Do we have agreement on a proposed text change for this sub-issue?
From: Richard L. Barnes [mailto:rbarnes@bbn.com]
Sent: Wednesday, February 10, 2010 3:37 PM
To: Thomson, Martin
Cc: Bernard Aboba; geopriv@ietf.org
Subject: Re: [Geopriv] Proposed text for 3825bis
Are you referring to the quantization error?
highvalue = value | mask
versus:
highvalue = ( value | mask ) + 1 (== lowvalue + scale)
(the all-0xFF high value vs. the next low value)
If that's the concern, then we might as well just adjust the fixed-point version to use the "+1" version.
I guess I'll agree that you're unlikely to get errors simply due to the fact that you're doing floating-point arithmetic in this case.
--Richard
On Feb 10, 2010, at 3:21 PM, Thomson, Martin wrote:
The error is absolute (it’s 2^-8 for altitude), it doesn’t make sense to say “up to”.
From: Richard L. Barnes [mailto:rbarnes@bbn.com]
Sent: Thursday, 11 February 2010 10:16 AM
To: Thomson, Martin
Cc: Bernard Aboba; geopriv@ietf.org
Subject: Re: [Geopriv] Proposed text for 3825bis
Good point on the quantification: Suggested fix:
s/up to floating-point errors/up to a possible error of 2^-25/
On Feb 10, 2010, at 3:10 PM, Thomson, Martin wrote:
Hi Richard,
I notice that there is a small error in the first paragraph.
Section X: s/versions 1 and 2/versions 0 and 1/
I’ll also note that the difference between setting bits and using exponents of two is quantifiable. The error in the upper value is 2^-25.
