If you started emacs with –Q, it shouldn’t be using your .emacs, correct? That would seem to indicate it’s not a problem with your .emacs. Would think it’s a difference in your emacs installations or environment. What differences between the working and non-working emacs machines can you think of? (other than Win7 vs. Win 8).
I’m not familiar with slime, but from the traceback you show it looks like the problem is with url-retrieve (in url.el). (Are you behind a firewall? On one system but not the working one maybe?)
When I try that command (without slime) I get:
#<buffer *http proxy.us.abb.com:8080*>
And ‘nil’ in the message window. I’m not sure if that’s a good test or test result though.
Have you tried ‘eww’ or other emacs web browser? With eww I can get to that url ok (and have my proxy set up to get through firewall).
This page (https://github.com/tkf/emacs-jedi/issues/83
) suggests a similar problem was due to insufficient timeout value when creating an external process from Emacs on Windows, but it looks like that may be specific to the EPC package for Emacs or emacs-jedi. Perhaps it could be related.
Can you run other commands in Emacs that launch external processes? What other commands might make use of url-retrieve or make-network-process, or open-network-stream?
I see you’ve asked this question in a lot of places online. If you find an answer, I hope you add the resolution to them for others’ benefits.