Benchmarking (UK style)

[dougandsonia]

Would there be any scope for adopting the Benchmarking idea here? The old OS trig points would be a good basis, especially as there is/was a plan on the part of the OS to remove them. I seem to remember a Society for the Protection of Triangulation Points a few years ago. Admittedly, most of these would be on the top of hills, rather than in town squares.

 

Any thoughts?

 

The nearer your destination, the more you keep slip sliding away.

[+naffita]

The Ramblers have a log of most of the trig points, it was open to everyone but it seems to be members only now, at http//nmc-ramblers.org.uk When this subject was first brought up a few months ago I had a look and they were very well documented, with easy instructions on how to find them. Not very difficult to find and it would not be much of a challenge.

 

naffita

[+The Hornet]

Have you noticed that we now have the 'useful' feature of being able to list all our benchmarks on our 'My Cache' page whilst no longer being able to see all our regular logs.

 

I've mentioned the problem in the the following threads

http://opentopic.Groundspeak.com/0/OpenTopic?a=tpc&s=1750973553&f=6770936793&m=9960961925

http://opentopic.Groundspeak.com/0/OpenTopic?a=tpc&s=1750973553&f=6770936793&m=7480991925

[+The Hornet]

Have you noticed that we now have the 'useful' feature of being able to list all our benchmarks on our 'My Cache' page whilst no longer being able to see all our regular logs.

 

I've mentioned the problem in the the following threads

http://opentopic.Groundspeak.com/0/OpenTopic?a=tpc&s=1750973553&f=6770936793&m=9960961925

http://opentopic.Groundspeak.com/0/OpenTopic?a=tpc&s=1750973553&f=6770936793&m=7480991925

[+paul.blitz]

I've noticed that too, good if you live in the US but not so good for all us Brits!

 

Team Blitz

 

Ok, we've found it, now where did we put the car?

[+paul.blitz]

I agree with naffita, the places wouldn't be very interesting, i know of several trig points that are feet from the road! What about those on private land (See our cache for details,http://www.geocaching.com/seek/cache_details.asp?ID=29455)

[iryshe]

If you guys can find a database of benchmarks in the UK, I'd be happy to add them.

 

Jeremy

 

Jeremy Irish

Groundspeak - The Language of Location

[el10t]

quote:

---

Originally posted by Jeremy (Admin):

If you guys can find a database of benchmarks in the UK, I'd be happy to add them.

---

 

I thought this had already been supplied (was it by Chris & Maria?)

 

Edited - I found the relevant discussion. There is an update in it that says you are already working on it Jeremy

 

-----------------

el10t

mobilis in mobili

 

[This message was edited by el10t on August 01, 2002 at 03:08 PM.]

[el10t]

quote:

---

Originally posted by Jeremy (Admin):

If you guys can find a database of benchmarks in the UK, I'd be happy to add them.

---

 

I thought this had already been supplied (was it by Chris & Maria?)

 

Edited - I found the relevant discussion. There is an update in it that says you are already working on it Jeremy

 

-----------------

el10t

mobilis in mobili

 

[This message was edited by el10t on August 01, 2002 at 03:08 PM.]

[+Chris n Maria]

Team Tate and I did some work on this and establish a file of the Benchmarks in the UK.

 

I think things came to a halt as it got confused between the file I sent and the OS Passive Network (more information on the Ordanance Survey Site). IMHO the passive network is not what we want as the database of trig points has a lot more points to it and seems more interesting as it is mostly points that arn't used any more. The accuracy of the data has already been tested and looks good.

 

I think at the time Jeremy was working on the EBook stuff and it got pushed to the back burner. I'll resend the file again Ref:GEN-020517.28917 as it is possible jeremy lost it in his inbox

 

Chris

"We're not lost - we just don't know where we are"

London & UK Geocaching Resources: http://www.sheps.clara.net

 

[This message was edited by Chris n Maria on August 02, 2002 at 12:57 AM.]

[+Chris n Maria]

Team Tate and I did some work on this and establish a file of the Benchmarks in the UK.

