I started in construction by picking up nails in my father’s cabinet shop. Been using a radial arm saw since I was 12. By 17 I dropped out of school and worked with Norwegian house framers. These guys were awesome fast. I learned quick and by the time I was 19 I was running my own crew. The money was much better.

That’s when I became concerned about material takeoffs. They were done so shoddily. I quickly found that  if I didn’t do my own material list it was going to be screwed up and it would cost me real money. If the girders were the wrong length, that floor system was not going to be built until more lumber arrived. Usually not until the next day. When you are paying people by the hour you have to keep them busy.

When I first started taking off blueprints I used a scale ruler and a yellow tablet and a calulator. It was a pretty laborious task. Adding up all those measurements. I must say I grew to particularly dislike scale rulers. What was to like? It had 12 different scales on it. You had to fumble with it to find the scale you needed. And 2 scales appear on each ruler coming from opposites ends.

As I started doing more and more material takeoffs I realized the scale ruler  was one of the biggest obstacle to getting it done fast.

The other obstacle was the calculator. While it was light years ahead of adding numbers up by hand, when you used the calculator to add up numbers, most times you would do it twice to see if you made a mistake. If the second time you added it up you got a different answer, then you did it a third time and hoped it agreed with one of the first 2 totals

Before long you also got tired of typing in the same sequence of math over and over again. And then doing it twice to check the answer. Every job started from scratch and every job re-invented the wheel.

In 1978 I got a TRS-80 computer. I was experimenting with wind generators for a few years. The math was very complex and a simple BASIC program replaced the calculator by doing hundreds of calculations per second. It wasn't long until I got the idea “why not do this with estimating”.

My first gadget for estimating with the computer was a variation of a scale ruler. I called it the “Letter Scale”.  Every 1/4 inch had a number or letter. It started at 0 and was labeled  1 2 3 4 5 6 7 8 9 0 A B C etc.

Then with the help of a computer program I could type in a length with one keystroke. I could type in a list of lengths just by entering a string of characters for example “8BB” would mean 1-8 and 2-12’s. returning an answer of 32 as the total.

The next gadget I made to increase the speed of material takeoffs was this digitizing pen. It hooked up to the TRS-80 via the cassette port. It basically strobed the port with a rotary switch and a 9 volt Battery. It worked great. Click on the picture for a more detailed explanation

My next gadget was a digitizing arm. This was very cool, you could put the pointer anyplace on a 18x24 blueprint and read its location to within 1/16 inch

I was rolling along doing takeoffs for builders and lumberyards using this system until one day I walked into an office to pick up a check and the builder said to me “You do terrible takeoffs!” After I picked my jaw up off the floor I blurted out “What?” He said “You cost me thousands of dollars, all the floor beams were wrong and I had to order a lot more lumber”

The problem really happened because of a very stupid idea, “blueprints with options”. This blueprint had an optional 2nd floor plan that required more cantilevered joist. To make a long story short, they put the wrong second floor on the house. Even though my material list gave the location of each set of joist called out by the room they were over.

I thought I had done everything possible to give as detailed and accurate a material list as you could reasonably do.

Yet the misunderstanding of the material list by his framer cost him money even after he already paid me good money for an accurate material takeoff.

THE AH-HA MOMENT

The reason words don’t work well on material lists is because blueprints are spatial and you have to search that space for a reference. thats a lot of information to sort through. a big red circle on a blueprint saying “ HERE” is a lot easier to grasp then to find a piece of text somewhere on the page.

One picture is worth a thousand words.

I knew then that the only path to material estimating nirvana was with graphic annotation that showed actual placement of material.

So the quest became how to get and use the actual blueprint images.

The most obvious way was to use a blueprint scanner. But these were very expensive (over $10,000) so I wound up taking plans to a blueprint place and have scans placed on CDs. It cost about $3 per page

3 dollars a page does not sound like much but the time to get there and back added to the cost.  I needed a way to get that picture quality in my office without driving anywhere. (or spending 10 grand)

Small flatbed scanners had the resolution I needed but could not take a picture of a full size blueprint (24x36). In 2003 I built a scanning camera from the inards of astra 2000p flatbed scanner. It still used the same software that came with the scanner. It didn't know it was converted into a camera and was scanning the blueprint from 30” away. It made a 40 mb image. Even the tiniest writing possible was clear as a bell in the image. It was so clear that you could print it out full scale and read everything on the blueprint.

As well as it worked it had one serious drawback. Each scan took about a minute and a half which does not seem like a great deal of time but it adds up. Especially if you have 20 - 30 pages to shoot. Taking that much time to get the pictures in made taking it to a service bureau look a lot more attractive.

In the mean time digital cameras just kept getting better and better. By the time 5 mega pixels cameras rolled out I could see they were going to be a lot faster to use. At first I pinned the blueprint to the wall to take pictures. It worked but you had to be really careful lining the camera up straight and holding it steady etc.

I tried using a tripod and easel but clearly there had to be a better way. They took up a huge amount of space and there was always the danger of tripping over a tripod leg and sending a $800 camera crashing to the floor.

In 2004 I invented snapscanner and it completely changed how blueprints were brought into the computer. it was by far the fastest and most accurate way yet. The only faster way was to use CAD files and PDFs