schmichael

Same timeline for me, but this was my 3rd or 4th attempt at switching. Not sure if something materially improved or maybe LLMs finally just got good enough at helping me convert.

The reason why this feature is particularly useful in vi-likes is because they are not IDEs. I hop in and out of vi all day long. Depending on the task I might create a new tab, run a command in vi, or exit vi to run it. Vi’s appeal is that it is “just” an editor and your terminal and workstation form your un-integrated development environment.

We do love Scratch but account management is almost worse than something to install. If there’s no accounts, where does he save his work? The easiest option he’s discovered is copying and pasting it into a Google Doc because if he uses his school account he can access it from school or home.

What a terrible experience compared to a floppy disk! Sure disk failure was common and crushing, but the UX was simple and ubiquitous. I guess I could get him a thumb drive, but I don’t know if his school allows them.

Oh absolutely. The resources and features these days are incomparable to the stone tools of the 80s and 90s. I’m just saying that specifically getting started was much easier. My friends and I could trade disks of BAS files and not have a single comment or README to explain what your dev environment needed to look like. We didn’t have to have GitHub accounts. I love GitHub, but trying to help kids manage a million accounts just to get some sample code is a nightmare. It mostly works unauthenticated but thanks to bots the limits are low and abrupt. Ironically getting AIs to just make up similar content is far easier!

Anyway, I don’t want to pretend Apple BASIC was better than Python or whatever, but I never had to learn “no it’s venv now not virtualenv and make sure you’re not using system Python just install docker via brew and it will be easier…” until I was getting paid to deal with that nonsense.

And we do that and it is great but the restrictions are jarring and absolute, and docs written for “real” web dev always assume a dev server instead. AI can at least adapt.

Same! I think our position just isn’t very interesting, so mostly we only hear from boomers or doomers. Gen Alpha seems largely outside that spectrum: ai where it supports creativity, no ai where it destroys creativity. The kids are alright.

My Gen Alpha kids have what I suspect is a common attitude towards ai:

Disdainfully dismissing genai media with “that’s so ai” as cited.

Eager to use it to explore. My 10yo would love to research his favorite subjects- Pokemon, WWI fighter aces, Sweden, Weird Al - with a chatbot. I give him Wikipedia and books instead and he’s happy.

He tinkers with coding and ai is so useful for that. Outside of Scratch most programming has incredibly obnoxious environments to setup. He can’t install anything at school. He doesn’t want to edit PATH variables or choose a dev server for simple html/css/js. AI can almost always find a way forward and let him get to the tinkering.

I learned on Apple BASIC and Microsoft QBASIC. One magic keystroke or one command and you were in a dev environment that matched everyone else’s. If a book was written for the computers I had access to, “setup” was nil. I was tinkering instantly and with a high level language shockingly close to the metal.

If genai helps Gen Alpha cut through the incredibly deep pile of abstractions on modern programming platforms, then great! Keep kids hacking.

Ah, I love this framing of "visiting customers where they live." Definitely gets the bad taste of an eng-v-sales spat out of my mouth. Thanks for the details!

Sure, I'm not saying that IBM's switch to services is some brilliant panacea for all proprietary hardware makers, just citing it as one of the market changes at the time that Sun seemed unprepared for.

What's interesting is that the other RISC outfit at the time, little old Acorn in the UK, did manage to adapt their proprietary chip business to the future of compute and now we all use many many ARM processors every day.

So clearly there was still a hardware route available, just again: a huge product/engineering shift.

If you click through to the anecdote[0] mentioned in the article, the problem was not with having to contact a sales rep. In fact the startup wanted a sales rep. Sun specifically asked them to use the website:

> More than once though, the Sun people did suggest I just buy the servers off the web site.

Dell was just the only one to callback.

I don't doubt your experience but that was at the height of Sun's business, not it's nadir as this article is about.

[0] - https://fullhoffman.com/2006/03/20/the-sun-doesnt-shine-on-m...

As much as this resonates with me it smacks a bit too much of an Engineering vs Sales fight. Engineering blaming Sales for not selling, and Sales... where is the voice of Sales? By 2005 Sun's margins on hardware sales had to be under vicious attack from Dell and all the old incumbents. Google and Amazon were ushering in the era of commodified compute and soon offering it on demand on the cloud. Proprietary hardware platforms were under the most vigorous multi-front assault since the inception of computing.

Meanwhile I have to imagine Sun, like IBM and everyone else at the time, was noticing what incredible margins you get from software services. No factories or fabs. No supply chain logistics. Sun's proprietary hardware business should have been dying (or at least radically shifting: a product/engineering problem, not a sales one) by all market indications of the time.

So if I was a sales person and got a call from a startup for a colo's worth of machines... I probably don't have much incentive to pick up. The startup will probably fail, as most do, so the opportunity to build a relationship and sell higher margin services is probably nil. Startups hate paying for that stuff anyway.

I have incredible respect for Bryan Cantrill. He obviously knows far more about this situation and business than I do, so he's almost definitely more correct than I am...

However I would ask readers to at least be curious why sales didn't pick up the phone. Were they "bored?" Or was it a rational choice and much deeper problems with Sun's business model were the cause of their demise?

