I was doing some precalculations on a survey that I am going to be working on in a plat from the early 1960s that has a lot of curves in it. I noticed that a lot of the curve figures didn't work very well and figured that they didn't have COGO at the time.
After some research, I see that the COGO as we know it was developed in the 1960s at MIT as part of their Intergrated Civil Engineering System or ICES. Does anyone know what was the first commercially available COGO package for use by the Civil/Survey community and when did it come out?
I'd bet Bruce Carlson could answer that question and I would not be surprised if he was in the group that created it.
Have a copy of the Surveyor1 for Apple II (Franklin Ace) by Carlson Software dated 1983.
My copy of Surveyor1 manual for IBM PC and compatibles is dated 1985
Stanley Trent and John Morehead are penciled inside the cover.
Being a bit pre Cogo I can remember curve calculations weren't really a problem.
HP helped, but it was just simple maths.
Survey Computations was a subject studied and sat as an exam.
We had tables to set out curves or just work from scratch.
I did such once.
A tight curve in bush in a sensitive area where dead scrub etc was seen as a black mark environmentally.
I blithely applied the curve deflections and proceeded to clear lines and peg chords.
After a while I realised something was wrong as I seemed to be getting tighter and tighter into the curve.
I'd added full deflection as opposed to half for the subsequent bearing to setout the next chord.
Most embarrassing.
It was my first year so we do need to make allowances!
I have an 1871 commercially available text with Latitude and Departure tables, so I must assume that coordinate geometry has been in use much longer than that edition.
Paul in PA
If you consider the Pythagorean theorem the basis and first cogo formulae, then 4000 years is good. Of course the two point form only dates to about 1731 in Alexis Clairauts "Reserches sir les Courbes a Double Courbes" (researches on curves of double curvature). Silvestre Lacroix gets credit for the direction (azimuth inverse) formula. The adoption of latitudes and departures to a surveying problem can be credited to Richard Norwood in his 1637 book a "Seamanship Practice". The concept of totaling up latitudes and departures into a local coordinate system dates back to Thomas Burghs 1724 publication of "A Method to Determine the Areas of Right Lined Figures Universally" which includes his invention of Double Meridian Distances. The conversion of polar-rectangular coordinates is first found in Newtons 1736 "The Method of Fluxons" in the "first method" (not the seventh method often quoted in the web). And the first true use of plane coordinates on a national scale dates to 1744 when Cesar-Francois Cassini used rectangular spherical coordinates on the Carte de Cassini of France.
I neglected to mention the formulae for the transformation between coordinate systems which can be attributed to Francis van Schooten in his second translation and expansion of Descartes "Geometry" between 1659 and 1661.
Lacroix should receive extra merit for providing many of the 'basic' formulae of coordinate geometry (mid point, etc.) in his series of textbooks.
My first exposure to computer cogo was CEADS on a wang computer, 1986. (Civil Engineering And Design System)
JaRo, post: 412654, member: 292 wrote: My first exposure to computer cogo was CEADS on a wang computer, 1986. (Civil Engineering And Design System)
I couldn't remember the name of it, but yes, the Wang had coordinate routines...Late '70s.
In the 1950's and 1960's, computers were starting to become the surveyor's tool to replace what had either been laborious hand computations using logarithms and trig tables, or a somewhat streamlined process using slide rules and hand cranked or electric Frieden and Monroe calculators. Programs such as ICES COGO, from MIT, were being developed. COGO provided the basis for much of the coordinate geometry software in use today. Also at this time, early in-house mainframe computers and time-share services such as Control Data Corporation (CDC), Technical Advisers from Michigan, and MacAuto from McDonnell Douglas in St. Louis, sprang up to meet the hardware demand. Many command driven, batch type software programs were developed to run on these mainframe computer systems.
During the late 1950's, Lloyd Jones (now deceased), an employee of Greenhorne & O'Mara (G&O), an influential engineering and surveying firm in the Washington, D.C. area, decided to develop a coordinate geometry program. He thought that the software he was using at the time contained too many redundant operations. Jones developed a program he called POint GeOmetry or simply POGO. The original program was written in Basic, then ALGOL and finally in FORTRAN. Jones, along with Mack Kelly, LS, a senior surveyor with G&O and the author of the original program manual (now retired and interviewed for this article), developed many of the early routines and procedures. Because the software was being developed in-house, new commands were constantly being incorporated into the program to meet surveyors' special requests. As the command list grew, so did POGO's reputation as a successful software product. By the late 1960's, POGO and the remaining G&O Civil Engineering and Land Surveying (CELS) applications were being marketed by CDC. I have met many people from around the country who have used these programs as a part of their engineering studies or at some time during their professional careers.
In the early 1980's, G&O's computer science department developed and marketed a DOS-based version of POGO, followed in the mid to late 1980's by a DOS-based interactive graphic version known as iPOGO. This interactive graphic program could generate command and output files while working with computer screen graphics. The interactive version also allowed the user to write custom user commands or macros that would process user-defined variables through the native iPOGO commands. This was and still is a very innovative feature. G&O stopped marketing iPOGO for DOS around 1990.
http://archives.profsurv.com/magazine/article.aspx?i=618
When I was in school in the 1980, we had ICES COGO running on a main frame with key punch cards. It was no fun. In our surveying lab, we had an HP 9815 with a HP 9871A Printer Plotter. I learned to make the HP 9815 do all kinds of tricks. I always wanted to give my HO 41 all the power of the HP 9815 but never finished that project. When I graduated and got a job, my first employer had a Wang that no one knew how to use. I remember watching one of the draftsmen loading an intersection program into his HP 11C then removing it and loading a traverse program multiple time doing a sub division plat. I tried to show them how easy it was to do on the HP41 but they weren't willing to spend the $ to get a HP41. I figured out why the next week when they laid the field crews off.
