NavList:

RE : Index-error-vs-index-correction-Hirose-mar-2020-g47237

Dear Paul,

Yes, at the beginning of every single year during both College (4 years) and 5 year Engineer's degree cursus, I was reminded this good sense rule : from a recorded value we get rid of / remove / suppress its observational error to get the error free observed value.

Hence:

Recorded value MINUS error = correct value (freed from error) . Or , as an equivalent :

Recorded value PLUS correction = Correct value

Hundreds if not thousands of "Correction tables" have been published in all areas. Such Corrections can be either positive or negative, and from recorded values you always ADD the published applicable / relevant corrections with its own sign.

I have very seldom heard of "Error Tables". But I remember I once did and - lo and behold - there was a note specifying that such "errors" were to be subtracted from recorded values to derive true values.

This debate is interesting since this is probably the first time in years that this topic is being addressed here.

At least this is the way I have been trained.

I also think that Jean Meeus would probably share this same view point.

It would also be interesting to see what the Handbook of Mathematical Functions by Abramowitz and Stegun says on the matter. My own copy still in a carboarb box further to my recent move.

The fact that NAL seems to use an "error" the opposite way could possibly come from a typing error (!) or from an oversight from the writer..

Best Regards to you Paul, and to all of you,

Antoine