I just noticed two anomalies re the way Therion records survey length. Data recorded like this... data cartesian from to northing easting altitude 7 7a 90 -99 4.5 7a 9 97 -9 4.5 9 10 31 -22 0 are reported as zero length surveys in the Therion survey-list output. A bug? Depends on point of view I guess, cartesian implies no actual distance measured, so zero survey length, but the definition of two points does have some parallels with a 'shot' and hence survey length. I contrast, 'no survey' leg lengths are recorded as accumulating survey length. ie flags surface fix 1 -41.1724 172.4155 572.6 fix 2 -41.20492 172.2789 574.04 fix 3 -41.21414 172.1513 574.04 etc data nosurvey from to 1 2 2 3 3 4 These are reported by Therion as accumulating survey length. I can see that the desirable behaviour of each of these scenarios might be different depending on one's particular intention. In the case of the particular data where I found this issue, I would prefer the; - underground cartesian data to accumulate length - surface nosurvey data to accumulate length, as it in fact does (I am joining up points from a gps trail) but in general I would expect; -underground nosurvey data to record zero length, because it will generally be a sound or sight connection with no actual survey, hence not surveyed length. I have not checked any of the other data survey formats. Other peoples thoughts on this? Bruce
On Sun, Aug 24, 2014 at 09:18:20AM +1200, Bruce wrote:
A bug? Depends on point of view I guess, cartesian implies no actual distance measured, so zero survey length, but the definition of two points does have some parallels with a 'shot' and hence survey length.
I don't think that is implied - it's just a different way of specifying measurements between two points. When we added it to Survex originally the idea was it provided a way to specify measurements which didn't fit any of the other supported data styles, without forcing people to invent tape/compass/clino data from such measurements.
I have not checked any of the other data survey formats. Other peoples thoughts on this?
FWIW, Survex treats cartesian data as counting in the survey length by the same rules as other measured data styles, and nosurvey data never counts towards the surveyed length (it's assumed to be used for things like visual connections or other unsurveyed links like you say, and really the clue is in the name). I hadn't thought about something like you GPS trail example where an instrument produces a series of absolute positions before. Perhaps a measured version of "nosurvey" would be useful, though I'm not sure we yet have something which can do that underground. Radiolocation could theoretically, but practically it takes too long to set up at each position to be a sane option for surveying a series of points along a passage using it, unless things have evolved a lot since I was last involved. Cheers, Olly
I hadn't thought about something like you GPS trail example where an instrument produces a series of absolute positions before. Perhaps a measured version of "nosurvey" would be useful, ...
Or maybe a new shotflag "Ignored" to compliment the existing "Duplicate", "Surface", "Approx" and "Splay" flags. "Ignored" would mean the shot does not count for length and depth statistics. Maybe the defaults for all shot datatypes except NoSurvey could be "Not ignored" And the default for NoSurvey could be "Ignored". However the user could explicitly change the state of the "Ignored" flag if their usage does not suit the defaults. This change, while it would change the behaviour of some existing datasets, would bring Therion in line with Survex (if I understand you correctly) and also allow the user flexibility to specifically ignore or not some centreline data. A pain for the Therion authors no doubt. Bruce
24. 8. 2014 v 8:33, Olly Betts <olly@survex.com>:
Survex treats cartesian data as counting in the survey length by the same rules as other measured data styles, and nosurvey data never counts towards the surveyed length (it's assumed to be used for things like visual connections or other unsurveyed links like you say, and really the clue is in the name).
This is the way how to recreate missing data set from published map, isn't it. m.
24. 8. 2014 v 8:33, Olly Betts <olly@survex.com>: Survex treats cartesian data as counting in the survey length by the same rules as other measured data styles, and nosurvey data never counts towards the surveyed length (it's assumed to be used for things like visual connections or other unsurveyed links like you say, and really the clue is in the name). This is the way how to recreate missing data set from published map, isn't it. m. That is how I have used it to date. Bruce
On Sun, Aug 24, 2014 at 11:55:29AM +0200, Martin Sluka wrote:
24. 8. 2014 v 8:33, Olly Betts <olly@survex.com>:
Survex treats cartesian data as counting in the survey length by the same rules as other measured data styles, and nosurvey data never counts towards the surveyed length (it's assumed to be used for things like visual connections or other unsurveyed links like you say, and really the clue is in the name).
This is the way how to recreate missing data set from published map, isn't it.
You're replying to a paragraph talking about both the cartesian and nosurvey styles, so I'm not totally sure which you mean by "this". I've certainly done that with the cartesian style, but if you don't want the recreated data to be measured, you can flag it as "duplicate", though one could argue it should count towards the surveyed length - if there's a map it presumably was surveyed, but the data has been lost. That's certainly the case for some older CUCC finds. I don't think this is really a good use of the nosurvey style - you could calculate where the reinvented stations would be and *fix them, then add nosurvey legs for the passages, but that doesn't play well with adding fresh survey data in later - you probably want to trust the newer data more than the reinvented, but with *fix, the reinvented data is treated as the (under)ground truth. Cheers, Olly
I am sorry for reopening this thread, but I would like to fix this issue and I have a question. We have some passages, that were surveyed extremly precisely using laser theodolite and station coordinates were calculated in its own software. We usually import them using fixes and data no-survey and want these shots to be added to total survey length. Do you have an idea, how to treat this situation and to stay compatible with survex? S. On 24 August 2014 12:33, Olly Betts <olly@survex.com> wrote:
On Sun, Aug 24, 2014 at 11:55:29AM +0200, Martin Sluka wrote:
24. 8. 2014 v 8:33, Olly Betts <olly@survex.com>:
Survex treats cartesian data as counting in the survey length by the same rules as other measured data styles, and nosurvey data never counts towards the surveyed length (it's assumed to be used for things like visual connections or other unsurveyed links like you say, and really the clue is in the name).
This is the way how to recreate missing data set from published map, isn't it.
You're replying to a paragraph talking about both the cartesian and nosurvey styles, so I'm not totally sure which you mean by "this".
I've certainly done that with the cartesian style, but if you don't want the recreated data to be measured, you can flag it as "duplicate", though one could argue it should count towards the surveyed length - if there's a map it presumably was surveyed, but the data has been lost. That's certainly the case for some older CUCC finds.
I don't think this is really a good use of the nosurvey style - you could calculate where the reinvented stations would be and *fix them, then add nosurvey legs for the passages, but that doesn't play well with adding fresh survey data in later - you probably want to trust the newer data more than the reinvented, but with *fix, the reinvented data is treated as the (under)ground truth.
Cheers, Olly _______________________________________________ Therion mailing list Therion@speleo.sk http://mailman.speleo.sk/mailman/listinfo/therion
participants (4)
-
Bruce -
Martin Sluka -
Olly Betts -
Stacho Mudrak