Have there been any efforts on a SQL-like language with a more consistent grammar and a closer connection between syntax and semantics? Having this sort of guide is useful to sort through the language we already have, but it certainly seems like there should be better options available in this space.
Thanks for the link! Doing some reading on my own, it does seem like this is an area ripe for improvement. I imagine part of the issue today is that there are now so many tools which abstract over the writing of SQL, thus reducing the impetus to innovate in this area. There are also individuals who make their career on essentially understanding SQL in depth, and understanding the databases which run it, and so have a vested interest in the status quo not improving. (Not to say that these people are bad for this. Simply that their context is such that they are more likely to act in opposition to change, rather than in support of it.)
Have there been any efforts on a SQL-like language with a more consistent grammar and a closer connection between syntax and semantics? Having this sort of guide is useful to sort through the language we already have, but it certainly seems like there should be better options available in this space.
D spec
There is that, but don’t think anyone’s made a native Postgres version of it ;)
Edit: had a hard time making the link correct, especially as I read my Lobsters on the phone.
Thanks for the link! Doing some reading on my own, it does seem like this is an area ripe for improvement. I imagine part of the issue today is that there are now so many tools which abstract over the writing of SQL, thus reducing the impetus to innovate in this area. There are also individuals who make their career on essentially understanding SQL in depth, and understanding the databases which run it, and so have a vested interest in the status quo not improving. (Not to say that these people are bad for this. Simply that their context is such that they are more likely to act in opposition to change, rather than in support of it.)
I’ve heard that changing the parser in Postgres for this wouldn’t be an impossible feat.
I think the hardest part is getting people to care. The easiest route might just be to transpile it, not support natively.