Re: (fwd)

New Message Reply About this list Date view Thread view Subject view Author view

From: John Halleck (
Date: Thu Feb 07 2002 - 05:51:15 CET

Return-Path: <>
Received: from localhost (localhost []) by (Postfix on SuSE eMail Server 2.0) with ESMTP id 88ECB9F12 for <>; Thu,  7 Feb 2002 06:12:01 +0100 (CET)
Received: by (Postfix on SuSE eMail Server 2.0, from userid 28) id 98DD09F08; Thu,  7 Feb 2002 06:11:58 +0100 (CET)
Received: from localhost (localhost []) by (Postfix on SuSE eMail Server 2.0) with ESMTP id A13E39F12 for <>; Thu,  7 Feb 2002 06:11:57 +0100 (CET)
Received: from ( []) by (Postfix on SuSE eMail Server 2.0) with ESMTP id E799C9E76 for <>; Thu,  7 Feb 2002 06:11:53 +0100 (CET)
Received: from localhost (nahaj_at_localhost) by (8.9.2/8.9.2) with ESMTP id VAA18101; Wed, 6 Feb 2002 21:51:16 -0700 (MST)
Date: Wed, 6 Feb 2002 21:51:15 -0700 (MST)
From: John Halleck <>
To: R Knapp <>
Cc: John Halleck <>, CaveXML <>
Subject: Re: (fwd)
In-Reply-To: <>
Message-ID: <>
Content-Type: TEXT/PLAIN; charset=US-ASCII
Precedence: bulk
X-Virus-Scanned: by AMaViS perl-11

On Wed, 6 Feb 2002, R Knapp wrote:

> Date: Wed, 06 Feb 2002 21:10:59 -0500 (EST)
> From: R Knapp <>
> To: John Halleck <>
> Cc: CaveXML <>
> Subject: Re: (fwd)
> > > - Should length be metric or English? Either choice w/could incur
> > > conversion back to the other "base" for misc calculations (length, depth, etc).
> >
> > Which English? US feet or International (ISO) feet?
> > At least their is only one metric system.
> True.
> > (And before you think "who cares about the difference between US and ISO feet?"
> > note that Larry Fish and Compass have had problems on just this issue for the people
> > that need output referenced to their state plane coordinate system.)
> Would the use of meters provide a solution?

  A solution? No... you still have to know which state plane coordinate systems
  are US feet and which are International feet. (Lists are availiable for those that

  But, it does mean that those of us that generally really don't care don't
  have to worry about it. (I.E. my didly tools don't need to do anything
  whatsoever with the issue.)

> > > - Should angles be degrees or radians? Radians have the advantage of being
> > > friendly to computers (most functions like rads and not degrees) but again
> > > there might be a conversion back for misc calculations (rose diagrams, etc).
> > I'd vote degrees. My weak argument is that at least degrees can be represented
> > exactly, any conversion to radians has to truncate.
> Logical.
> > > Maybe there could be an attribute/element that specifies this information at
> > > the top of the XML file?
> > As soon as it is specified at the top, then every diddly tool down the pipe has to
> > be aware of it.
> I think even if the canonical units will always be the same, it should be specified
> somewhere, just so all tools interacting with the format can extract the correct
> information without any ambiguity.

  (And XML does support constant default values.)

> > Yep. Which is the reality. (But remember that if length A has standard deviation
> > of Sa and Length be has standard deviation of Sb then the two added together have
> > a standard deviation that is NOT Sa+Sb, it is sqrt (Sa*Sa+Sb*Sb).
> It's been too long since I've done any real statistical anaylsis. Maybe at some point
> I'll have some free time (maniacal laughter)

  :) By the way, the variances of the shots add... which is why you see a lot
  of advanced network adjustment stuff derived in terms of variances.

> > I've generally not been posting things to the list, since I've come to feel that my
> > pedantic picking at assumptions was inhibiting some folks from posting.
> I think it has been very helpful. There is a lot of detail I would have glossed over (or
> ignored until it bit me) if you didn't keep me honest. Thanks.

  I'm quite willing to try to keep people honest... I hope that they return the favor.

> Richard Knapp

New Message Reply About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b30 : Thu Feb 28 2002 - 23:00:00 CET