3 ms·

From the comments on the blog: "I'm reminded of the guy who decides that there should be one standard because there are n divergent implementations.

by engtech 14y ago

From the comments on the blog: "I'm reminded of the guy who decides that there should be one standard because there are n divergent implementations. So he goes and writes his own. Now there are n+1 divergent implementations." That is probably the most likely outcome, but kudos to Jeff for trying. The idea of Markdown is great, but I found the implementation of links is less than obvious. (haven't tried it in 4 years, so there was probably other issues that I had that I've forgotten) The problem I inherently always end up having with "parses to HTML" syntax conventions is there are always warts where the syntax is harder to remember than the HTML it is supposed to parse to.

raganwald 14y ago
This guy? http://xkcd.com/927/ http://xkcd.com/927/
fusiongyro 14y ago
You can hardly blame links for that. There was no widespread hypertext format before HTML for conventional renderings to appear in. Most of the rest of Markdown has analogues in earlier plain text formats and is fairly visually suggestive. Common Lisp managed to unite several divergent implementations of Lisp (MacLisp, InterLisp, Lisp Machine Lisp, etc.) under a single common specification. This was possible because the Common Lisp standard was not created by some third-party out of dissatisfaction with the other guy's stuff but by significant representatives from all the big camps. The desire was for better interoperability, not imposing ideals on the competition. I think this effort, if Gruber supports it, is likely to succeed simply by incorporating most of the community. In fact, it could be even easier, because one of the principle motivators for the divergent implementations of Markdown is simply to create concrete specs around a concrete grammar. This doesn't mean there won't be divergences (Scheme, Clojure, OpenLisp, etc.) it just means that those divergences will be principled ("we reject the size", "we desire modern FP techniques") rather than accidental ("we used this regex instead of that to tokenize emphasized text").

by engtech 14y ago

raganwald 14y ago

This guy? http://xkcd.com/927/ http://xkcd.com/927/

fusiongyro 14y ago

You can hardly blame links for that. There was no widespread hypertext format before HTML for conventional renderings to appear in. Most of the rest of Markdown has analogues in earlier plain text formats and is fairly visually suggestive. Common Lisp managed to unite several divergent implementations of Lisp (MacLisp, InterLisp, Lisp Machine Lisp, etc.) under a single common specification. This was possible because the Common Lisp standard was not created by some third-party out of dissatisfaction with the other guy's stuff but by significant representatives from all the big camps. The desire was for better interoperability, not imposing ideals on the competition. I think this effort, if Gruber supports it, is likely to succeed simply by incorporating most of the community. In fact, it could be even easier, because one of the principle motivators for the divergent implementations of Markdown is simply to create concrete specs around a concrete grammar. This doesn't mean there won't be divergences (Scheme, Clojure, OpenLisp, etc.) it just means that those divergences will be principled ("we reject the size", "we desire modern FP techniques") rather than accidental ("we used this regex instead of that to tokenize emphasized text").