Rules of Thumb - Sheldon Wolfe

    A huge problem that continues to grow is that we have too much information. When American architects formed AIA, 150 years ago, construction was much simpler; mechanical systems hadn't changed much since the Romans used them 2,000 years ago. Since then, countless new materials and processes have been introduced.

    Life was simple for architects of those early years, much of their time being spent detailing ornamentation. In 1905, a local university building of 112,000 square feet was built using a steel frame, with brick, marble, granite, and terra cotta. The construction documents comprised 58 drawing sheets and a 51 page project manual. By today's standard practice, hundreds of pages of drawings and a project manual of at least two volumes.

    We all know that, at least in theory, today's designers must understand and comply with a growing collection of building codes, local regulations, and zoning requirements; they must keep abreast of the latest in building materials; and they must know what's in the standards published by many organizations. No easy task, this - in fact, it's impossible - so we focus on the big things and hope for the best. To keep things moving, we must carry in our heads the really important stuff, the rules of thumb. Following is a collection of such rules I have offered to young professionals for many years.

    What to draw. If it comes in a box, don’t waste time detailing it. Do spend time showing how it fits in. Example: Don’t draw detailed sections of windows, with all of the pieces that make up the sash and frame; do make sure to detail how the window fits in the opening and how it is flashed.

    Draw only what is needed; but draw everything that is needed. This takes a little thought, but helps the drawings get done right the first time. And, it helps the bidders, who don’t have to wade through a lot of information that isn’t necessary to find what it is we really want.

    Where does the information go? People who work at the site don’t even carry specs, let alone read them. Put the information they need on the drawings, and everything else in the specs.

    Defined terms. If defined in the contract documents, the terms furnishinstall, and provide can have distinct meanings. While the difference between furnish and install is fairly obvious, the common definition of provide is not, so avoid problems by using furnish and install rather than provide. In a single-prime contract, there is only one contractor, but there may be many subcontractors.

    Drawing notes. General drawing notes often repeat, and often contradict, each other, as well as the project manual. Eliminate redundant notes. Use the same term for a given product throughout; use the same term that appears in the specifications. Used too often, "Unless noted otherwise" suggests you don't know what's in your own documents; how can the contractor be expected to know? Why preface some notes with the word “Note”? Ask yourself what each note means. Example: “Fill with concrete and paint.” Notes such as “fasten securely” and “see specs” are unnecessary. Don't use brand names. There is no need to say "Provide countertop" or "Install trim"; just indicate what the product is.

    Assignment of work. That's part of the contractor's job.


    I have a spelling checker, it came with my PC.
    It plainly marks four my revue, mistakes I cannot sea.
    I've run this poem threw it, I'm sure your please two no;
    Its letter perfect in it's weigh, My checker tolled me sew.

    © 2015, Sheldon Wolfe, RA, FCSI, CCS, CCCA, CSC

    Agree? Disagree? Leave your comments at

    Time to brush up your resume? - Sheldon Wolfe

    It's been ten years since my firm took the plunge and began moving from AutoCAD to Revit. There was a lot of behind-the-scenes research and discussion in the preceding year, after which a test team was assembled and trained. A real project was selected for live-fire testing, and we were on the way. About two years later, we did our first all-discipline project. In the next two years, the entire production staff received a full week of training. By the time the economy collapsed in 2008, Revit was our primary program, and today, it is used for virtually all of our work.

    When the decision was made to commit to Revit, a few of our users made a presentation to the rest of the office, showing some of BIM's capabilities. Many of those who watched were impressed by a simple demonstration that showed simultaneously a plan, an elevation, and an isometric view of part of a model. The presenter showed that moving a door in any one of the views changed the other views in real time. 

    As I watched, I remember thinking, "Someone is going to be out of a job." It should be no secret that, as firms become more familiar and more efficient in their use of BIM software, they will no longer need those people who formerly translated the changes made on one drawing to related parts of other drawings. From there, it's not difficult to imagine a program, or a collection of integrated programs, that would allow a single designer to operate without any support staff. Carry that thought a bit further, and it is quite possible to do away with structural, mechanical, and electrical engineers.

    We all like to think we're essential, but computers and automation have been putting people out of work for a long time, and it seems the rate is increasing. And, even though many people accept this as fact, it's common for them to believe that their jobs are safe. But are they?

    Nearly anything that is repetitious is now done by machines, controlled by computers. Entire factories now require only a few humans to watch the process, and even their jobs are in danger. It's interesting that many of the jobs left to humans are basic services, or manual jobs that are too varied or complex for computers - at least for the moment. In high school, I worked in a Ward's warehouse, a huge building full of thousands of products. At the time, it would have been difficult to conceive of a way that machines could find, select, and deliver those products as well as a human. To see how even these jobs are being replaced, watch this video about Amazon's new warehouse: The only humans still at work are stuffing shipping boxes, something a computer will probably be able to do within a couple of years.

    Some people argue that all of this automation frees us from menial work, and will allow us to pursue more interesting work. That may be true, but in most cases, the people put out of work cannot simply move on to a job that requires more education and experience. That's clear in the case of those who work in warehouses or factories, but it's also true of people with years of college education and experience. Will the staff architect move on to become a programmer for AutoDesk? Possibly, but not without more education.

    The problem is, computers are not limited to simple jobs. If you can define how to do something, you can program a computer to do the same thing. Watching robots in an assembly line, it's clear they can perform complex operations. And while computers and robots once were built to do just a few things, current models can be reprogrammed as required for different jobs, and some now are able to learn and reprogram themselves.

    What about your job? We talked about staff architects already, but what about engineers? They already rely on computers to do all the calculations that were done manually many years ago. Don't you think it's possible for a computer to analyze a BIM model, evaluate various structural systems, and choose the one that's best for the project? Couldn't the computer also be able to compare several HVAC systems, plumbing designs, and electrical options, and choose the best? Someone may have to tell the computer if cost or performance is more important, but even that decision could be automated. Hardware specifiers amaze me with all they know, but again, if you can describe how they decide which hardware to use, a computer can do the same thing - and it can be done in the architect's office.

    Surely, there is no way to completely eliminate architects! Don't be too sure. Early in October, I watched an interesting video ( that discussed the possibility of a computer completely designing a building based on program requirements, site conditions, and building codes. I'm sure architects will object, saying there's no way a machine could infuse the building with the subtle expression and style that could come only from a human. Well, maybe, except that the majority of buildings don't have much style, or have a style that strongly suggests use of a cookie cutter. Throw in some of that innovative design that is indistinguishable from the aftermath of a tornado, and I'm not sure architects we would know if a building had been designed by an architect or by a computer. Furthermore, I suspect that the program could contain several recognized style options, so a given building could resemble Gothic, Romanesque, Chicago, Art Deco, Postmodern, or any of the Revivals.

    What about construction workers? In the past, everything was done in the field, but more and more work is moving into factories. Modular construction further reduces the need for on-site workers, and 3D printing may eliminate more. With the right information, we won't need estimators or schedulers, and driverless trucks are in our future. Sensors on building components and maintenance items will tell computers what needs to be done, and robots will do it.

    The bottom line is - the bottom line. Companies don't exist to hire people; they exist to make money for their owners. At first glance, robots look expensive, but if a robot costs $25,000 and must be replaced after two years, the cost works out to about $6.00 per hour - if it works only eight hours a day. No one knows how all this will play out, but it's sure to be interesting.

    So maybe it's time to update your resume - or have a computer do it for you.


    © 2014, Sheldon Wolfe, RA, FCSI, CCS, CCCA, CSC

    Agree? Disagree? Leave your comments at