If you do away with the point number what are you gonna paint on the control point? A bunch of circles with "Monument" written by them, is gonna be pretty useless.
Would you have examples of observation ID's?
I'm with Willi - I see little reason the identifiers can't be alphanumeric if the software is appropriately written.
However as a sometimes software developer I'd also insist that the identifiers contain no blanks or certain special characters and be limited to some maximum length.
On the other hand, it is easier to program with numbers. My own least squares programming project was done with point numbers and it is on the "someday" list to add the capability of recognizing alphanumeric identifiers. Internally, they have to be turned into array index numbers by some table or algorithm that I haven't taken the time to write.
Not on hand, I'm not sure that it does - I just figured there is something in the binary. Even if it doesn't, it's not that hard to do so.
I think if you export a JXL and look at it with a web browser you can see it?
I can't paste it here because the XML messes with the HTML. You can see in the JXL file ID="0000001b". Different lines of entry (observations, station setups, target HI changes) get different identifiers and are counted up as the file progresses.
My point number was 104008321, it was registered by Trimble as 0000001b.
I don't use point numbers, I use point designations (alpha-numeric). For example, EOP01, EOP02, etc. GPS points always have designations like 14013AA, 14013AB, etc, and NSRS points get their PID, for example KX2400.
I do believe every point should have a unique designation, at least within a project. Control points (typically GPS and traverse) have unique designations across projects. In the example I mentioned above, the 14013 is a job number, and the AA, AB represent points in the project. I have a lot of projects that we go back to each year or every other year. It is important to be able to have unique identifiers, so for example 05159G (a GPS point) was used again last week, it gets that designation again. In actuality, many control points have dual designations because they are often already named something else. For example, every one of the 39 dam deformation networks that we survey have an M1, M2, etc (these are reference points, either disks or pedestals). So, if we occupy the with GPS, they get a GPS number. The database I create for every project has a station name field and a GPSID field (and a description field), the latter is used in all adjustments and comps because it is unique. Once I export results from the database, I can use the station name if that is what the client needs. Sort of like the station designation (non unique)/PID (unique) system at NGS.
While we are on the subject of point numbering/naming, I would like to know why the practice of giving different names to points when hitting them again is so common. For example, GPS1 on the first occupation and GPS1A on the second, GPS1B on the third, etc. Or running a traverse as 1-2-3-4-5-6, where 6 is really 1. I have ALWAYS called a point by the original designation when hitting it again. I have heard the justification that that way they can check misclosures. I just finished a project where we used the VRS and double occupied all points (separated by several hours). The DC gives the misclosure on the second occupation, and I just store another (in trimblespeak). I don't know or care what it did to the coordinate, I only use raw observations and compute/adjust the raw data.
And I have seen users always number their GPS points 1,2,3,4...on every project. Then they someday connect the two projects and a cluster-!@#$ results.
why would the traverse and GPS static and OPUS solutions for the same point have different point designations? (see my post further down)
> While we are on the subject of point numbering/naming, I would like to know why the practice of giving different names to points when hitting them again is so common.
I think because software used to require it. I know TDS didn't like it when you tried to use a # again, and I think older versions of SurvCE were the same way. The earliest Trimble and Leica data collection I used allowed you to "store another" in Trimble speak.
What's funny is people automatically started using different numbers when the software required it, but now that it's not required it's hard to get people to switch back. At least it has been for me. Ironically data collection has came so far that I can directly emulate the use of a field book, and when we used field books we didn't use different point numbers for redundant obs.
I also think if people knew the percentage of surveyors that just use a TXT file dump instead of analyzing their raw data it would be disheartening.
I like to use & re-use point ID's, much like you described. The challenge becomes, when you are running multiple crews, keeping the different classifications of accuracies automatically sorted. This is one aspect where I think a blending of GIS technique could benefit. Store the observation with a quality attribute and have it intelligently linked to the LS analysis program. The downside of automation, though, is that it usually funnels you in to one brand/workflow (e.g. Trimble).
In TDS you store another observation when it flags the point number conflict.
Should have clarified, I haven't used TDS since it was on my 48. 🙂
Reserve blocks of Pt# to various types of shots.
Ex. Topo 1001-up, Control 100-999, temp 1-99, etc.
Also sub blocks to different crews or areas.
Each setup starts at next even 100.
Ex. 1001-4999 Crew A, 5001-6999 Crew B.
Labeling peg books, notes, maps
Painting ID on ground
All better and less confusing with Pt#
Yes in old days DC's were limited in field length.
How's a crew gonna stake culd-de-sac curb if the office-generated points all have the same description but no point numbers? :bored:
Not true at all!! The point number is the order that F2F reads the codes. Reverse the direction of the shots and you will find voids and missing segments.
Tom, youalmost understand how F2F works but you are missing a major point and that is "How does F2F read the codes?".
Try this; draw a multi-sided building figure as a closed poly. Now let's assign some points to each PI of the building. Start with 100....BC, 102.....BC, 103....BC, then pretend that you took other field shots and now have returned to the building, continue at the next PI with 225.....BC, then 224......BC, 223.....BC, 222.....BC, then you took other shots and again you come back with 654....BC, 677....BC, 678......BC CLO.
How does that line work come out for you??
It's ALL about the point numbers and the reversal of direction on points 222-224. The program reads the crd file in the numeric order of the file.
A rhetorical question?
Some good stuff in your post later on. I might have to borrow some of your ideas. After all, I visit here to learn, not preach.
