| Imagine a table `Foo` with the schema `FooId, name, date, favorite_color, active` Now, you want to pull the ID for the latest `foo` for a specific date, but you don't know any of the column names. The modern workflow looks like this You write `SELECT * FROM Foo` then you say, "Ok, now I can get autocomplete" `SELECT FooId FROM Foo` "Ok, now I can write the where clause" `Select FooId From Foo WHERE date=?` It becomes an exercise in moving the cursor around just to get the autocomplete going. If you are really familiar with the schema, not a problem. But if you just remember a few details about it, then you are stuck in this weird back and forth cursor moving thing. That's why it'd be more ergonomic to have something like `From Foo Select FooId where date=?` Because you never need to move your cursor and you could get all the autocomplete you need at the right times. This becomes especially true when writing joining statements FROM Foo f
JOIN Bar b ON f.FooId = b.FooId
SELECT f.FooId
WHERE b.active = 1
|