“Never” is a very strong word. What is your recommendation based upon, exactly?
This is my code. It costs me nothing to tell people my future plans with the code. It costs me nothing to tell people that my plans are most probably off. But it will gain people something - they would be able to plan for the future better.
“Politeness costs nothing and gains everything” - is a better theorem (than yours one). I would rather tell it as: politeness usually costs nothing and gains at least something, but that’s just an engineer correcting a salesman’s statement.
But your plans can change.
Maybe you change your priorities, maybe you discover that a feature is harder to implement than what you tought and you must reduce its scope.
Or maybe you realize it was a bad idea after all and decide to implement a different feature.
If you have promise the original feature, instead of happy users for the improvement, some of them will be angry because of what you "took from them".
That's why I think the real politeness is to never promise nothing concrete until you are actually ready to ship it.
THIS SOFTWARE IS PROVIDED „AS IS“ AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.