Dear Charlie I don't remember this discussion. Perhaps I commented it at the time! Anyway, I think the three you proposed before are well-defined and suitable for proposal in CF if you know of actual use-cases for them - do you? To propose a modification to the convention, please open a trac ticket.
I'm not sure about > sum_absolute_value u The data values are representative of > a sum or accumulation of the absolute values over the cell. I can see what you mean, but I can't think of a practical example of it. What is the use-case? Best wishes Jonathan ----- Forwarded message from Charlie Zender <[email protected]> ----- > Date: Tue, 16 Feb 2016 16:05:57 -0800 > From: Charlie Zender <[email protected]> > To: David Hassell <[email protected]> > CC: CF Metadata Mail List <[email protected]> > Subject: Re: [CF-metadata] New cell_methods: mabs/mibs/mebs? > User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.11; rv:38.0) > Gecko/20100101 Thunderbird/38.5.1 > > Dear CFers, > > About a year ago I proposed three new cell-methods (mibs/mabs/mebs). > The proposal received some discussion, and what I infer from silence > as acquiescence. To those three methods I now propose adding a fourth: > tabs = total absolute value whose CF mehod would be encoded as > sum_absolute_value (or, if people prefer, total_absolute_value). > tabs is analogous to the existing "sum" statistic but is computed > with absolute values. Its utility is in computing statistics of > quantities whose magnitude not signedness is the key metric. Its > omission in my original proposal was an oversight. > > The full set of recommendations could be implemented in CF by > inserting the following line into Table E.1. Cell Methods > > http://cfconventions.org/Data/cf-conventions/cf-conventions-1.7/build/cf-conventions.html#appendix-cell-methods > > cell_methods: Units: Description > maximum_absolute_value u Maximum absolute value > minimum_absolute_value u Minimum absolute value > mean_absolute_value u Mean absolute value > sum_absolute_value u The data values are representative of > a sum or accumulation of the absolute values over the cell. > > Thoughts? > > Best, > Charlie > > On 2/20/15 00:55, David Hassell wrote: > >Hi Charlie, > > > >>the strings "maximum_absolute_value", "minimum_absolute_value", > >>and "mean_absolute_value". I suggest CF adopt this, or some > >>variant pursuant to discussion. > > > >Many apologies - for some reason, I failed to register this crucial > >sentence when I replied. I suppose that it's good that the same names > >were arrived at without prior knowledge. > > > >No objections from me, then. > > > >All the best, > > > >David > > > >---- Original message from Charlie Zender (04PM 19 Feb 15) > > > >>Date: Thu, 19 Feb 2015 16:39:15 -0800 > >>From: Charlie Zender <[email protected]> > >>To: David Hassell <[email protected]> > >>CC: CF Metadata Mail List <[email protected]> > >>Subject: Re: [CF-metadata] New cell_methods: mabs/mibs/mebs? > >>User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 > >> Thunderbird/31.4.0 > >> > >>Hello David, > >> > >>I use mabs/mebs/mibs as shorthand, not as cell_methods. > >>I suggest, and NCO implements, cell_methods with the longer > >>versions that you prefer. The command line operators of NCO > >>accept either full or abbreviated versions (to save typing > >>when conducting the operation itself). > >> > >>Now I see what you mean by the sentence in the appendix. > >>It could be read either way, and I read it the wrong way. > >>So I like your suggestion to clarify it. > >> > >>Charlie > >> > >>On 02/19/2015 04:27 PM, David Hassell wrote: > >>>Dear Charlie, > >>> > >>>I for one have no objection, in general, to new cell methods - I don't > >>>think that there are enough. > >>> > >>>Your suggestions (mabs/mibs/mebs) are clearly well defined, though I'm > >>>personally not so keen on the use of abbreviations. I've not seen > >>>these terms before, and wouldn't have guessed what they all mean. This > >>>is contrary to all of the other cell methods, which are unabbreviated > >>>and, I suspect, nearly universally understood. > >>> > >>>I dislike typing as much as anyone, but spelling them out is only 1 to > >>>4 characters more than typing standard_deviation, the current longest > >>>method name: > >>> > >>>standard_deviation > >>>mean_absolute_value > >>>minimum_absolute_value > >>>maximum_absolute_value > >>> > >>>These terms seem nicely self describing to me. Do you think this is an > >>>option? > >>> > >>>>There appears to be an error in the draft 1.7 document. The sentence > >>>>describing Appendix E (the cell-methods appendix) says "In the Units > >>>>column, u indicates the units of the physical quantity before the > >>>>method is applied." Actually the units column entries are valid > >>>>_after_ the method is applied. Variance is the only method for