Ad-hoc SQL workday be nice too. PowerQuery is nice (great, even) but it doesn’t feel like it’s a part of Excel - and while it’s language is better than SQL, I really don’t want to have to learn a whole new syntax to express the same simple ideas.
It’s funny though because so many people who crank on excel don’t know what types are. And don’t know what variables are. And don’t have the layers of mental abstraction that programmers who want types have and need.
So the expected protections don’t really happen. So while it’s nice to not convert “01” to “1” automatically, lots of errors will probably be caused by people trying to add “01” and “02” and getting “0102” and being super confused.
I do think that bringing new syntax to the table is justifiable if it brings new concepts with it.
I agree, power query is essentially a separate product and doesn’t feel first class. I think that probably explains its relative obscurity, despite being highly powerful.
JavaScript is strongly-typed though, it just lacks a way to declare types. One can easily restrict types in script code (using `typeof` for primitives and `instanceof` for object prototypes).
People refer to it as "weakly-typed" because most (but not all) of the built-in types have implicit conversions to other types; and many of which are non-obvious: https://github.com/denysdovhan/wtfjs
Yes, I like this point of view a lot. Like probably the majority of HN, I dread the very idea of a .xls file, but most problems are related to people not understanding the tradeoffs as opposed to shortcomings of Excel itself. "Usability by non-technical people" is a respectable if underappreciated dimension.
So it’s not so much to exclude, but to figure out how to use it and include it.
I think Excel is like the saturated fat of the data world. Widely used, found bad in many situations, tried to be removed completely, replacements caused more harm (sugar), now figuring out the right kind and what combinations are good.
This is the key insight and one that the other replies somewhat hilariously ignore.
An Excel replacement is going to be "ugly" to programmers' eyes. It has to be since this is what makes it useful to the wider population.
What it needs to be is a bit better.
I’ve been waiting years for someone to solve this as there are serious flaws that I’d like fixed.
But all the “improvements” end up breaking other things. For example, airtable seems cool but its cost makes it non-comparable (eg, email a spreadsheet to a million people, a million can “fork” try pricing that with airtable; save a spreadsheet to disk, archive forever, etc).
I think that everyone can program, but few will be professional programmers. And many of the fixes for Excel are assuming that all people who program should act like professional programmers. And this isn’t possible or likely.