[MetaCRS] Quick and Dirty CSV File

Landon Blake lblake at ksninc.com
Thu Nov 5 13:13:06 EST 2009


This is harder than I thought. :]

Martin wrote: "- I'm not crazy about the "Northing:nnnn" format. Seems
hard to parse, and what exactly is being communicated by this prefix? I
think Norm is on the money with his suggestion that the CRS determines
the 
interpretation of the ordinate values."

The ordinate prefix eliminates the need to remember the order of
ordinate values, and would also give the parser a clue to the format for
the value that follows. I don't need to worry about parsing a value in
DMS format for an ellipsoid height that is always in meters.

I don't know about Python or Javascript, but you can split the ordinate
prefix from the ordinate value pretty simply with the built in
String.split function.

Still, if the consensus is that this it too hard to parse, we can drop
the prefix.

Martin wrote: "Slim down the column names. Eg
Expected Destination CRS Ordinate 1 -> Dest Ord 1 Suggest 3D Distance
Tolerance -> Distance Tol"

I don't see why this is necessary. Abbreviations make things harder to
understand, and we are just dealing with plain text files. If there is
no file format limitation on column name length, why not spell it out?

Martin wrote: "Is there agreement about the concept of a 3D distance
tolerance? It seems to me that it's better to break out the tolerances
for each axis, if that's desired."

I have no preference. This was just my suggestion. I can change it to a
tolerance for each axis.

Martin wrote: "Instead of saying "None" or "Not Provided" just leave the
field empty (null)"

Now this makes parsing harder! If we leave "none" or "not provided" each
row will always have the same number of values. If we don't include a
value each row could have a different number of values or cells. This
could make parsing a lot more difficult. For example, if I use the
String.split function in Java I can get an array of Strings where the
comma determines what goes in each element of the array. I can then use
the same index position to always access the desired array element
without having to look at the other elements in the array. I hope this
makes sense.

Martin wrote: "Would it be better to normalize the authority prefix out
into a separate column? This will make the table more amenable to
querying in database environments."

This is a good suggestion. I can make this change.

Landon
Office Phone Number: (209) 946-0268
Cell Phone Number: (209) 992-0658
 
 
-----Original Message-----
From: Martin Davis [mailto:mbdavis at refractions.net] 
Sent: Thursday, November 05, 2009 10:11 AM
To: Landon Blake
Subject: Re: [MetaCRS] Quick and Dirty CSV File

I'm going to post my comments here:

- I'm not crazy about the "Northing:nnnn" format. Seems hard to parse, 
and what exactly is being communicated by this prefix? I think Norm is 
on the money with his suggestion that the CRS determines the 
interpretation of the ordinate values.

- Slim down the column names. Eg
Expected Destination CRS Ordinate 1 -> Dest Ord 1
Suggest 3D Distance Tolerance -> Distance Tol

- Is there agreement about the concept of a 3D distance tolerance? It 
seems to me that it's better to break out the tolerances for each axis, 
if that's desired.

- Instead of saying "None" or "Not Provided" just leave the field empty 
(null)

- Would it be better to normalize the authority prefix out into a 
separate column? This will make the table more amenable to querying in 
database environments.

Martin


Landon Blake wrote:
>
> I busted out a quick and dirty CSV file for discussion. I attached it 
> as a text file with comments. I also attached it as a Microsoft Excel 
> Spreadhseet without comments and with some formatting. I can also make

> it available in Open Office Format.
>
> A couple of things came up when I was working on the file:
>
> - I provided a suggested 3D tolerance in the units of the expected 
> coordinates. Does anyone see a problem with using these units instead 
> of the source coordinates?
>
> - I added a comments section. I figured we could add comments there, 
> without commas, of course. This will make a good "metadata" field for 
> each row if needed.
>
> - I didn't have to work with degrees-minutes-seconds ordinate values, 
> but some rows might need to. I would suggest we use a format like 
> 37-57-23.66554 and -121-45-18.55265 for ordinate values that need to 
> be in this format.
>
> - I also propose we use the negative sign to indicate west longitude 
> values and south latitude values.
>
> I will await discussion before making revisions to this file.
>
> Landon
>
>
>
> *Warning:
> *Information provided via electronic media is not guaranteed against 
> defects including translation and transmission errors. If the reader 
> is not the intended recipient, you are hereby notified that any 
> dissemination, distribution or copying of this communication is 
> strictly prohibited. If you have received this information in error, 
> please notify the sender immediately.
>
>
------------------------------------------------------------------------
>
> _______________________________________________
> MetaCRS mailing list
> MetaCRS at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/metacrs
>   

-- 
Martin Davis
Senior Technical Architect
Refractions Research, Inc.
(250) 383-3022



Warning:
Information provided via electronic media is not guaranteed against defects including translation and transmission errors. If the reader is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this information in error, please notify the sender immediately.


More information about the MetaCRS mailing list