Yep. I've written a piece about the limitations of GPS, based on the information provided in this thread...we'll see whether the local running group newsletter sees fit to post it.
Fellow measuring mavens, we really need to come up w/a position statement on these GPS units runners are wearing...and spreading their numbers around as if gospel. I'm seeing and hearing snatches of talk about GPS numbers regularly w/regard to race distances and the sooner we, an officious voice of the national governing body, can address the situation, the quicker we can allay questions & concerns of the (Garmin) Big Kid's Toys wearing public. The position statement should be distributed to all major running outlets including The Running Network, Runner's World, Running Times, Dead Runner's Society (I'm not kidding about DRS).
Posted by request of Gene Newman:
Policy on GPS Measurement: In response to requests to formulate an RRTC policy on use of GPS (Global Positioning System) for course measurement, Gene Newman announced the following: GPS is never acceptable for measuring a race course. GPS may be used for measuring a calibration course, provided that the GPS device used is a professional surveying-quality instrument (these typically cost $30,000 to $60,000), and is actually operated by a licensed surveyor. Coordinates determined by GPS may be useful in documenting positions of points along a race course, although only as a supplement to the distances from landmarks which are specified conventionally for documenting point locations.
To understand the reasoning behind this policy, professional surveying-quality GPS devices can achieve accuracy within a centimeter, whereas consumer GPS devices of the sort that could be mounted on a bicycle (or, worse yet, that a runner might run with) have errors on the order of several meters (sometimes as much as 20 m)—and that’s just in measuring the location of a single point. Errors in measuring a course length are much greater, as the course must be approximated by making point measurements at some number of points along the course, and the overall error depends on how densely the points are sampled, the errors in locating the individual points (inevitably mis-positioned somewhat to left or right of the true path), and the algorithm used in calculating the course length from all the point locations. Also, some points along the course may not have clear views of the satellites used in making GPS measurements, resulting in very large errors in locating those points. For all these reasons, GPS is not acceptable for measuring course lengths.
Policy on GPS Measurement: In response to requests to formulate an RRTC policy on use of GPS (Global Positioning System) for course measurement, Gene Newman announced the following: GPS is never acceptable for measuring a race course. GPS may be used for measuring a calibration course, provided that the GPS device used is a professional surveying-quality instrument (these typically cost $30,000 to $60,000), and is actually operated by a licensed surveyor. Coordinates determined by GPS may be useful in documenting positions of points along a race course, although only as a supplement to the distances from landmarks which are specified conventionally for documenting point locations.
To understand the reasoning behind this policy, professional surveying-quality GPS devices can achieve accuracy within a centimeter, whereas consumer GPS devices of the sort that could be mounted on a bicycle (or, worse yet, that a runner might run with) have errors on the order of several meters (sometimes as much as 20 m)—and that’s just in measuring the location of a single point. Errors in measuring a course length are much greater, as the course must be approximated by making point measurements at some number of points along the course, and the overall error depends on how densely the points are sampled, the errors in locating the individual points (inevitably mis-positioned somewhat to left or right of the true path), and the algorithm used in calculating the course length from all the point locations. Also, some points along the course may not have clear views of the satellites used in making GPS measurements, resulting in very large errors in locating those points. For all these reasons, GPS is not acceptable for measuring course lengths.
Before we get into a long-winded exercise on this, consider that the task seems to be to convince people who do not understand much about measurement that the calibrated bicycle method is more accurate than their fancy toy that promises accuracy. We will not change minds. “A man convinced against his will is of the same opinion still.”
I don’t see surveyors getting scared that they may lose their house-lot measuring business to energetic runners who just run around the periphery and pronounce the lot size. Why should we get into endless wrangling with people whose information comes not from study, but from some misguided idea that because something involves electronics and satellites that it is somehow as accurate as they wish it to be? We have worked hard to develop the system we have and it is the worldwide standard.
People used to use automobile odometers too. I remember reading that someone used one of these, and touted the distance as accurate because it was measured with a brand-new Buick.
To my knowledge, no consumer-grade GPS manufacturer has made the claim of accuracy that some of the runners seem to think is true. Wishing doesn’t make it so.
The answer to the GPS adherents is “consumer-grade wrist-worn runner GPS units do not have the accuracy required to measure race courses.” Leave it at that. There is no simple, easy-to-understand explanation. And a complicated one will not be understood.
I don’t see surveyors getting scared that they may lose their house-lot measuring business to energetic runners who just run around the periphery and pronounce the lot size. Why should we get into endless wrangling with people whose information comes not from study, but from some misguided idea that because something involves electronics and satellites that it is somehow as accurate as they wish it to be? We have worked hard to develop the system we have and it is the worldwide standard.