 

I think things came to a halt as it got confused between the file I sent and the OS Passive Network (more information on the Ordanance Survey Site). IMHO the passive network is not what we want as the database of trig points has a lot more points to it and seems more interesting as it is mostly points that arn't used any more. The accuracy of the data has already been tested and looks good.

 

I think at the time Jeremy was working on the EBook stuff and it got pushed to the back burner. I'll resend the file again Ref:GEN-020517.28917 as it is possible jeremy lost it in his inbox

 

Chris

"We're not lost - we just don't know where we are"

London & UK Geocaching Resources: http://www.sheps.clara.net

 

[This message was edited by Chris n Maria on August 02, 2002 at 12:57 AM.]

[+The Northumbrian]

Is the Bench marking open to all or is it just charter members? there was at the start , some mention of these extra's only going to charter members.

[+Chris n Maria]

I'm pretty sure it's open to all. Charter Membership gives you the pocket queries and the ability to change your tag in the forums, plus a nice warm feeling of helping the site

 

Chris

"We're not lost - we just don't know where we are"

London & UK Geocaching Resources: http://www.sheps.clara.net

[+Chris n Maria]

I'm pretty sure it's open to all. Charter Membership gives you the pocket queries and the ability to change your tag in the forums, plus a nice warm feeling of helping the site

 

Chris

"We're not lost - we just don't know where we are"

London & UK Geocaching Resources: http://www.sheps.clara.net

[+jeremyp]

The OS has (or had) two sets of benchmarks: the trig points used for triangulating horizontal position and another set of benchmarks for determining height above sea level.

 

The trig points are usually situated on the tops of hills and so many of them will be in good locations (if you like locations with a good view). Several existing caches have trig points near them. They could be used to make virtual caches with quite easily. Some would be good and some not so good, but you can say that about normal caches anyway.

 

-------

jeremyp

The second ten million caches were the worst too.

http://www.jeremyp.net/geocaching

[+Longitude]

Is benchmarking a way of checking the accuarcy of our receivers? If so can we rely on the OS trig points as being exactly where they say they are bearing mind they were placed before GPS was around?

[+jeremyp]

The trig points were set using the OSGB36 datum. I don't know what the accuracy is, but it is probably higher than the accuracy of a handheld GPS. So as long as you set your GPS to OSGB36, then you can use them.

 

Note that the OSGB36 datum is designed for the UK and so it is fixed wrt the British landmass. WGS84 - the datum used by the GPS satellites - is designed so that the average movement of the whole Earth's crust wrt it is 0. This means that the UK is actually moving slowly North East although the rate of movement is insignificant wrt the accuracy of a handheld GPS unit. Anyway, in the end the trig points will have moved far enough so that the conversion from WGS84 to OSGB36 is wrong, but I assume that firmware upgrades and new GPS units will adjust the conversion.

 

-------

jeremyp

The second ten million caches were the worst too.

http://www.jeremyp.net/geocaching

[Moss Trooper]

The OS have stopped maintaining the majority of the OS monuments as they are now some what defunked due to the GPS system.

 

They have a number of active stations, these are constantly updated. (cant remember how many offa the top of me head)

 

There are about 900 passive GPS stations. These are trig points, height datums, rivets etc. and will be maintained.

 

The accuracy of these passive stations is down to the millimeter in location and height.

 

If you want to find a one near you to check out your GPS Go here.. http://www.gps.gov.uk/

 

You will have to register to access it, (registration is free.) and search the passive station database.

 

There is also a load of info about OSGB and GPS

 

On another note about OSGB36 and WGS84, and this was confirmed by OSGB when I contacted them.. The national grid on OS maps is OSGB36 while the Lat/Long on OSGB maps IS WGS84

 

Moss de Boss... Sorta

[Moss Trooper]

The OS have stopped maintaining the majority of the OS monuments as they are now some what defunked due to the GPS system.

 

They have a number of active stations, these are constantly updated. (cant remember how many offa the top of me head)

 

There are about 900 passive GPS stations. These are trig points, height datums, rivets etc. and will be maintained.

 

The accuracy of these passive stations is down to the millimeter in location and height.