I went to work for another surveyor, then next week and he also had a Wang. I used it some but never liked it. He bought an HP86 B with Wild Volume (C). It was amazing that all of the programs worked just like the HP 9815 but had alpha numeric prompts. I really liked that Box. The first year that I used that computer, we used one box of fan fold paper, one floppy disk, and one dot matrix ribbon. Then we got a Geodimeter Total Station and Data Collector. After than, it was a box of disks a month, a box of paper and a ribbon.
James Fleming, post: 412661, member: 136 wrote: The interactive version also allowed the user to write custom user commands or macros that would process user-defined variables through the native iPOGO commands. This was and still is a very innovative feature.
I don't know when Lewis & Lewis' LLCOGO first appeared -- my initial encounter was around 1985 -- but it also offered macro programming that we used to great advantage.
Going back to the original poster's problem, I've found that Caltrans maps dating back to 1940s and 1950s (possibly earlier, but I don't have as much experience with them) are very accurately calculated. Even with the large radius curves that they often used, their figures generally close within a hundredth or two. I don't know what they used to calculate them, but it worked.
COGO has always been there, but until the widespread use, notice I didn't say availability, of calculators/computers solving SIN/COS accurately for angles very close to 0, 90, 180, etc. Applying/reapplying the trig table value just increased the error.
Why we used to round our answers to the nearest 1/10th instead of a 1/100th. We knew better.
Why the curve data doesn't always work perfectly on old maps/plans. Ya know, I can't get the D, L, and R to agree.
Why we used to field adjust the last few hubs in a curve. The little 'd' wasn't quite right either, or the 20" gun, or the tape...
Why I sleep very well at night when an old pin on a big curve is within +/- a couple feet of where it 'belongs'
Steve
I posed your question to Dustin (Dusty) Smith, the creator of the BenchMark Survey System (AST). Dusty created BenchMark in early 1980s, and I used it in 1983 on an Apple computer.
Below is his remembrance ....
"ItÛªs generally accepted that the Olivetti P-101 was the first cogo calculation device out there for surveyorsÛ? 1967/1968 time periodÛ? it used those magnetic cards we all grew to love and hateÛ? sold for $4000 and, while it was crude, it shrunk an hoursÛª worth of calculation time to about 15 minutesÛ?. Nothing since has ever had that kind of ROI.
Other devices of that time period that I competed with were made by Wylie, Wang Labs, HP, and Monroe to name a few.
The Olivetti 101 was eventually succeeded by the P-602 which had a cartridge tape storage device called the MLUÛ?"
I still use BenchMark, and have never had a problem with it that the well written manual can't solve....
As an after thought regarding curves not fitting....
I was taught to always, always, always hold the tangent directions in and out .....
and always hold the tangent distances....
Let the length of curves "be a variable" as a function of delta and tangent lengths....
Seldom did anyone create a curve with those un-godly tenths and hundredths in them.
Curves were almost always written on a plan in 10 foot increments. More than likely, an even number that
corresponded to the old "Rail Road Curve" templates, we used for drafting.
According to one of my classes, Rene Descartes, a French mathematician, invented analytic geometry around 1619, which is the basis for modern day COGO.
To show you what an old guy I am, I started writing COGO programs in 1965 on an IBM (punch cards) when I worked for FDOT. The drafting supervisor didn't believe it would work consistently and everything I did had to be checked by hand. From there I moved to the Wang 360 (also punch cards), and then later the Wang 2200 (big floppy disks), which was really impressive for the time. I copyrighted COGO in BASIC on January 4, 1982 and started selling it.
Check out Charles L. Millet who was a professor at MIT . Had the opportunity to meet with Prof. Miller on several occasions he was a really smart guy. Prof. Miller created a engineering company named CLM CEAL (Charles L Miller. Civil Engineering Automated Library) great cogo program still have it on one of my computers.
http://news.mit.edu/2000/miller
Drafting Dan, 1956, created by RAH, often considered the first mention of CAD
Jim Frame, post: 412671, member: 10 wrote: I don't know when Lewis & Lewis' LLCOGO first appeared -- my initial encounter was around 1985 -- but it also offered macro programming that we used to great advantage.
Going back to the original poster's problem, I've found that Caltrans maps dating back to 1940s and 1950s (possibly earlier, but I don't have as much experience with them) are very accurately calculated. Even with the large radius curves that they often used, their figures generally close within a hundredth or two. I don't know what they used to calculate them, but it worked.
We have lots of print-outs in the files. For some reason, I think they sent them over to Caltrans for computations. We also have legal pad sheets where Knute Nelson converted the deflection angles from the field notes into bearings and reduced the chained distances to horizontal then whoever ran the mainframe (I assume) would compute and adjust the traverse.
I think 75 might be our State Archives agency number but I can't find that written down anywhere. It seems like Parks was 72. We aren't a Caltrans District. Which reminds me...Caltrans Districts never change, you can go in a District office and they'll have pictures of all the District Engineers back to the beginning. Parks and Forestry (now Cal Fire) have scrambled Districts, Regions, and Operating Units over time.
Some days, I feel like I invented COGO.