People used to use automobile odometers too. I remember reading that someone used one of these, and touted the distance as accurate because it was measured with a brand-new Buick.
To my knowledge, no consumer-grade GPS manufacturer has made the claim of accuracy that some of the runners seem to think is true. Wishing doesn’t make it so.
The answer to the GPS adherents is “consumer-grade wrist-worn runner GPS units do not have the accuracy required to measure race courses.” Leave it at that. There is no simple, easy-to-understand explanation. And a complicated one will not be understood.
quote:Originally posted by Pete Riegel:
Before we get into a long-winded exercise on this, consider that the task seems to be to convince people who do not understand much about measurement that the calibrated bicycle method is more accurate than their fancy toy that promises accuracy. We will not change minds. “A man convinced against his will is of the same opinion still.”
I don’t see surveyors getting scared that they may lose their house-lot measuring business to energetic runners who just run around the periphery and pronounce the lot size. Why should we get into endless wrangling with people whose information comes not from study, but from some misguided idea that because something involves electronics and satellites that it is somehow as accurate as they wish it to be? We have worked hard to develop the system we have and it is the worldwide standard.
People used to use automobile odometers too. I remember reading that someone used one of these, and touted the distance as accurate because it was measured with a brand-new Buick.
To my knowledge, no consumer-grade GPS manufacturer has made the claim of accuracy that some of the runners seem to think is true. Wishing doesn’t make it so.
The answer to the GPS adherents is “consumer-grade wrist-worn runner GPS units do not have the accuracy required to measure race courses.” Leave it at that. There is no simple, easy-to-understand explanation. And a complicated one will not be understood.
Seems like we are fast approaching a formal statement suitable for publication. Something that I could send to local/regional running mags or blogs. What do you think?
Tom & all,
I e-mailed Scott with basically the same thing Pete expressed.
The botton line is the GPS is not an acceptable tool for measuring. We will never convince the general public that their toy is not as accurate as they think.
I e-mailed Scott with basically the same thing Pete expressed.
The botton line is the GPS is not an acceptable tool for measuring. We will never convince the general public that their toy is not as accurate as they think.
Gene,
I think you might convince the general public if you include examples like the one I posted in the Electronic Measurement forum. It could save you about 1000 words.
I think you might convince the general public if you include examples like the one I posted in the Electronic Measurement forum. It could save you about 1000 words.
quote:Originally posted by Pete Riegel:
...consider that the task seems to be to convince people who do not understand much about measurement that the calibrated bicycle method is more accurate than their fancy toy that promises accuracy. We will not change minds. “A man convinced against his will is of the same opinion still.”
I don’t see surveyors getting scared that they may lose their house-lot measuring business to energetic runners who just run around the periphery and pronounce the lot size. Why should we get into endless wrangling with people whose information comes not from study, but from some misguided idea that because something involves electronics and satellites that it is somehow as accurate as they wish it to be? We have worked hard to develop the system we have and it is the worldwide standard.
People used to use automobile odometers too. I remember reading that someone used one of these, and touted the distance as accurate because it was measured with a brand-new Buick.
To my knowledge, no consumer-grade GPS manufacturer has made the claim of accuracy that some of the runners seem to think is true. Wishing doesn’t make it so.
The answer to the GPS adherents is “consumer-grade wrist-worn runner GPS units do not have the accuracy required to measure race courses.” Leave it at that. There is no simple, easy-to-understand explanation. And a complicated one will not be understood.
Pete, we don't need to 'convince' people, we should describe & educate them about the limitations of their devices & just a little about how & why we do things as we do. Whether we 'convince' them is a matter of how well the 'position statement' is crafted. We have no control over what readers may 'read' into our statement, what they choose to believe.
I don't assume GPS wearers won't understand something if it's explained succinctly. If our position is made clearly & public enough, we will have served several purposes: spreading the word is always good; education is always good; our group, RRTC, could emerge from USATF shadows and appear proactive and well-meaning. RRTC lurks in the background of the road racing world. Few know what we do or how we do it. We can provide a public service by going to our audience rather than expect them to come to us.
Your examples of surveyors and car odometers to support your argument are specious (although we continue to have too many courses measured by cars). People know the limitations of cars and how precise surveyors are...they don't know much about our methods. The explanation how we do things needn't be complicated. I submit to you the volume of views and posts on this thread (the most ever?) as evidence there are many out there who are interested in our reaction to 'consumer-grade' GPS numbers vs certified courses.
As I stated in a letter to you, I can't understand the sentence in the convention-generated 'position statement' that starts, "Coordinates determined by GPS..." If I can't understand it, I'd guess most of our target audience won't either. Also, we need to remove the qualifying terms & paranthesis that precede '(or worse yet, that a runner might wear)' in paragraph #2 and replace w/'or that a runner might wear'.
