4 ms·
IMO the greatest change would not be in tcp/ip and below, but in http and up. If you had no knowledge of the current web, and I would pitch you html, css and js
by nubbee 12y ago
IMO the greatest change would not be in tcp/ip and below, but in http and up. If you had no knowledge of the current web, and I would pitch you html, css and js, you'd probably think it was a bad idea. Why would you use interpreted scripts over binaries if speed is mission critical. First you encapsulate your data in a markup language only to be parsed/converted back by the receiver, ridiculous. It also lacks semantics to make dynamic web applications readable for machines, not to mention the XSS vulnerability in the markup language that would make server side application unnecessarily complex. What a weak proposal, would never work.
- bradleyy 12y agoYes, but that's the thing: it does work. And anything that you created from whole cloth, empirically, would be highly unlikely to work.
- jephir 12y agoI agree, there aren't that many problems in IP compared to the www stack. Imagine if we had bytecode instead of Javascript, the web would be years ahead in terms of being able to create games, video editors, etc. and other high performance apps in the browser. Javascript has basically become a poor substitute for an actual assembly language.
- ddispaltro 12y agoI actually disagree. I think the "view source" menu item probably helped the web evolve faster. I think the trade-off made tons of sense at the time.