Linas Žvirblis <[EMAIL PROTECTED]> wrote: > Neither have I. But gnumed, for example, currently contains entries in > "Tools". This is wrong by default, so I do not consider that a feature. > > They will have to be moved anyway, so why not to "Science"? This would > keep them organized and, most importantly, in predictable locations.
Agreed. >> You didn't put everything there that contains "analysis" in its name - >> probably you took only "general purpose" things. Hm. > > Correct. Anything that can process and/or present generic data. Is there > anything wrong with having such section? No, at least not if it can be clearly defined. >> Currently, grace(6), which you forgot, is in math. > > "Math" was outside my scope. > >> I think that there >> is no clear border between spreadsheet calculators (Openoffice Calc, >> gnumeric, ...), which can do a little data fitting and plotting, and >> full-featured data fitting and plotting programs (fityk, mn-fit,grace), >> which often can do less data manipulation, but still some. > > I think there is. They are targeted at entirely different user groups. Hm, in biological sciences, people often use Excel even for presentation graphs (I guess mainly because it's there). For complicated graphs, they switch to a "real" plotting program, also if you need to do nonlinear least-squares fitting. Currently, the free plotting/fitting programs I know of are in fact clearly different from spreadsheets, because they lack a decent user interface for data manipulation. But that may change (it was promised for a future version of grace, and I have not yet checked out grace6). I don't say that it's impossible to make a clear distinction, but one needs to be careful when phrasing the descriptions. Note also that one can use algebra systems with numeric capabilities (maxima,...) for data analysis, too. >> qtdmm looks like it should be in electronics. kboincspy is probably >> less scientific than a performance monitor - after all it's the central >> server who does the serious evaluation of the data that the distributed >> clients calculated. > > It is really up to package maintainer to decide. I am not familiar with > any of those, so that was only a rough guess. Yes, but it makes the list of packages suitable for "Data analysis" smaller; I'm not sure we need it. Regards, Frank -- Frank Küster Single Molecule Spectroscopy, Protein Folding @ Inst. f. Biochemie, Univ. Zürich Debian Developer (teTeX)
