This is a really interesting project, and I love the idea of a different kind of shell. Why do we need to pretend we’re working on a 1980’s terminal in 2014? Interacting with the shell in a more stateful way (just a hidden, searchable command history isn’t enough) is a great idea.
But, if you’re just curious, this project makes it really hard to casually check out. (Unless you’re already a rubyist who also likes to mess with emacs, I guess.)
You need to set up ruby, probably rvm, install a bunch of gems, then run some ruby script that copies something to /usr/bin (really?) and finally you’re ready to start the program. Which just launches a daemon and prints a page of some ultimately less than helpful README. And unless you’ve got the right version of ruby (version 1.9.3 isn’t enough, you need 1.9.3-r<something higher than I had on my mac> ), it’ll just crash after that.
The next step is apparently to access the web server you just started to get instructions on how to modify your emacs/whatever setup to actually get things to work, in emacs' case, but adding el4r to talk to the xiki daemon. I managed to create files with emacs that the web server saw, but gave up after too many minutes of futzing with trying to figure out how to execute commands in place like in those screencasts. I guess I should have followed my first impulse when I was installing ruby and it advised me that I’d better “HTTParty hard” or whatever. I would’ve saved myself an hour on a nice vacation weekend.
To end on a positive note, let’s hope the xiki kickstarter goes well and they can spend some time on packaging and documentation. Don’t just rely on screencasts!
Yes, trying to set it up was definitely disappointing. I was really keen to try it out but I didn’t get anywhere. I’m not sure Kickstarter will succeed judging by the contribution curve, which is a pity.
Wow! This is a great solution to a problem that I’ve never ever had!