Octoth0rpe

The value of the Trump presidency for his voters was never really about those issues, so much as who he was promising to hurt.

> They haven’t been able to fix it with many multiples of that.

which may actually be the problem. I suspect that there is actually some ideal ratio that could be calculated of Input Fields / Dev, LoC / Devs, or maybe Unique Pages / Dev, or some mix of all of the above. Some of the metrics I hear out of places like airbnb absolutely blow my mind (>5000 engineers! wtaf are they all doing?!?). I can sort of see the #s at google, MS making sense given the breadth of the problems they are solving, but other places, not so much.

But at least they're passing on all the savings to the renters, right?

....Right???????

Project size is obviously going to be a factor, but so is machine specs. It's much more noticeable on a spinning disk. One can partially compensate for the project size aspect by opening vscode as far into your project as possible (eg, the api subfolder) rather than at the root. No real solution if you don't have an ssd though.

Because of how policies are grouped together in many countries (particularly the US), the fight against fascism is necessarily concurrent with the fight against climate change.

The ipad pro with keyboard is $1300, more than 2x the price of the Neo. Definitely not the same product category.

Up until a week ago, the base m4 mini (16gb ram/256gb ssd) was $399 at microcenter, now $499. Pretty shocking how good of a value that is IMO.

Well, better to be off by that much here than on my next jira ticket size estimate.

thanks

Part 2 has benchmarks: https://maderix.substack.com/p/inside-the-m4-apple-neural-en...

6.6 FLOPS/W, plus the ability to completely turn off when not in use, so 0W at idle.

> But it’s clearly not here just yet.

That's not clear to me, and apparently not to Block.

> There’s definitely going to be a shift over time.

Yep, and as the transition happens, it'll probably be bell-shaped. Someone will be on the left side of the bell, and someone will be on the right side. Block has chosen to be on the left side.

The current administration does not seem to be considering the majority within their own party considering how unpopular the current approach to immigration enforcement is. Or for another example, the glycophosphate/MAHA situation.

> but I still think most people would try to do their best to avoid firing nukes.

"most people" are not in the positions that matter. A significant portion of the people who are in a position to advocate for such a decision believe that:

- killing people sends em to heaven/hell where they were going anyway; and that this is also true for any of your own citizens that get killed by a counterstrike.

- the end of the world will be the best day ever

That might impose enough effort to stop most slop from being spammed everywhere.

I think it might be enough to be able to verify that a photo was taken by a particular manufacturer's camera, not necessarily a specific camera/owner.

Would you like to claim a limited time license for a /s for your reply? The use of this /s can be revoked at any time. You may only view the /s on a limited number of your own devices. A public display of this /s without prior written consent immediately invalidates your license to this /s, and you may be subject to a lawsuit in a specific court in West Texas where you must show up in person at a particular date with 48 hours notice.

> which is far from being a primary feature,

I don't know that I agree. If the adblocking functionality has gaps, then I will likely not use it for reading stories, and if I'm already opening up HN in my browser w/ ublock origin, then I might as well stay there and comment. Reading the article may not be the 'primary feature', but strong adblocking does seem like table stakes.

> Meta runs ads, and they can use AI to help advertisers create content, target people, engage, etc.

It is hard to imagine the level of spending they are doing if that is the sum total of their use case: shoring up a moat for which there really aren't any significant competitors in the first place. It seems like it can only be justified by eventually rolling out some kind of subscription service for... something, but for the life of me I can't think of what they might be able to actually sell to people or corps.

> They need to their eyeballs and not lose them to OpenAI, Anthropic or Google.

At this point I'm not sure how they could 'lose eyeballs' to those 3. There doesn't seem to be any kind of market overlap. Unless we're talking about the very abstract sense of doing _anything_ other than use a meta product is a potential lost eyeball in which case you might as well add the national park system to the list of people they can't lose to, and I don't think that's a useful way to talk about the cost/benefit of Meta's ai spending spree.

The current leader for me for worst questions suggested by Meta's AI was on a photo someone took of some conspiracy theorist's van with the spraypainted message "THEY EAT BABIES IN DENVER". The suggested questions from their AI were:

- Baby-eating restaurants in Denver

- Denver's unique food scene

wtaf meta.

Beyond that, I simply don't see how Meta can possibly ever monetize their investment in AI. People are and will continue to be willing to pay OpenAI, Anthropic, google, microsoft. No one will pay Meta for their AI. And if their investment was only a couple million and they got some useless suggested questions out of it, whatever. But the size of their investment sure makes it look like someone thinks they'll make money off of it.

Great point, thanks!

@IronsideXXVI, are you open to changing to gpl v3? Otherwise, there is probably a decent set of filter lists with an MIT license somewhere. The goal is for you to NOT become a filter list maintainer, and by piggybacking off an already respected set of lists, you'd build user trust in your adblocking.

> Built-in ad blocking — a precompiled WKContentRuleList blocks 14 major ad networks (DoubleClick, Google Syndication, Criteo, Taboola, Outbrain, Amazon ads, etc.) right in the WebKit layer. No extensions needed. Toggleable in settings

This is a good start, but I think a better approach would be to piggyback off of ublock origin's lists. Hopefully less maintenance that way too.

From this book in particular, I love the scene with everyone sitting around talking about how horrifying the concept of cheese is. The rest of the quartet is wonderful, with the second book (A Closed and Common Orbit) being the MVP IMO.

>Sometimes the actors have terrible pronunciation

I'm sure this could be easily dismissed as a regional (planetary?) accent.

I would say that this is the _setup_ for the bobiverse novels, not the plot. The plot is a whole hell of a lot larger at this point.

> They're a fun read.

Strong agree.

I believe they're referring to running Zed entirely in a browser. This opens up possibilities like using zed for something like codepen, or embedding it into a git web frontend like gitea. Many projects like this basically embed vscode, a rare benefit of being an electron app which Zed is not.

> Unless you believe that logic and "thinking" are somehow magic, and thus beyond the realm of computation

Worth noting that significant majority of the US population (though not necessarily developers) does in fact believe that, or at least belongs to a religious group for which that belief is commonly promulgated.

I'm currently using opus in Zed via copilot (I think that's what you're recommending?) and tbh couldn't be happier. It's hard to imagine what better would look like.

> (cc / codex / cursor / copilot (jk, nobody should be using copilot)

I seem to be using claude (sonnet/opus/haiku, not cc though), and have the option of using codex via my copilot account. Is there some advantage to using codex/claude more directly/not through copilot?

> we should more readily abandon GPU vendors

This was so much more practical before the market coalesced to just 3 players. Matrox, it's time for your comeback arc! and maybe a desktop pcie packaging for mali?

I haven't looked at a cost analysis recently, but it's possible that we basically already have $2k/month models, if they were priced to be even slightly profitable.