Spreadsheets like Excel are clearly low$-$code platforms and often no$-$code. At a base

level, they are usable by almost any educated person; they have a layered$-2$D $($table with

sheets$)$ model, an easily understandable instant$-$calculation semantics and a rich library

of functions with a variety of plugins also available. Modern spreadsheets have migrated

to the cloud, like other low$-$code platforms.

But spreadsheets can grow fantastically in complexity $[9,10].$ Excess complexity can

arise through five different types of scaling: the number of formulas, complex arrange$-$

ment and interconnections of those formulas, massively complex formulas $($many lines

of text with no way to lay it out and comment it$),$ macros $($in Visual Basic for the case

of Excel$)$ and plugins.

Spreadsheets tend to be very fragile $($proneness to bugs if modified incorrectly$),$

and difficult to understand, with subtle differences in formulas not easy to notice, even

if the spreadsheet makes some attempt to warn users. There is difficulty separating

concerns: separate sheets or files can be used for this, but traditional languages are much

more flexible in this regard, allowing almost$-$limitless flexibility in arranging files. It is

almost impossible in spreadsheets to do proper detailed documentation $($e$.$g$.$ of complex

formulae, patterns of cells, and so on$)$; this is only possible in macros which have a

traditional programming$-$language structure. Reuse of formulas in different spreadsheets

is also not generally practical, so they tend to be cloned with the consequent problem of

bug propagation