We ought to prepare a position statement, let Garmin review it for comment, splice it all together and release it. The road racing season is heating up and we're sure to start hearing more & more from GPS wearers.
It’s true that I don’t see a large need to educate the public on this subject, but that should not stand in the way of an effort being made. In RRTC we have traditionally let the person who has the interest take the lead and do the work with minimal interference. Obviously I am not the person to take on the job.
Both Scott and Mark see a need here, and I don’t see any reason why whichever of them wishes to do so should not just do it. I’d suggest clearing it with Gene, but I would expect that he would not micromanage any effort that’s made.
Both Scott and Mark see a need here, and I don’t see any reason why whichever of them wishes to do so should not just do it. I’d suggest clearing it with Gene, but I would expect that he would not micromanage any effort that’s made.
Have a look at the New York Times article in the Wed. issue on GPS and Glonass and Baidu and Galileo. The U S does not have a monopoly on satellite navigation. Make sure you read the last two paragraphs. Their site is http://www.nytimes.com/2007/04/04/business/worldbusines...4a8&ei=5070&emc=eta1. If that doesn't get it for you go to nytimes.com and type in the author's name - Andrew Kramer.
Some say the only reason the U.S. military made the GPS system available to everyone was to, ironically, maintain their monopoly. Once they made the U.S. system available, it removed the commercial benefit for anyone else to develop their own system, making it difficult to convince the public to fund such a massive project.
Just great, local headlines from a Flint race held 7/28: 'Runners get a little extra at Tuuri'.
'...a couple people said the course could be long. The course was overdistance just enough for the elite runners to take notice. (Winner Susan) Larsen, whose husband (Eric) was charting her progress was a good sport about the extra .09 miles of the posted 6.2 mile course.
Method used to ascertain the extra .09 long: a runner-borne GPS unit. Nice.
I quizzed the race reporter who confirmed my suspicion GPS was used by runners to come up w/the 'extra' distance. However, it was not mentioned in the story what method was used to 'measure' the course raceday. Sloppy, very sloppy reporting AND yet another example of runners whining about race distances recorded on GPS units.
'...a couple people said the course could be long. The course was overdistance just enough for the elite runners to take notice. (Winner Susan) Larsen, whose husband (Eric) was charting her progress was a good sport about the extra .09 miles of the posted 6.2 mile course.
Method used to ascertain the extra .09 long: a runner-borne GPS unit. Nice.
I quizzed the race reporter who confirmed my suspicion GPS was used by runners to come up w/the 'extra' distance. However, it was not mentioned in the story what method was used to 'measure' the course raceday. Sloppy, very sloppy reporting AND yet another example of runners whining about race distances recorded on GPS units.
I think people are right. We need a statement, or maybe two of them.
A succinct statement stating that the RRCA certification method is the most accurate method known, and that all certified courses includes a SCPF factor. All of them. That tests have shown that portable GPS units are not as accurate as the RRCA method. That for a more detailed discussion as to exactly why, please read www.RRCA.org\xxxxx\yyyy
That way the standard text is always accurate because the underlaying data and details can be kept up to date on the website.
THEN, on the website we include a full length comparison of the two measurement methods in language that laymen can understand. This could include diagrams as to why a GPS plotting points around a soothe curve produces and estimation of one side of a polygon, why there is a rounding error in the addition of hundreds of straight segments that make up a GPS distance, and why runners often don't run the shortest course.
A succinct statement stating that the RRCA certification method is the most accurate method known, and that all certified courses includes a SCPF factor. All of them. That tests have shown that portable GPS units are not as accurate as the RRCA method. That for a more detailed discussion as to exactly why, please read www.RRCA.org\xxxxx\yyyy
That way the standard text is always accurate because the underlaying data and details can be kept up to date on the website.
THEN, on the website we include a full length comparison of the two measurement methods in language that laymen can understand. This could include diagrams as to why a GPS plotting points around a soothe curve produces and estimation of one side of a polygon, why there is a rounding error in the addition of hundreds of straight segments that make up a GPS distance, and why runners often don't run the shortest course.
I have had the same experience here in Portland (for the first time) with at least two runners claiming the 2007 Portland Marathon course was long. One had the distance at 26.6 miles the other at 26.65, both using Garmin GPS watches.
As was said earlier, they paid good money for these devices and the fact that the course was certified seems to not mean that much to them. They did manage to accept the course accuracy once I explained the procedures used to measure the course.
Frustrating to have your work challenged when you've measured over 100 courses.
As was said earlier, they paid good money for these devices and the fact that the course was certified seems to not mean that much to them. They did manage to accept the course accuracy once I explained the procedures used to measure the course.
Frustrating to have your work challenged when you've measured over 100 courses.