3 ms·
brian_goetz: >Perhaps I can shed some light on this apparent contradiction. >We discovered, unfortunately quite late in the game, that the design was flawed.
by kernal 2y ago
brian_goetz:
>Perhaps I can shed some light on this apparent contradiction.
>We discovered, unfortunately quite late in the game, that the design was flawed. How was this discovered? By actually using it in a substantial, real-world internal project. But these flaws had little to do with "someone didn't like the syntax"; they were much more fundamental to the model.
>Unfortunately, most of the feedback we received on public channels was of the "someone didn't like the syntax" variety (at varying levels of constructiveness, not all with positive coefficients). It did not derive from actual experience or from deep analysis; the supporting arguments provided were emotional, not technical; and it merely re-trod, with little new perspective, already-explored ground. It was literally a thousand drive-by, "blink" responses.
>The net effect of this was not only unhelpful, but anti-helpful. I get that every developer has opinions about syntax, and seemingly some fraction of them will literally explode if they don't express it out loud, but such discussions are like an invasive species: they suck up the oxygen that is required for actually useful feedback to survive. (Or maybe the useful feedback was buried in there, but just got lost in all the noise.)
>So the answer to your implied question is that its possible to be both: the feature can be discovered to be flawed, while at the same time the publicly available complaints were "nothing", as you say. It is sad, but possible.