Ask HN: What language/framework you thought was the future but ended up dying?
20 comments
Honorable mention would be Java an in particular "agent"-frameworks. Thought it would be the future of automation. Perhaps people actually use it, I just haven't seen it implemented anywhere.
I had a similar thought. Obviously Java "the language" is, indeed, widely used and did become "the future"... but certain parts of the broader Java vision didn't quite pan out: the "mobile code" and "agents" stuff in particular. "Java in the browser" was big for a time, but ultimately faded. Which I actually find quite disappointing on many levels.
All of that said, I don't think it's out of the realm of possibility that some of the "agent framework" stuff comes back into vogue at some point - at least in specific niche domains - although it may or may not involve Java.
I had a similar thought. Obviously Java "the language" is, indeed, widely used and did become "the future"... but certain parts of the broader Java vision didn't quite pan out: the "mobile code" and "agents" stuff in particular. "Java in the browser" was big for a time, but ultimately faded. Which I actually find quite disappointing on many levels.
All of that said, I don't think it's out of the realm of possibility that some of the "agent framework" stuff comes back into vogue at some point - at least in specific niche domains - although it may or may not involve Java.
I actually almost like WPF and would still consider it if I had to do any an Windows-only desktop apps, but one of the few things that drove me away from it was the method of styling components. It made me almost like CSS.
Coming from MFC, Delphi and Winforms, I've never really enjoyed WPF. I felt so bloated (still does) and I hated the Visual Studio experience too. To me, it's in the same category as Java. Slow, fat and overengineered.
Microsoft were far too quick to abandon it and move on to something else. I suspect it's made a lot of developers wary and sceptical of new technologies coming from them.
For a brief amount of time I thought Dart would eventually push through in the web. It was interesting to see it's resurgence in the domain of mobile recently though.
On the other hand I didn't exapct Go to go anywhere. I remember briefly paltong with it a few years before it became popular and kinda just set it aside because I thought it was just some random toy language from Google.
On the other hand I didn't exapct Go to go anywhere. I remember briefly paltong with it a few years before it became popular and kinda just set it aside because I thought it was just some random toy language from Google.
I actually tried to adopt Dart v1 line at a startup buying into the WebComponents (too) early. The weekly updates broke compatibility so much and often that it had to be abandoned. Just as well as it didn't survive or perhaps because.
Yes, Go picked up pretty unexpectedly. But now I believe there are good reasons for that. Though I do not expect core Java developers to accept it anytime soon.
I very much like Dart, especially when compared to the other options available. The fact that Flutter exists makes it so much more appealing as well.
I knew a few people who were really into Dart before flutter, back when you could use Dartium to run Dart code in a browser (for testing and development only) without transpiling. I dont know if was ever based in much reality, but a few of them were convinced that Google was planning to eventually put a full Dart runtime in the mainline Chrome browser. There may have also been rumors spread about an earlier Microsoft Edge doing the same, if I'm recalling correctly. Of course, none of that never happened.
Meteor seemed interesting but poor leadership doomed it to maintenance mode. I guess it’s apollo, Now? Unfortunately I think it had the same leadership. Pitty for the various brilliant people there who bought into the vision
I had the displeasure of working on a production app built on Meteor and it basically pushed me from “I’m pretty flexible with tools that let me take shortcuts for productivity gains” fully to “I only want to use tools that make bad choices as painful as possible”
A rich internet application (RIA) framework--any of Adobe AIR, Silverlight, JavaFX. But they competed and none gained enough traction to become dominant. This happens over and over with universal toolkits. The current one is html+js web app/electron. Not my fave in terms of developer ergonomics or resource efficiency.
polymer + haskell
If I might expand the question to be about technologies in general, I'll submit OpenSocial[1] and Widgets[2] as two things I thought would be widely adopted, but ultimately weren't.
[1]: https://en.wikipedia.org/wiki/OpenSocial
[2]: http://w3c.github.io/packaged-webapps/api/Overview.html
[1]: https://en.wikipedia.org/wiki/OpenSocial
[2]: http://w3c.github.io/packaged-webapps/api/Overview.html
Enyo was an unfortunate victim of the failure of the TouchPad. The actual JavaScript framework was/is great and was actually the real precursor to React. Some of its core component based principles made it there. The original Enyo team breathed some life into it again by releasing Enact, a React based framework inspired by the original Enyo.
Old school Enyo is hard to work with these days and there are many many dead ends and the community is majorly lacking but for a sheer JavaScript framework it’s one of the originals and the best.
https://enyojs.com
Old school Enyo is hard to work with these days and there are many many dead ends and the community is majorly lacking but for a sheer JavaScript framework it’s one of the originals and the best.
https://enyojs.com
From languages I would have thought that golang and rust would make larger inroads, but maybe that is still to come.
Honorable mention would be Java an in particular "agent"-frameworks. Thought it would be the future of automation. Perhaps people actually use it, I just haven't seen it implemented anywhere.