3 ms·
SVG would make a lot of sense here - but could you settle for content-hash-named transclusion of arbitrary bitmap images as a Unicode escape? It's been propose
by gojomo 8y ago
SVG would make a lot of sense here - but could you settle for content-hash-named transclusion of arbitrary bitmap images as a Unicode escape?
It's been proposed!
http://www.unicode.org/L2/L2016/16105r-unicode-image-hash.pdf http://www.unicode.org/L2/L2016/16105r-unicode-image-hash.pd...
- Analemma_ 8y agoOh god why. This is like the 21th century update to Zawinski's Law: every data interchange format will eventually expand to encode arbitrary binary data. Unicode is a text standard and even including wingdings symbols and emoji was pushing it, please kill this in the cradle.
- oconnore 8y agoThink about the 1700s, when printing presses had existed for a few hundred years (since 1439) to transfer data between people. Books were expensive, though. Would you have been able to imagine that in 1975, printed paper would be created each morning, and hauled around to everyone's door step in time for breakfast? Now think about us sitting here in 2018. Computer rendered text has been around for a while (ASCII, UTF, .doc, .html, etc.) In another 200 years, the way we transfer, encode, and view those formats will be completely perfected and probably unrecognizable to us. They will probably be ubiquitous in a way that seems completely alien to us now, and all of the implementation issues we worry about in 2018 will completely fade away (proprietary vs. open, security vulnerabilities, compatibility, accessibility, etc.). With this in mind, being reluctant to add enrichment to how document transfer works seems like it's obviously on the wrong side of history.