<div dir="ltr">Johan Tibell wrote:<br><div class="gmail_extra"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div>I think the biggest user will be Haskell code in general, </div>
</blockquote><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div>because you can now use |> instead of ".", so we </div></blockquote><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div>should be optimizing for that, not for any particular </div></blockquote><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div>library.<br></div></blockquote><div>
<br></div><div>I agree with that sentiment.</div><div><br></div><div>The main reason I proposed what the lens people</div><div>are using is because the lens world is such a</div><div>good testing bed for this.</div><div>
<br>
</div><div>Lens covers quite a large swath of interesting</div><div>types, paradigms, and use cases. And of course</div><div>a large number of combinators and operators.</div><div>There seems to be a good size group of people</div>
<div>hacking on all that and using it in practice.</div><div><br></div><div>For something as basic as this, all the little</div><div>details matter. All of those details have gotten</div><div>some real-world testing, and comparison with</div>
<div>their alternatives. Making it flip ($) instead of</div><div>flip (.). The precedence. The fixity. And also</div><div>the color of the bikeshed. It turns out that the</div><div>difference between a one-character name</div>
<div>and two-character name really is significant</div><div>in practice, more than you might realize,</div><div>even without factoring in the grand</div><div>operator-naming scheme in lens.</div><div><br></div><div>Thanks,</div>
<div>Yitz</div></div></div></div>