 

If you want to find a one near you to check out your GPS Go here.. http://www.gps.gov.uk/

 

You will have to register to access it, (registration is free.) and search the passive station database.

 

There is also a load of info about OSGB and GPS

 

On another note about OSGB36 and WGS84, and this was confirmed by OSGB when I contacted them.. The national grid on OS maps is OSGB36 while the Lat/Long on OSGB maps IS WGS84

 

Moss de Boss... Sorta

[+jeremyp]

quote:

---

On another note about OSGB36 and WGS84, and this was confirmed by OSGB when I contacted them.. The national grid on OS maps is OSGB36 while the Lat/Long on OSGB maps IS WGS84

 

Moss de Boss... Sorta

---

 

That I find incredible. It's not just a question of moving the equator and meridian. The two systems actually assume that the Earth is a different shape. Moreover I can't find anywhere on a Landranger D series map that even mentions WGS84. Of course you'd never get a lat/long off one that was accurate enough to tell the difference.

 

-------

jeremyp

The second ten million caches were the worst too.

http://www.jeremyp.net/geocaching

[+naffita]

There is a list of 6557 trig points at http://www.36haroldstreet.freeserve.co.uk/trigpoints.htm.

 

I have the list as a sortable excel file if anyone wants it. It's about 850kb so it should be ok to e-mail.

 

I couldn't find any reference to the benchmarks, I seem to remember them as a broad arrow with a bar across the top? They might be more like the US benchmarks and need searching out.

 

naffita

[dougandsonia]

What a knowledgeable bunch! I just thought ... and then all this OSGB 36 / 84 stuff, not to mention rivets. I'm mightily impressed.

 

Thanks guys.

 

Doug

 

The nearer your destination, the more you keep slip sliding away.

[Moss Trooper]

Why should it be incredable!!

 

It actually makes common sence. It doesn't take to much effort to print maps that have 2 different Datum systems.

 

As for not finding WGS84 on their maps.. thats why I contacted them.. to confirm..

 

I do disagree about the difference it makes.. If you use 1/25000 maps it can make a significant difference.

 

Moss de Boss... Sorta

[+Team S-J]

Why not do something like the www.confluence.org project and only have one log per trig point. That way it all comes under one virtual log with a single log for every trig point. As a first come first log process. I only suggest this as I had a first at Ye Ole Survey Monuments - Walton Hill. April 2 by Team S-J I even uploaded a photo. So that's me started! As to how to regulate having only one log per trig point is another matter. Any takers?

[+Chris n Maria]

quote:

---

Originally posted by Chris n Maria:

I'll resend the file again Ref:GEN-020517.28917 as it is possible jeremy lost it in his inbox

---

I resent the file on Friday - havn't heard anything yet.

 

Chris

"We're not lost - we just don't know where we are"

London & UK Geocaching Resources: http://www.sheps.clara.net

[+Chris n Maria]

quote:

---

Originally posted by Chris n Maria:

I'll resend the file again Ref:GEN-020517.28917 as it is possible jeremy lost it in his inbox

---

I resent the file on Friday - havn't heard anything yet.

 

Chris

"We're not lost - we just don't know where we are"

London & UK Geocaching Resources: http://www.sheps.clara.net

[+jeremyp]

quote:

---

Originally posted by Moss Trooper:

Why should it be incredable!!

 

It actually makes common sense. It doesn't take to much effort to print maps that have 2 different Datum systems.

 

As for not finding WGS84 on their maps.. thats why I contacted them.. to confirm..

 

I do disagree about the difference it makes.. If you use 1/25000 maps it can make a significant difference.

 

Moss de Boss... Sorta

---

 

My point is: you can't print a map with two different datums. I agree you could overlay a WGS84 grid on an OSGB36 map, but the lines of the grid wouldn't be straight.

 

The point regarding the accuracy of lat/long readings on an OS map was nothing to do with the resolution of the map. OSGB36 and WGS84 can disagree by up to 200 metres. This can be discernable even on a 1/50,000 map. The point was that the lat/long axes are not divided up into enough divisions and the lat/long grid lines do not go across the map itself. You'd need some very long rulers to get an accurate fix.

