Reinforcement learning, maximise rewards? They work because rabbits like carrots. What does an LLM want? Haven't we already committed the fundamental error when we're saying we're using reinforcement learning and they want rewards?
I think it's your responsibility to control the LLM. Sometimes, I worry that I'm beginning to code myself into a corner, and I ask if this is the dumbest idea it's ever heard and it says there might be a better way to do it. Sometimes I'm totally sceptical and ask that question first thing. (Usually it hallucinates when I'm being really obtuse though, and in a bad case that's the first time I notice it.)
> I think it's your responsibility to control the LLM.
Yes. The issue here is control and NLP is a poor interface to exercise control over the computer. Code on the other hand is a great way. That is the whole point of skepticism around LLM in software development.
> GNU did not have a working system until Linus released Linux in 1992. They had pieces and components which were worthless on their own.
People were installing GNU onto existing Unix systems because GNU was better than they were distributed with. Maybe they did that with components of BSD Net/1 - no one has ever told me they did but it probably happened - but that was definitively post GNU.
Anyway, I'm not sure if this matters so much to the debate. Stallman was reacting to a change. He rambled politically and wrote some code to back it up because he used to be able to do things, and now he could only do them if he would write some code and win some allies.
Linking against GPL code on a backend server which is never distributed - neither in code or binary form. (Because what might happen tomorrow? Maybe now you want to allow enterprise on prem.)
My favorite one I think is the Internet Explorer/Google Chrome "Same shit different - " one, because it's obviously recent and somehow iconic of the sort of person who reminisces about the old web, and clearly narrowcasting to such people.
I'm trying to brainstorm an answer. My best guess is that SSH is obsoleted by disposable instances. You can spin up a new instance for every version of your configuration, transition to it, and dispose of the original (or set it aside or whatever). That way, you could probably have a reasonably complete tech career and only ever use ssh as an implementation detail of git.
I think OP's point isn't to prevent toxic assholes from saying whatever righteous things and fighting whatever bad fight, but to limit bot/inorganic/foreign contributions from made up people - basically to make it "one person one voice".
I kind of like the idea of "one person one voice", but I have two problems with it, which I think will block me from accepting it.
One is that the cost of it seems much too high, even if you can change it to allow the use of chosen aliases (I don't think it matters what a "one person one voice" system calls an authenticated member). I don't really trust everyone who I have to give my ID details too, and this is just one more bit of stress for so little gain.
The second is that the benefits will never be realised. In an election, one person one vote doesn't work when half the population doesn't vote; you need almost everyone to come, otherwise it's the strongest opinions not the mainstream opinions that dominate. And I'm quite sure we'll see the exact same thing here, but in spades, and faster. If you don't like the opinion, you just don't show up. Once the centre of the social media is sufficiently different from the centre of the community, there will be the sort of bullying and self censorship you foresee and it will spiral out of control.
In case someone gets the misapprehension that there is a contrast between systemd and launchd in terms of the "well documented" attribution, systemd configuration is also well documented e.g. man systemd.timer etc. I didn't know if launchd has an equivalent of timers, but it does and I've just read `man launchd.plist` "StartCalendarInterval" and compared it with `man systemd.timer` "OnCalendar". I would have said they're about equal. Launchd is more concise, but systemd talks a lot about the interactions with other settings and edge cases.
As for ini vs xml, I've generally found xml is a crueller syntax for humans than ini. At the time I started using systemd, it was a bit funny - the last time I'd been editing ini files was on Windows 3.11. But I think ini and toml are now once again reasonably common so I forgot about how out of place it felt at the time.
I've never taken it as wish that employees have some cult like adoration, but as a team building exercise. I don't like it, it's cringe, but it's nothing worse than a bit of meaningless cringe. I have heard the theory of the fine line between a CEO and a cult leader, but I've never worked at a company that came anywhere near that fine line. Every CEO I've worked with has known that we're there due to a mutually beneficial agreement.
I don't know what other solutions there are to TOFU, but maybe it's nice if there's something like a standardised /.well-known/ssh-keys.json path for public ssh servers like github and pico.sh.
There’s SSHFP, but it’s off by default and assumes an attacker can’t modify dns, though most mitms would be executed with dns and dnssec deployment is generally a disaster.
Currently their host key page is only linked once at the bottom of their page and isn’t referenced in any onboarding docs, so effectively onboarding encourages “yolo”, and if users aren’t savvy they’re likely putting other things at risk, whatever their keys happen to also have access to.
The other argument that comes up here then is “well mitms are rare so this doesn’t seem like a big problem in practice”, however there are actually great targets here, for example you go to a conference and hijack the WiFi, then spend your time in hallway track advertising these services to your targets. This kind of thing has a high success rate.
The web improves on this problem with PKI, though similar phishing tactics exist in a similar situation where you encourage people to sign up explicitly guiding them to an incorrect domain, but propensity for using search in address bars strongly helps resist this too.
SSH is terrible for this use case, no matter how it makes people feel.
