Re: """Standard""" software (was: Re: Early-2014 Macbook Air and Linux Mint Cinnamon)

Le 27-06-2026, Your Name <[email protected]> a écrit :
> On 2026-06-27 18:22:31 +0000, Charlie Gibbs said:
>
>> In fact, I prefer LibreOffice's spreadsheet to Excel, primarily because 
>> when I pull in a CSV file it automatically sets the column widths to 
>> something reasonable, rather than requiring you to manually select the 
>> hilariously-misnamed "Auto format".
>
> Column widths are easily fixed quickly in Excel. Just click in the 
> corner "cell" above the row numbers and left of the column letters to 
> select the entire spreadsheet, then double-click the divider between 
> any column.
>
> Similarly for row heights, although you may need to turn on the 
> text-wrapping for all the cells first.
>
> The biggest issue with Excel is that it is simply not accurate, and 
> therefore untrustworthy. Because Microsoft stubbornly chose to do 
> decimal arithmatic in stupid way, that they claim is a "feature", it 
> means miniscule errors creep in, which can then get bigger when using 
> those error cells in other calculations. No other spreadsheet (or 
> database) that I've used has had this problem because they do decimal 
> arithmatic properly.

For me, the biggest issue with Excel opening CSV is because it assumes
things and changes the document based of those fucking assumptions. And
once it's change, you can't do anything about it and you have to manage
your CSV file before opening it with Excel.

Concrete examples:
- UTF-8 versus latin9 => as a French there are a lot of diacritical
  signs in the texts and I can't tell it which encoding it should use.
  So, I have to manage that before opening it.
- Coma versus semicolons => In France the decimal separator is the coma,
  so we can't use the coma to be the field separator. So we are using
  the semicolon. And when the field separator is not the one expected by
  Excel, everything is messed up and there is no way to change that. I
  know there is a way to convert a cell, but if the cells on the right
  are not empty, everything is lost in those cells. So it needs to be
  treated before opening it with Excel.
- French dates versus English dates => when Excel sees something it
  knows, it consider it must be adapted. So when there is JAN, for
  janvier in French, which is the same as January in English, everything
  is fine.  But, when there is FEV, for février in French, which is not
  the same as FEB for February, then Excel doesn't know what to do and
  let it like that. In the end, half of the dates have been changed and
  the other half remain identical. So it's impossible to manage and it
  needs to be treated before opening it in Excel.

And, for the record, in French faux means at the same time scythe and
false. And, guess what, Excel can be confused because when some things
are searched in Excel, having faux on a cell can make the ligne just
disappear. And some French cities have changed their name for that
reason:
<https://www.leparisien.fr/societe/faux-roche-grigny-pourquoi-huit-communes-vont-changer-de-nom-le-1er-janvier-prochain-10-08-2024-5T4PROQXUVFONGEN36FC2735JQ.php>
Sorry, it's in French, but it explains why some French cities have
change their names with that part:
« On s’est rendu compte que dans certains fichiers numériques, notamment
les tableurs, Faux disparaissait ou était traduit en anglais et devenait
false »

Which can be translated in something like:
"We have realised that in some of the numeric files, like spreadsheets,
Faux disappeared or was translated in English and became false."

Some other articles spoke about Excel, I don't know about OpenOffice or
OnlyOffice. But what I know is that OpenOffice and OnlyOffice let me
chose the field separator and the encoding before doing anything. So if
everything is messed up, it's my fault and start again doing the right
choices.

-- 
Si vous avez du temps à perdre :
https://scarpet42.gitlab.io