-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 On 07/05/16 10:44, Jon Fairbairn wrote:
Something that is missing from this type of discussion is any reference to design rules, agreement on which should be made before any suggestion like this.
The one this violates is “never make language design decisions to work around deficiencies in tools” I am entirely sympathetic to the argument. But I also have to read and write Haskell a lot, and to suffer from something that can seemingly be so trivially fixed *hurts*. Especially when languages like F# and Elm have taken a more practical stance and just fixed it. It's another "Haskell isn't practical/serious" argument from engineers -- and like it or not, engineers are the main users of the language.
We should definitely avoid success at all costs. However, I don't think the cost is too great in this instance. Especially considering that there is already precedent for this in import and export lists. Finally, I'd like to add that I appreciate your post and angle, and hope that the committee take it to heart in general. The design should be guided by (but not slave to) first principles. - -- Alexander alexander@plaimi.net https://secure.plaimi.net/~alexander -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBCgAGBQJXLdMHAAoJENQqWdRUGk8BRZkQAOcxrn2oBuJMZkOsDN3vN4z7 GkeDbf4N6EB//xeQxWYvGkzcFoVLsKY6CASZ4pthRKEn0hE7EgP8q7MZOuoTECt8 fvECtpL50v43GFQ8aY6thytKkPWIPuhaN9gL5jrH5S/Q5dS4cieaxEqdh27RBSYZ FFiLlOrpVPQcuIzo2EVVVReApbatNIu3LsUXWW0wLeEjqudEVRt4yLK6WMAdSdKO C5OMuZoU2G0Cw4LOQV5tDnUfj9RS06+AJIEUUvgk7xbhU46ZaNQHqde1pTDJ3w/K Iq3g5WFndt8JZfAN/2eTLCmWj4EsX/iBMDunQqwlmAxlez5qYIgNlnlB9a7guj/n 9BT9zLuxnc/SsKB33KIkFSlVyQsssBXWOqvRcQ6ND4yi26rTLJry0WvEZ5IDmdg+ hRdKKb2YvyHLBuqMp7rIBQ7fwfCegWfJpp4A+GsCbkKXfz2SZJCBdDT3UZMkgwIQ qXr4CRESKssGXh/ezz+vXltxTAPArfDb97vULr8PeIrMoBFacIR6PVhSxHS7aSeE Hbh7zD6S6ynobtFBReQo4KYGlXUOC9nNDktnpjUAc12xQoRwVqirAasJ5D1oMwkO LB6L6YyHVlfcqb8ZfbFAHpblCCrbuw4f3pi6a9YVUy4d9ZyhX0K0RFSEWcHxE1rG yWFHxsMF+/bf7AM+tX9z =/DKF -----END PGP SIGNATURE-----