 

Lastly: if a map says "uses OSGB36" datum on it, then all position references must use that datum. Most people aren't going to think "hmmm I wonder if the OS are keeping something from us, I'll give them a ring and find out". They will just use the reference and assume it's what the map says.

 

That's why I find the statement that the OS use both datums on their maps (simultaneously) incredible.

 

-------

jeremyp

The second ten million caches were the worst too.

http://www.jeremyp.net/geocaching

 

[This message was edited by jeremyp on August 05, 2002 at 12:49 PM.]

[Moss Trooper]

Errrrrrr.. If it is impossible why are OSGB printing them.. ????

 

You will notice that they don't print a grid for WGS84. could be because they recon 100mtrs difference isn't that much to worry about, or that to many lines clutter an already cluttered map.. and add to conffusion.

 

As for deciphering position from the grid.. a simple romer is all that is required.. and this can be made from a piece of paper, a pencil and something with a straight edge.

 

All positions MUST use that datum.. why.. because you decide that it is incredible that an OS map can not use two datums.!!!

 

I find it incredible that you can be so stuck.. only ONE datum is used on any one map, why? In todays technology, which you seem to be more into than most, you are so narrow minded.. remenber.. the printer prints what he is given.

 

If the OS say that the maps have a duel co-ord system then why doubt it? It is a simple matter. OSGB coords on one rule and WGS84 on the other which is the case.

 

If you check you will find that on a 1/50,000 map Latitude is about 100mtrs different over 40 km between bottom and top and Longitude is about 100 mtrs out over 40 km from left and right. This showing that they do not follow the OSGB grid.

 

If the Lat/Long do not follow OSGB36 then what do they follow???

 

I admit they do not publish that the Lat/Long is WGS84.. their failing.. but it is a fact. Why question it.?

 

Moss de Boss... Sorta

 

PS.. if in doubt phone OSGB

 

[This message was edited by Moss Trooper on August 05, 2002 at 04:50 PM.]

[+lathama]

hav these been added to the list?

[+jeremyp]

quote:

---

Originally posted by Moss Trooper:

Errrrrrr.. If it is impossible why are OSGB printing them.. ????

---

They aren't.

 

I have Landranger sheet 184 which by happy chance has the 2deg west longitude going through it. This lines up perfectly with the west edge of the SU grid square - I checked top and bottom with a straight edge. Actually this is not a coincidence, the OS grid baseline is W2d. I plugged in a grid reference to Map source (SU 00000 50000) and converted to WGS84. The longitude came out as W 2 00.08 which is doesn't seem much but is actually about 98 metres. When I zeroed the minutes and converted back to a grid ref I got SU 00098 50000 which is resolvable on a 1:50000 Landranger map (a difference of 1 in a 6 figure ref).

 

I also have Explorer 174 - a 1:25000 map with the prime meridian marked on it as a green line right across the map. The line passes though the 0 longitude on the scale at the top of the map. It also passes through TL 37840 15000 which is on the top edge of the map. The 4 I estimated because the green line looked to be just to the left of the middle of the interval between 378 and 379. This time I changed the position format from OS grid ref to deg mins, noted the position, then changed the datum to WGS84 and created a new waypont with the same position. It was 120 metres away and on changing back to OSGB and OS groid ref had a ref of TL 37952 14956. The Eastings were different by 112 metres which is easy to resolve on a 1:25000 map.

 

The above leads me to conclude that the W2d line is almost certainly based on the OSGB datum and the 0d line is definitely based on OSGB.

 

In my opinion, the OS person you spoke to misunderstood your question.

 

-------

jeremyp

The second ten million caches were the worst too.

http://www.jeremyp.net/geocaching

[+Chris n Maria]

deleted....

 

[This message was edited by Chris n Maria on August 06, 2002 at 10:44 PM.]

[Moss Trooper]

Well Jeremyp I think the only sensible thing to do on this occasion is to agree to differ.. otherwise it could get very very technical and could slide into a lslanging match and that is one thing I wouldn't like cos then I would have to censor my self ( Not a pretty sight)

 

Agreed!

