I found the following text in the NIST fact sheet on labelling: [begin quote] "Rounding -- In all conversions for the purpose of showing an equivalent SI or inch-pound quantity to a rounded inch-pound or SI quantity, or in calculated values to be declared in the net quantity statement, the number of significant digits retained must be such that accuracy is neither sacrificed nor exaggerated. Conversions, the proper use of significant digits, and rounding must be based on the packer's knowledge of the accuracy of the original measurement that is being converted. In no case shall rounded net contents declarations overstate a quantity; the packer may round converted values down to avoid overstating the net contents. (See the attached Appendixes A & B from the UPLR.) Note: When as a result of rounding SI or customary inch-pound declarations calculated based on the conversion factors in Appendix A, the resulting declarations are not exact, the largest number will be used for enforcement purposes to determine whether a package contains at least the declared amount of product." [end quote]
http://ts.nist.gov/ts/htdocs/200/202/metric.htm -- Terry Simpson Human Factors Consultant [EMAIL PROTECTED] www.connected-systems.com Phone: +44 7850 511794 > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:owner-usma@;colostate.edu] On Behalf > Of Nat Hager III > Sent: 08 November 2002 12:17 > To: U.S. Metric Association > Subject: [USMA:23184] RE: Metric Forum > > Follow-up here: > > This may have inadvertently provided incentive for the over-precise > rounding > in SI declarations. Since the SI declaration carried legal weight, > manufacturers simply looked it up to 8 decimal places in the NIST > handbook, > and quoted verbatum. Simple, basic CYA. > > What would be interesting is if the larger declaration ruled, with > *neither* > stated to more than 2 significant figures... > > Nat > > > > -----Original Message----- > > From: [EMAIL PROTECTED] [mailto:owner-usma@;colostate.edu]On > > Behalf Of Nat Hager III > > Sent: Thursday, 2002 November 07 20.32 > > To: U.S. Metric Association > > Subject: [USMA:23166] RE: Metric Forum > > > > > > > Are you sure you heard this right? Did they use a term like 'more > > > prominent' rather than 'larger'? The UK required the controlling units > > > to be shown 'first'. > > > > > > > No, that's correct as stated. It was in 1992 FPLA, that in the case of a > > rounding disagreement between SI and Imperial, the larger > > declaration ruled. > > It provided an incentive for manufacturers to get the conversion right. > > > > Whether it was enforced or not is another matter. > > > > Nat > > > > > > > >
