3 ms·
This is really nifty as a proof of concept, major kudos to the authors. However, it completely kills accessibility, so it's probably best to avoid for any serio
by TheAceOfHearts 9y ago
This is really nifty as a proof of concept, major kudos to the authors. However, it completely kills accessibility, so it's probably best to avoid for any serious projects.
I've gotta ask, since I've never used Java applets... Wasn't this already possible back in 1995? Have we come full circle?
- pavlov 9y agoYes, around 1996-97 there was a lot of hype around desktop applications delivered as applets. E.g. Corel Office for Java: http://www.edm2.com/index.php/Corel_Office_for_Java http://www.edm2.com/index.php/Corel_Office_for_Java
- jeffreportmill1 9y agoIt was possible, but there were issues with performance, compatibility, security and configuration. Everyone had a different version of the Java plugin implemented with different platform Look-And-Feels (if you had the plugin at all). And then the browser vendors went to war against plugins. CheerpJ is a much better approach - doesn't require a separate plugin, doesn't have version problems, has inherent JS-sandbox security, uses the same look-and-feel and performs great (after initial download). For a large project, it's nice to have all the benefits of desktop development with a strongly typed language, but still have the ability to publish to the browser.
- switchbak 9y agoOne of the biggest impediments was that Microsoft had this crappy implementation of Java on IE that was simply useless. Making a portable (modern at the time) applet that worked in IE was essentially impossible. I have no doubt MS did that on purpose. All said, I think having a standardized binary ASM model in the browser is the right way to go, even if it is a couple decades later :)