 

Moss de Boss... Sorta

[Moss Trooper]

Well Jeremyp I think the only sensible thing to do on this occasion is to agree to differ.. otherwise it could get very very technical and could slide into a lslanging match and that is one thing I wouldn't like cos then I would have to censor my self ( Not a pretty sight)

 

Agreed!

 

Moss de Boss... Sorta

[+Chris n Maria]

I have made the files available in various formats at http://www.sheps.clara.net/TrigPoints.html

any feedback would be apreciated.

 

Chris

"We're not lost - we just don't know where we are"

London & UK Geocaching Resources: http://www.sheps.clara.net

[el10t]

Hi Chris

Have you had any feedback yet from Jeremy over whether these are going to be incorporated into the benchmarking part of the site?

 

-----------------

el10t

mobilis in mobili

[+Chris n Maria]

quote:

---

Originally posted by el10t:

Hi Chris

Have you had any feedback yet from Jeremy over whether these are going to be incorporated into the benchmarking part of the site?

 

-----------------

el10t

_mobilis in mobili_

---

Nope,

 

No response at all so far - he is a busy man though!

 

Chris

"We're not lost - we just don't know where we are"

London & UK Geocaching Resources: http://www.sheps.clara.net

[+Chris n Maria]

I've posted a question on the Geocaching.com Discussion board hopefully we will get a reply there.

 

Chris

"We're not lost - we just don't know where we are"

London & UK Geocaching Resources: http://www.sheps.clara.net

[+Chris n Maria]

quote:

---

Originally posted by Chris n Maria:

I've posted a question on the http://opentopic.Groundspeak.com/0/OpenTopic?a=tpc&s=1750973553&f=6770936793&m=8450908925&r=8450908925#8450908925 hopefully we will get a reply there.

---

Still no reply - perhaps it might e best if a few more people added comments to the above discussion. Unless of course EL10T and us are the only people who want to play this game.

 

Chris

"We're not lost - we just don't know where we are"

London & UK Geocaching Resources: http://www.sheps.clara.net

[evilrooster]

Added my question - before even reading your suggestion that we do so.

 

evilrooster

-the email of the species is deadlier than the mail-

[el10t]

We need as many different people as possible to lobby the discussion thread and keep it near the top of the list.

 

-----------------

el10t

mobilis in mobili

[el10t]

In the event that Jeremy either does not acknowledge our request or is unable to implement it I wonder whether we could host a loggable database of these points on the UK geocaching resource page? I am guessing that it doesn't have to be a highly complex system because the whole thing is populated in advance with the waypoints. We just need to be able to accept logs against each one and a tally of how many benchmarks people have found. Oh, and also presumably a search function.

 

What do people think? Could someone on the "techie-team" explain the issues involved in a system like this?

 

-----------------

el10t

mobilis in mobili

[+jeremyp]

The ideal way to do it would be to have a database and some cgi or other server side scripting to populate it when you fill in the form and also to create the HTML for display.

 

I could put a prototype up on my site quite easily as it has mysql support 'built in'. I expect the geocacheuk site has something similar. Could somebody mail me the list of trig points so I can have a go.

 

mailto:geocachingATjeremyp.net (Change the "AT" to an at )

 

-------

jeremyp

The second ten million caches were the worst too.

http://www.jeremyp.net/geocaching

 

[This message was edited by jeremyp on August 12, 2002 at 07:50 AM.]

[+jeremyp]

The ideal way to do it would be to have a database and some cgi or other server side scripting to populate it when you fill in the form and also to create the HTML for display.

 

I could put a prototype up on my site quite easily as it has mysql support 'built in'. I expect the geocacheuk site has something similar. Could somebody mail me the list of trig points so I can have a go.

 

mailto:geocachingATjeremyp.net (Change the "AT" to an at )

 

-------

jeremyp

The second ten million caches were the worst too.

http://www.jeremyp.net/geocaching

 

[This message was edited by jeremyp on August 12, 2002 at 07:50 AM.]

[el10t]

They're on Chris website - see the similar thread in the uk geocaching resource web page forums.