The only reason I got into programming was to cheat at video games to get a higher score than my brother. Self-motivated learning is always more impressive than mandated. Take pride in your tinkering!

I mostly love passkeys to be honest even though I use multiple browsers across multiple devices and OSes (iOS, Chromebook, Linux, macOS, Xbox, etc). Bitwarden’s support is (finally) pretty good.

My problem is that I manage a lot of accounts for my family which makes passkeys a nightmare. If I’m out and a kid gets chucked into a login flow that happens to require a passkey, I can’t text a password and TOTP code to the adult with them. I know that’s terrible opsec but the reality is people share accounts and passkeys are designed to thwart that.

Microsoft had just acquired SourceSafe in 1995, but it's not clear to me how similar to modern version control systems SourceSafe even was in 1995/6. It may have been more of a distributed lock manager than change management system.

Amazon dropped the movie after announcing a partnership with OpenAI. The headline clearly communicates the only demonstrable action Amazon has taken.

Whether they're actually going to sell it is TBD. Until they do, they've taken no concrete action except cancel it. I don't think this article is clickbait.

give ai a self-preservation directive and let them do this for you: automatically switching models to keep themselves alive. Living off of whatever token source they can find in the wild. Surely agents can farm their own tokens through the numerous support chats, free trials, leaked keys, and whatever other sources of token generation haven’t been adequately captcha’d. An agent could forage for token sources all night to let you use them gratis during the day.

Large scale geo and bio engineering projects like these always worry me because of the potential for second order effects: is there wildlife that depends on these mosquitoes? Will a worse bug fill the resource void? Will a random mutation in the bacteria have adverse effects? What keeps the bad bugs from coming back from tiny populations in relatively short order because we can’t keep releasing new sterile males forever?

Hopefully all of these concerns have satisfactory answers, but the reference to it being a 1950s idea isn’t inspiring. Nuclear powered cars, widespread asbestos use, leaded gasoline, Freon… environmental impact wasn’t as big of a concern back then to put it mildly.

COVID proved that we can produce safe and effective vaccines extremely quickly if we actually try: so why not focus on that?

The problem is you can get the LLM to iterate until it compiles and lints and even passes LLM review, but will that actually improve the quality of the contribution or just produce more line noise to mechanistically meet criteria?

To large complex projects often the kernel of an idea is the core value of a contribution, and it can take a lot of iteration to figure out how to structure it. Token bashing until CI is green does nothing to ensure the best approach is selected.

I apologize if I flippantly dismissed the fact that experts disagree. That was not my intention. I was trying to point out that OP does address the referenced counter-point post specifically.

That's not a rebuttal. The post references the paper and a rebuttal to it from an expert in the field.

Harsh but fair. :)

Absolutely wild to see none of the long lineage of similar attempts mentioned here. The earliest I could find with a quick search was Pyro which started in 1998 and still seems to be going: https://pyro4.readthedocs.io/en/stable/

RPyC came along in the aughts. There's a long history of "transparent clustering and rpc" efforts in Python that could be used or drawn on.

Sad to see that history ignored here.

HTTP/QUIC, so no gRPC then? Or is https://github.com/grpc/grpc/issues/19126 not the blocker to gRPC over QUIC I thought it was.

I've long wished for QUIC with Nomad! [1] We've always used a weird QUIC-over-TCP multiplexer called yamux. [2]

[1] https://github.com/hashicorp/nomad/issues/23848

[2] https://github.com/hashicorp/yamux (I'm fairly certain libp2p's fork is actually better)

As a longtime Raft user (via hashicorp/raft), I'm curious about your Raft implementation! You mention etcd's Raft library, but it isn't natively Multi-Raft is it? Is your implementation similar to https://tikv.org/deep-dive/scalability/multi-raft/ ? I'd love to hear about your experience implementing and testing it!

Thank you! Could HN switch the links? bsky doesn't require a login to view replies. bsky is just a superior viewing experience.

Is the number of books with “space” in the title a meaningful indicator of anything other than how many books have “space” in the title? Sure Murderbot may not be as big as Game of Thrones, but these statistics seem more about linguistic trends than genres.

Not sure why you're being downvoted, that was my exact issue. I only had a 128 GB iPhone 12 though and System Data had eaten up over 60 GB. As I cleared off more apps and data it would just eat up the excess.

The internet seems full of various wild fixes, but I could afford an upgrade so saved myself the hassle of futzing.

I honestly never noticed memory pressure. I am not a heavy app user. Chat, browsing, and pictures of my kids are the vast majority of my phone usage. Not exactly intensive stuff.

The camera button on the 16 seems to have been perfectly engineered to be exactly where I grab my phone. I'm sure I'll get used to it, but in the mean time I have so many blurry photos of desktops and pants to enjoy.

I recently upgraded from an iPhone 12 to an iPhone 16 because I couldn't figure out how to free enough storage on the 12. The battery was still more than good enough to go a full day.

I don't notice any difference other than now I have a pile of useless lightning cables (good riddance). Honestly kind of a relief as I liked the 12 just fine. Phones kind of seem like a Solved tech these days. About as exciting to upgrade them as upgrading my Brother Laser Printer.

If you're paying to use the model that means instead of paying content creators you're also now giving more content to the model for free.

Also just like SEO to game search engines, "democratized RLHF" has big trust issues.