5 ms·

How useful are these discussions? In the end, boats still prefers prefix and others still prefer postfix. This is pretty fake discussion. Everyone already has a

How useful are these discussions? In the end, boats still prefers prefix and others still prefer postfix. This is pretty fake discussion. Everyone already has a position and pretend to be open to discussion.

steveklabnik 7y ago
They exposed all the options. That’s very useful. They made their case. That’s very useful. This is the most discussed feature of Rust ever. The team spent a lot of time and energy on it, way way way more than any other feature. Several team members flip flopped positions several times.
- jnordwick 7y ago
  And the team seemed to ignore the users. The two polls I saw, both had the method syntax below 50% approval and below the other front runners. It seemed to me they basically said, "Sure, tell us want you want so we can note it and ignore it." Kind of a real dick move. Why bother even asking for input? (For disclosure, I preferred postfix punctuation similar to `?` that most people hated - so I'm not gettng my way either.)
  - steveklabnik 7y ago
    Framing this as “the users” is disingenuous, again, a lot of users favored the decision too. The arguments of those users were taken into account, but the decision was made against it anwyay. Those polls were incredibly non-scientific, and don’t really mean anything.
    - jnordwick 7y ago
      And you had better polling besides just your cohort and what other said to you? This seems like a disingenuous argument, ending with you can't really trust anything but our own opinions. Yes, they were bad, self-selecting polls. You have nothing better, and I hear far more complains than people being happy with the decision. The polls both tended to agree though so that would lend some credibility. Maybe you should have just explained "We aren't really asking the users since we have way to show that any poll would be valid, but we will take some input and make up our mind from that." I think that is a very precise (and charitable) description.
      - steveklabnik 7y ago
        No, we didn’t have better polling. We don’t use polls. We did ask the users. Just not through polls. The teams always make decisions. That’s why they exist.
  - wtetzner 7y ago
    I'm in favor of the chosen syntax, but I was unaware of the poll, so I'm not sure that's a good representation of Rust users.
    - steveklabnik 7y ago
      Part of that is because the polls weren't made by the team; it was random other users. We don't vote in Rust, and polls have so many problems that we don't use them as a method to evaluate answers to questions.
      - jnordwick 7y ago
        So this decision is purely on the team - no saying "we took input from others and didn't do completely what we wanted to do.' In that case, you as a team basically saying that you were afraid of punctuation was intellectually shallow, lazy, and a terrible decision. It wasn't like a tailing @ or # would have turned rust into APL. You should go back and revision the `?` decision then under the same reasoning.
        steveklabnik 7y ago
        ok

steveklabnik 7y ago

They exposed all the options. That’s very useful. They made their case. That’s very useful. This is the most discussed feature of Rust ever. The team spent a lot of time and energy on it, way way way more than any other feature. Several team members flip flopped positions several times.

jnordwick 7y ago
And the team seemed to ignore the users. The two polls I saw, both had the method syntax below 50% approval and below the other front runners. It seemed to me they basically said, "Sure, tell us want you want so we can note it and ignore it." Kind of a real dick move. Why bother even asking for input? (For disclosure, I preferred postfix punctuation similar to `?` that most people hated - so I'm not gettng my way either.)
- steveklabnik 7y ago
  Framing this as “the users” is disingenuous, again, a lot of users favored the decision too. The arguments of those users were taken into account, but the decision was made against it anwyay. Those polls were incredibly non-scientific, and don’t really mean anything.
  - jnordwick 7y ago
    And you had better polling besides just your cohort and what other said to you? This seems like a disingenuous argument, ending with you can't really trust anything but our own opinions. Yes, they were bad, self-selecting polls. You have nothing better, and I hear far more complains than people being happy with the decision. The polls both tended to agree though so that would lend some credibility. Maybe you should have just explained "We aren't really asking the users since we have way to show that any poll would be valid, but we will take some input and make up our mind from that." I think that is a very precise (and charitable) description.
    - steveklabnik 7y ago
      No, we didn’t have better polling. We don’t use polls. We did ask the users. Just not through polls. The teams always make decisions. That’s why they exist.
- wtetzner 7y ago
  I'm in favor of the chosen syntax, but I was unaware of the poll, so I'm not sure that's a good representation of Rust users.
  - steveklabnik 7y ago
    Part of that is because the polls weren't made by the team; it was random other users. We don't vote in Rust, and polls have so many problems that we don't use them as a method to evaluate answers to questions.
    - jnordwick 7y ago
      So this decision is purely on the team - no saying "we took input from others and didn't do completely what we wanted to do.' In that case, you as a team basically saying that you were afraid of punctuation was intellectually shallow, lazy, and a terrible decision. It wasn't like a tailing @ or # would have turned rust into APL. You should go back and revision the `?` decision then under the same reasoning.
      - steveklabnik 7y ago
        ok