, then it gets legs, then it gets wheels, then it gets Expert-level-detailed. Its wild. If only it really went that way. It really does something like:
Concept: Well in Sketchup, we grabbed it from the 3d warehouse. (Its not the right size, it was just a good looking chair)
Schematic: So there needs to be chairs. In Sketchup we had enough chairs. So I think its good. The client needs this many chairs.
Design Development: Did we put any dimensions on that sketchup chair? The one I got from the manufacturer is way bigger, and I cant get 200 of those in this room. Well lets wait and see if they really pick this chair.
Construction Document: That’s the chair, and it doesn’t fit. Can we make this room bigger?
Surely, I’m writing this in jest, a bit. But a bigger problem I have with content isn’t that its readily available but not precise: Its that it doesn’t *mature,* it retires and gets replaced . Over and over. And every time we replace it we roll the dice on “do we have this correct?”
Its not a software topic, for me. I don’t believe in using Revit’s LOD as a Content “level” solution, it just isn’t built for that. But on a BIGGER scale, could we use something like Dynamo or the API to acquire the traits done in a concept design, and to recompile as a detailed model? Within Revit alone, its a challenge. With interoperability, its a nightmare. With cross-design-phase-team-changes, AND interoperability, AND software complications… Is it possible? I hope so.
We would love to know how you think Geometry Fidelity and Data Fidelity mature as a project goes on. These blank graphs will be around at @RTCBCS. Grab one and give us your thoughts, or MSpaint it and tweet it. Were going to compile them all after, to plot it.
Hope to see you all at @RTCBCS, and at @RTCNA directly afterwards.
-Aaron “Whos chair is this, and where are my socks?” Maller