 

-----------------

el10t

mobilis in mobili

[+John Stead]

Has this topic died a death - or what?

[el10t]

quote:

---

Originally posted by jstead:

Has this topic died a death - or what?

---

Not yet - we need to keep it alive to get the UK trigpoints incorporated on the site.

 

-----------------

el10t

mobilis in mobili

[+dancalvert]

Geocaching is a global sport, Lat/Long, based on the WGS84 reference elipsoid is the only globally recognised co-ordinate system. All map projections distort reality, because our planet is a (slightly flattened) sphere. In boiling a point in 3 dimensional space down to a 2 dimensional reference will always produce errors to some degree. Try and get the earth on to a flat bit of paper without squishing some part of it. WGS84 provides a 'best-fit' solution for global navigation. OSGB36 is based on the Airy 1830 projection, one that provides the most accurate bearing and distance referencing for the UK. (we are quite a long way north, so differences between a global projection eg WGS84 and one that gives the most realistic references for us will be more pronounced). If you want to get really anal about it I can give you tweaks to OSGB36 optimised for Wales, Scotland, and even the Isle of Man. They have all kinds of systems in the states (due to the sheer size of the landmass) and I think it is good that J.Irish has decided on WGS84 based Lat/Longs so as to enable a good reference for all of us.

[NeilG]

Hello JeremyP

 

Just for info, My German maps have use two datums WGS84 and Potsdam. An no, the UTM/WGS84 grid is not straight, confused me at first as I am used to using OS maps!

 

The UTM co-ordinates use WGS84 as do the lat/long, but the german Gauss Krüger system uses the Potsdam datum. Fortunately the only grid on the map is the UTM.

Cheers,

Neil

[+dancalvert]

NG, which do you prefer, not from a co-ordinate system perspective, but as a means of getting spatial information across, the German maps, or the OS ones?

 

Also, 'Oh no!!!!' now we have UTM (universal transverse mercator)!!!

[+jeremyp]

quote:

---

Originally posted by Dan:

WGS84 provides a 'best-fit' solution for global navigation. OSGB36 is based on the Airy 1830 projection, one that provides the most accurate bearing and distance referencing for the UK. (we are quite a long way north, so differences between a global projection eg WGS84 and one that gives the most realistic references for us will be more pronounced).

---

 

You're confusing the datum with the projection. WGS84 or OSGB36 do not say anything about how to project a coordinate on the surface of an ellipsoid onto a 2D map.

 

WGS84 and OSGB36 are datums which means they are an ellipsoid and a prime meridian drawn on that ellipsoid. The WGS84 ellipsoid is designed to be a best fit for the whole Earth. The OSGB36 is designed to be a best fit only in the region of the United Kingdom. Given the datum and a point in 3 D space, it is easy to convert that point into a longitude, a latitude and an ellipsoidal height (i.e. distance above or below the ellipsoid).

 

How the point is depicted on a flat map is a function of the projection used on the map. The most well know projection is the Mercator which is obtained by wrapping an imaginary sheet of paper in a cylinder round the equator of the ellipsoid and then drawing a line from the centre of the ellipsoid through the point on the surface and seeing where it hits the paper. The Mercator projection has the property of preserving bearings, that is if you follw a straight line on the map you are travelling on a constant bearing. Unfortunately, it also has the property of making distances look very big near the poles.

 

Ordnance survey maps use a transverse mercator projection based on the 2 degree West meridian (OSGB36) which means that the imaginary piece of paper is wrapped around the 2degree west line of longitude instead of the 0degree line of latitude. In the same way that a normal Mercator projection distorts distances at the North and South edges of a map, distances on OS maps are distorted at the East and West edges (cos it's transverse). In fact, distances are squashed slightly on the 2degree longitude line and stretched slightly at the extreme east and west edges of the OS grid, giving two lines of longitude where an OS map is exactly to scale. At the extreme east and west edges of the grid, the distortion is about 5cm in 1km (IIRC) i.e. a km square on the map would have a side of about 999.95metres if drawn on the ground.

 

-------

jeremyp

The second ten million caches were the worst too.

http://www.jeremyp.net/geocaching