BIM is a database – Naming is important

Delayed commentary on things learned and shared at a recent RevitDC meeting:

NAMING THINGS and SHARING (it’s a matter of INTEGRITY)
Naming comes up again. “Inconsistent” is the keyword. I am glad that I am not the only one who struggles with it. Shakespeare wrote “What’s in a name? That which we call a rose by any other name would smell as sweet…” ….NOT IN computing and a BIM (building information model) it WON’T!

Naming standards: a concept that could save billions of dollars!
More often than not, architects have provided drawings with non-standard names for the layers and “blocks” (which are nothing more than buckets). And these often contain attributes with yet more non-standard names. Trouble in data-city. [IMO, Autocad's over-reliance on the two “object-types” they call blocks and layers has created a stunting effect. Many former CAD users will need to unlearn them to use Revit.]

Make no mistake about it: CAD is a form of database, (albeit a limited dataset). But for a database to work decently, the rules must be created, followed, and applied consistently. (Ref: Database design manuals and texts) That said, think about the naming and organization that happens in the contract documents, never mind CAD. An article on BIM that I wrote for fellow engineers, but all stakeholders should read.

Should we require database design courses for BUILDING ARCHITECTS?
The building designers (architects) take notice: a change to a layer name can (and has) wreaked havoc on many things. All the way back to my spreadsheets! For example: countless times, even predating CAD, I have been surprised by changes made by an architect on the room names and numbers, and being forced to MANUALLY repair these on the mechanical, electrical and plumbing documents. Not to mention the many calculations that depend on these. Things like lighting and HVAC calculations. (This issue will hopefully die down in a few years, although I had expected this issue to be solved by now)

Blocks and layers are useful for rough collections of data, but they have always been “rough”. Even after 20 years of the technology. This is one reason why Autodesk bought out Revit. They knew that Autocad could not compare to Revit in the evolution of Building information Modeling.

Data organization apparently is not a favorite topic among architects and engineers. It is more interesting to IT pros and a few more forward-looking architects engineers like myself. But this is no less relevant to the building information model (design and contract documents).

So it happens that subject-matter experts and cross-trained folks, like myself, are breaking through the walls (pun intended) that separate the disciplines. Witness the XML schemas being developed by groups like ASHRAE (who are guiding gbXML) and the Associated General Contractors of America (directing agcXML). These standards will make the BIMs easier to use.

DISTRIBUTED DESIGN
Over the years, we have come to share our drawings by the use of the so-called “external reference”. Autocad called them “X-reference” plans; a way of sharing CAD ‘drawings’ between workstations. Many drawings may use or even “depend” on an X-referenced plan. The beauty of it: make a change in one place, and it will propagate to all the dependent views.

The increased popularity of BIM and BIM software demonstrates that a high priority is being given to real-time sharing. An understanding of the true benefit has FINALLY trickled up to many decision-making managers. No more silos of information! The lighting designer, HVAC designer and the architect can collaborate on the glass and shading all the time, not just at the weekly meeting!!! YAY!!! The cost consultants can do their cost analysis on an ongoing basis, not after the design is 80% complete and when the changes are most costly and disruptive.

What happens as the building design is developed – At RevitDC, I learned that Revit doesn’t implement automatic notification to other team members/users of changes. Someone said, “You don’t know what you don’t know”. YIKES! Something is not right about that.

Copy/monitor issues were mentioned. The use of the word “copy” is a confusing terminology that resembles the CAD problems of updating. Are we not talking about a central database?

I also heard about sluggishness. Why is bandwidth still being discussed ? Paying for a fast Internet connection is cheaper than the mistakes a slow connection can cause. At the Revit DC meeting they discuss only Autodesk Revit. So it appears that it has an ongoing problem with its communication architecture. Come on !!!

The distributed computing model came late to Revit, I think. Graphisoft’s Archicad and Bentley’s Building don’t seem to have this problem. Revit was not originally designed for distributed computing using large databases. It was intended for a smaller audience in the beginning. They had a communication issue at some point and it looks and feels like it has some sort of workaround ever since. Maybe Revit is showing signs of stress and strains and may be replaced?

Archicad seems to have the distributed data and updating part smoothed out. (Did they take a cue from Oracle?) I have heard that Archicad updates the distributed data almost seamlessly. Compare that to Revit, which seems to stop and start. Perhaps it goes back to the project data organization? Technical skill of the managers? Or is the Revit data scheme just plain bottlenecked?

CONNECTIONS
Kimon Onuma (of Onuma Software) mentioned the importance of “connections” at the RevitDC presentation. Connections are key indeed, and when I first heard this idea many years ago, I realized that good design creates good connections between objects, and having the objects themselves is only half the task. Learning this, like designing, is a practiced, intuitive, hands-on experience.

Connecting people and ideas is a cool thing.

Tags: BIM, Green Design, Software