Re: “Standard” software

On 6/27/26 19:03, Lawrence D’Oliveiro wrote:
> On Sat, 27 Jun 2026 18:22:31 GMT, Charlie Gibbs wrote:
> 
>> In fact, I prefer LibreOffice's spreadsheet to Excel ...
> 
> You’re not the only one. Actual researchers who have studied the issue
> have come to the same conclusion
> <https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1008984>:
> don’t use spreadsheets to try to do professional-quality data
> analyses, but if you must use a spreadsheet, LibreOffice Calc is
> preferable to Microsoft Excel.
> 
> In general, I absolutely love the recommendations that they make. The
> first one is a biggie:
> 
>      Scripted analyses are preferred over spreadsheets. Gene name to
>      date conversion is a bug specific to spreadsheets and doesn’t
>      occur in scripted computer languages like Python or R. In
>      addition, analyses conducted with Python and R notebooks (eg:
>      Jupyter or Rmarkdown) capture computational methods and results in
>      a stepwise fashion meaning these workflows can be more readily
>      audited. These notebooks can therefore achieve a higher level of
>      computational reproducibility than spreadsheets. Although this
>      requires a big investment in learning a computer language, this
>      investment pays off in the longer term.
> 
> Note that bit: “capture computational methods and results in a
> stepwise fashion meaning these workflows can be more readily audited”.


   Sounds totalitarian, at least totalitarian bean-counter !

   How about just hiring Good People ?


> Here I thought reproducibility was an absolutely non-negotiable
> foundation stone of scientific research, yet it seems people have been
> publishing results with nothing to back up their analyses other than
> an Excel spreadsheet.

   Like stats, such 'analysis' can be easily tweaked
   to LIE - to support someones political/ideological/
   positional CAUSE. Only look at what serves YOU and
   hurts your opposition.