which > >>>>this currently matters. This can be addressed independently of the > >>>>rest of the cell_methods suggestions proposed here. > >>> > >>>I think that this is OK. The column contains units after the method is > >>>applied, defined in terms of the original units ('u'). However, I > >>>agree that the terse description can mislead (as it did me just > >>>now!). How about replacing: > >>> > >>> "In the Units column, 'u' indicates the units of the physical > >>> quantity before the method is applied." > >>> > >>>with something like: > >>> > >>> "The Units column contains the units of the physical quantity after > >>> the method is applied, in terms of 'u', the units before the method > >>> is applied." > >>> > >>> > >>>All the best, > >>> > >>>David > >>> > >>> > >>>---- Original message from Charlie Zender (11AM 19 Feb 15) > >>> > >>>>Date: Thu, 19 Feb 2015 11:56:22 -0800 > >>>>From: Charlie Zender <[email protected]> > >>>>To: CF Metadata Mail List <[email protected]> > >>>>Subject: [CF-metadata] New cell_methods: mabs/mibs/mebs? > >>>>User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:31.0) > >>>> Gecko/20100101 Thunderbird/31.4.0 > >>>> > >>>>Dear CF-ers, > >>>> > >>>>The statistics mabs/mibs/mebs stand for "Maximum absolute value", > >>>>"Minimum absolute value", and "Mean absolute value", respectively. > >>>>They are similar to max/min/mean statistics, and they can be useful > >>>>in characterizing data when one wants positive-definite metrics. > >>>>mebs (unlike mean) does not allow positive and negative values to > >>>>compensate eachother. Unlike rms, mebs not does weight outliers > >>>>quadratically. NCO (version 4.4.8) implements mabs/mibs/mebs as > >>>>fundamental statistics (like max/min/mean/rms), and annotates the > >>>>cell_methods attribute of variables reduced by these statistics with > >>>>the strings "maximum_absolute_value", "minimum_absolute_value", and > >>>>"mean_absolute_value". I suggest CF adopt this, or some variant > >>>>pursuant to discussion. > >>>> > >>>>So I guess this is a request for discussion. > >>>>The relevant portions of CF are > >>>>http://cfconventions.org/Data/cf-conventions/cf-conventions-1.7/build/cf-conventions.html#cell-methods > >>>>http://cfconventions.org/Data/cf-conventions/cf-conventions-1.7/build/cf-conventions.html#appendix-cell-methods > >>>>The modifications that would be needed seem straightforward: > >>>>mention mabs/mebs/mibs in the text and then enlarge the existing > >>>>cell_methods table table by three rows. > >>>> > >>>>There appears to be an error in the draft 1.7 document. The sentence > >>>>describing Appendix E (the cell-methods appendix) says "In the Units > >>>>column, u indicates the units of the physical quantity before the > >>>>method is applied." Actually the units column entries are valid > >>>>_after_ the method is applied. Variance is the only method for which > >>>>this currently matters. This can be addressed independently > >>>>of the rest of the cell_methods suggestions proposed here. > >>> > >>>I think that this is OK. The column contains units after the method is > >>>applied, defined in terms of the original units ('u'). However, the > >>>terse description is misleading on first reading. How about something > >>>like: > >>> > >>>"In the Units column are the units of the physical quantity after the > >>> method is applied, in terms of 'u', the units before the method is > >>> applied." > >>> > >>>>Best, > >>>>Charlie > >>>>-- > >>>>Charlie Zender, Earth System Sci. & Computer Sci. > >>>>University of California, Irvine 949-891-2429 )'( > >>>>_______________________________________________ > >>>>CF-metadata mailing list > >>>>[email protected] > >>>>http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata > >>> > >>> > >>>-- > >>>David Hassell > >>>National Centre for Atmospheric Science (NCAS) > >>>Department of Meteorology, University of Reading, > >>>Earley Gate, PO Box 243, > >>>Reading RG6 6BB, U.K. > >>> > >>>Tel : +44 118 3785613 > >>>E-mail: [email protected] > >>> > >> > >>-- > >>Charlie Zender, Earth System Sci. & Computer Sci. > >>University of California, Irvine 949-891-2429 )'( > > > > > >-- > >David Hassell > >National Centre for Atmospheric Science (NCAS) > >Department of Meteorology, University of Reading, > >Earley Gate, PO Box 243, > >Reading RG6 6BB, U.K. > > > >Tel : +44 118 3785613 > >E-mail: [email protected] > > > > -- > Charlie Zender, Earth System Sci. & Computer Sci. > University of California, Irvine 949-891-2429 )'( > _______________________________________________ > CF-metadata mailing list > [email protected] > http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata ----- End forwarded message ----- _______________________________________________ CF-metadata mailing list [email protected] http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
