hotspur

Well yeah, almost certainly. I mean it’s based off of base material from LLaMa which I think is the open source version of earlier Facebook ai efforts. So it definitely used copyright material for training. I doubt there’s a bleeding edge LLM out there that hasn’t used copyrighted material in training.

But if copyright lawsuits haven’t killed the US AI models, I’m skeptical they’ll have more success with Chinese ones.

Haha this is so amusing. I’ll take that though over the blind confidence you get out of so many other products I guess.

How long before deepseek is banned by US gov for being too good—-wait sorry I mean a security risk?

Once you strip out all of the PR about democracy, freedoms, etc, the US really does fit pretty well as the single biggest threat to continued existence on earth: massive nuclear stockpile on a hair trigger, massive globe spanning network of carbon emitting and resource extracting enterprises, attempting to develop ai with minimal safeguards and autonomous weapons, etc.

There are other countries that do some or all of these things, but in aggregate, the US is, for now, gotta be the single largest threat. Possibly another country will take up this mantle in the future.

All that said, I’m not convinced at this point that it can be avoided. It may just be what happens one way or another with humans at these kinds of scales, or a natural progression of technology snuffing itself out. Doesn’t make it any nicer to contemplate though.

Yeah I basically agree with your point about the unpleasant logic behind such a move, and would only add that Greenland looks appealing if you’re trying to lock down the arctic from both sides of the continent—US has good arctic frontage on Alaska, and Greenland would bookend Canada and allow US more flexibility in countering Russia and expanding oil extraction.

I was trying to think about where this suddenly came from, and the first thing that kept popping up was Trumps current obsession with drill baby drill, the arctic is the last frontier for potentially easy extraction once all the ice melts and Canada, US and Russia have already been playing footsie there for a decade under the guise of science and commercial traffic trying to lay claim to stuff that was ignorable before.

Like some dude got in his ear and convinced him the future is in the arctic. It also adds some further explanation to Trump “joking” about making Canada a state. If it was just economic hardball / a new trade deal, they could leave it at tariffs and the like, but they keep saying they want to make it a state…

All of that makes me sick to my stomach, but as you say there is logic to it.

That’s a great info dump, thanks for posting.

Yeah that’s a fair point—like it’s fine as a brainstorming tool, but I’d never trust… uhh life and death details to it. And as OP mentioned, you can easily answer many of the questions with a web search.

It’s so amusing, I’m not a demolitions expert (tbh I would have thought an SF guy would have had better knowledge there, or at least better sources to go to) but the idea of trusting anything chatGPT said to help you lay out a plan like that… seems ludicrous.

Also very cool—the building in default modifiers to the command. Thanks for the great tips.

Wild had no idea—this is so cool. If you do this, does the original command also still function (so like I could rename to something easier for me, but hopefully transition to the real deal at some point/ properly follow help forums or suggested pasted commands)

Yeah honestly, for all the value this has, they might as well just dispense with any vetting whatsoever. We know it’s a farce.

I haven’t used competing apps to know, but as a forced teams user it is very sluggish, seems to break other ms apps half the time and has some strange and persistent design choices that irk me. It also crashes on its own, when I’m not using it 2-3 times a day.

It has improved in terms of features lately, but still feels very bloated and WIP most of the time. It still won’t let me control where video windows are, and I’ll never understand this.

This is our replacement for Skype, which was obviously feature deficient and getting old, but does what it’s supposed to do and doesn’t cause problems.

Not sure if there’s a good competing app in terms of video and slack functionality, integration into outlook and onedrive (both of which also annoy me and seem to be performing worse-over-time, but are unavoidable and sometimes useful.)

Well said-I feel the same.

I mean the real comparison is just: did she get enough votes, in states that Clinton lost, where if those people had all voted for Clinton, then Clinton would have won that state. I don’t know the answer, but even if the numbers did cover the margin, I think saying Stein is therefore a spoiler is problematic for a few reasons:

1. It ignores the very real number of voters who chose not to vote democratic or vote at all simply because of Clinton as candidate.
2. it ignores massive mistakes made by a hubristic campaign that couldn’t fathom losing to trump.
3. it supposes that people that voted green, would have gritted their teeth and instead voted Clinton, which is not a safe assumption.

Regarding OP’s argument: if Stein is a spoiler, than the libertarians are also spoilers. Since her being a spoiler assumes a majority of her votes would have gone democratic, we can take the same liberty and assume the libertarians would have instead opted for trump. If they had larger vote numbers than the Green Party got, as OP is saying above, then they cancel out greens spoiler-ness, and in fact represent a slight spoiler in favor of the democrats. I don’t really buy this read for the reasons I mentioned above, but OP’s point still kinda stands.

I’m not personally interested in voting for stein, I’ve heard enough weird stuff about her over the years that I’m not comfortable with her as a candidate. But I don’t buy the constant messaging that “third party votes are wasted votes”. My assumption with people that post these things is that they’re not suggesting it’s OK to not vote. And assumably, they also don’t want you to vote, but vote for the opposition. So it’s just the same old thing: vote the way I want you to.

Haha yeah also that.

Wild would never have guessed. Maybe because it’s possible for it to print as a continuous sheet, and I guess it comes in big rolls, etc.

Man, I wonder if it’s challenging to source a steady supply of paper for that thing…

If only someone could do something, anything… to stop this horrible tragedy!

No I didn’t know that, would be interesting to see more of them try it, just for curiositys sake.

It makes me wonder—would the dynamic change if there was only an upvote? So you could choose not to upvote, but the default action would be a neutral one, and if you liked/wanted to support/etc you could signal that.

I see tons of posts on here now that are downvoted to oblivion, because they are a legitimate article that says something a group doesn’t like. There won’t even be comments on the post. So like a Reuter article that discusses Palestinian casualties and no comments and like -20. This doesn’t seem like a super useful mechanism. Or at least, it’s just functioning today as a content preference “I don’t want to see this typed content” as opposed to “this is bad info, out of line with the community, etc.”

And despite ranking my list by either hot, or top day/six hours, I still see the downvoted posts regularly so the mechanic doesn’t even really do anything in terms of visibility. Or possibly there’s just too little content on a given community for it to get filtered out.