@orbfix I do mirror on gitlab.com and haven't been putting my new repos on github.com, but there's still the issue of SaaSS, so I don't use gitlab.com for anything but mirroring. I use cgit on my personal website (https://mikegerwitz.com/projects/). For GNU ease.js, I use Savannah.
@bjoern Thanks for the link. Self-hosting and federation are very important issues (I don't see my use in hosting my own #GitLab instance until it hopefully one day introduces federation; I just use cgit). I'll update my page to include mentions of that.
The GNU ethical repo criteria and my page do focus on software freedom---the issue of GitHub itself being non-free is an SaSS issue. It doesn't matter if GitHub was free software if it doesn't federate, because you'd still be using github.com, which you can't control. But it _does_ matter if the client-side JS is free software, since that code is executing on your computer, just as any other program.
Similarly, using GitLab on gitlab.com for anything but repo hosting has the same SaSS issues. But you're of course free to host your own. Unfortunately, without federation, we have a bunch of fragmented communities. It's a problem that I very much want solved.
So I agree that it's important. I disagree that software freedom is less important; they're related but separate issues that are both essential for different reasons and different types of freedoms.
After ~6 months, finally got an employer disclaimer of rights signed for FSF copyright assignments (which I already had, before the company I worked for was purchased). Executed by the CIO on my Birthday. What a coincidence. Happy Birthday to me.
To put this in perspective: it took ~1hr to get that signed at my company before the purchase. This has been a deeply frustrating process. Thanks to Ted at the @FSF for stepping in to bring it to a close.
Now I have countless things to catch up on code-wise, and no time to do it.
@blp I think that the current state of things gives us more of an opening to keep doing what we've been doing, being sure to adapt our message to the particular issues without sacrificing discussion of freedom. Our receptive audience has broadened depending on how the issue is approached and our stance conveyed.
I think corporations and governments are shooting themselves in the foot: they want control, but it's being taken to such extremes, that it's becoming painfully obvious the type of control they actually are exerting and the freedoms that users really are sacrificing. But the biggest problem is still getting people to realize that these are issues they actually care about. It takes more than a group of technology experts fighting for/against legislation to invoke change---it takes an uprising.
@fsf The phrase "Free as in Freedom" is more meaningful today than ever before. We often talk of users being "robbed" of their freedoms by non-free software, but many users aren't provided the opportunity to have something to be robbed of to begin with. Nearly everything users do is controlled and surveilled by corporations and governments as a feature, by default. Non-free software exposes and exfiltrates the most intimate aspects of our lives---it explores our thoughts, sits in our living rooms, and understands us better than we understand ourselves. Children are exposed to and taught to love and accept these software and devices before they can even crawl.
I've never felt more liberated by #freesoftware. But I've also never felt more concerned and sympathetic for users who are caught up in all of this. Most users don't even know that there are things to be concerned _of_, let alone where to even begin, despite the best intentions and predictions of many within our community.
And I've also never felt more compelled to do something about it.
I'd want to read them for the same reason I want to re-read SICP: I probably am familiar with most of the concepts by now, but it's worth refreshing and reading from different perspectives.
@1iceloops123 The issue with KRACK (assuming that's what we're talking about) exploits a 4-way handshake at the client level. In any case, this is a problem with WPA2 implementations that are correctly implemented---free software wouldn't of (and didn't) help here. Updates to wpa_supplicant (free software) fix the issue for GNU/Linux users.
(Btw my router does run free software, but ThinkPenguin unfortunately discontinued them.)
@cereal Your link requires JS to view; do you have an alternative link I can look at?
HTTPS will work as long as it's successfully initiated. If an attacker can inject packets, then they will be able to perform a man-on-the-side attack, where they can send a reply before the remote server sends its own.
Normally, when a user types "foo.com", if a webserver serves only HTTPS, it'll send a redirect to "https://foo.com". The MotS attack can reply first before such a redirect takes place.
HTTP Strict Transport Security (HSTS) was made to prevent this sort of thing. The strongest protection is the list of websites distributed with web browsers that force the web browser to connect over HTTPS, always:
@cwebber Take a break before you burn out, if you haven't already. Just make sure you accept that it's necessary to enjoy yourself and don't look at it as wasting time, or it'll be stressful rather than stress-relieving.