andrewl-hn

For starters, compensation is almost entirely location-based now. Situations where a person lives in cheap city / country but paid close to US rates were never common, but in last 5 years or so they disappeared almost entirely. Moreover, every company seems to know about local rates, they all source this data from the same sources, so if you're job hunting they all give you the same offer. And, since there are so much competition for positions companies are very reluctant to bid on candidates, and most offers do not allow negotiations. Even if paying you slightly above market would still save the company quite a lot of money they don't do it. So, situations where you receive the same offer from multiple companies, and all of them are "take it or leave it", became pretty common.

I haven't been on a job market for a long time, but this is what I hear from people around me.

Honestly, good luck with that. Mozilla spent past 15 years or so creating and killing new projects over and over. Right now I'd be hard-pressed to try anything they create because they would inevitably shut it down next year.

One other "AI" thing that I really liked came before the LLM boom. Mozilla made a speech-to-text engine called [DeepSpeech](https://github.com/mozilla/DeepSpeech) for running on-device when you use dictation in the browser. Apparently, it was discontinued last year, so I'm not sure what they use now.

Around me people use the verb google for that, too: "I googled AI / GPT and it told me ..."

I do the same thing, except I usually run a version for two-three years and skip some OS releases in-between.

My recent upgrade was from Ventura to Sequoia, and I will likely move to Golden Gate at some point next summer right before Sequoia's support period expires (this happens in July or August). Over years it meant that I avoided some of the famous bugs like the discoveryd fiasco in Yosemite (I went from Maverics to El Capitan).

The one downside is that I sometimes have a to go through a huge app redesigns or big feature changes that have been accumulating over several versions. I ignore most of them and probably don't get as much value from the OS as I could have.

Meanwhile there are many masters that make these tiles today - I know one personally! - and they all struggle to find clients.

I suspect they make a big model first. In this case it's Fable. Then they run the shrinker steps to make Sonnet and Opus. Sonnet is smaller, takes less time to make, so it got released first. Opus needed few more weeks to cook.

With this iteration they had a delay because when the Mythos was ready they had some sort of "Oh shit" moment and spent half a year adding safety guards to it. Then slowly rolled it out, but got another delay due to a government block. So, maybe the work on making Opus and Sonnet only started after they got a green light from the administration.

Presumably, now that they learned how to do this safety-wrapping the next iteration of Mythos / Fable / Opus / Sonnet is going to show up faster.

Something like that.

Just so you know, "move to Europe" is pretty complicated inside Apple ecosystem. You can't move your App Store account from one country to another while you have any subscriptions active. You cancel them, wait for them to expire, and only then you can switch. So, for example, if you subscribed to an app yearly in January, and then cancelled your subscription in May, you still have to wait until January of next year. Very annoying in practice.

It's less about platforms and more about games themselves. Games on Steam has to follow certain rules that are mostly consumer-friendly. And the fear is that some of these rules will be relaxed to increase Steam's revenue, and as a result the store will be flooded with ad-ridden gatcha-slop that is completely saturated mobile.

I wish I could edit my post.

Somehow everyone wrote to me about baldes. These are not the same, though. Blade servers were mounded into units of 4u, 8u, etc, they occupied a portion of the overall cabinet and still had to do "plumbing" for power and networking behind the chassis to the rest of the cabinet or to the rest of the datacenter. A full-cabinet blade rig would have multiple 8u blade units and some off the shelf units for networking, storage, etc. Yes, you could mix and match different components based on your needs, but that also meant that there were extra wires, cables, mounting rails, and more importantly - all these different components ran a mix of software that had to integrate using common denominator protocols and speeds.

Steve rightly mentioned the integration below, and I didn't put it in my message because I kinda assumed that we include software in this discussion too.

HP in 2005 had an army of programmers writing all sorts of firmware and software and another army of hardware engineers, too. They could have made an Oxide computer back then, and it would sell really well. But they didn't, and none of their competitors did despite this being an obvious product (in hindsight), an THIS is what I find interesting.

This design feels very obvious-in-hindsight. Consolidate power adapters and networking, replace cabling with pluggable slots. It's something similar to what IBM mainframes or Sun cabinets could've looked like. Somehow hardware giants like Dell, HP, SuperMicro, etc didn't make a product like this, even at their peak in 2000s or during cloud boom in 2010s. I wonder why?

Beautiful machine, and fun to see Illumos heart still beating inside!

A lot of it depends on what kind of company / industry you work in. For example, in Rust space there's big divide between LLM-heavy users and light/no-LLM users. Main rust repositories: the language, standard library, formatter, linter, LSP - all forbid LLM-generated code, and lots of third-party libraries, especially low-level and security related, follow suit. However, application code written in Rust is often LLM generated. So, if you work with Rust and you contribute to core Rust projects you have to write a large portion of the code yourself.

But when I talk to my friends who work with TypeScript it seems like they all moved to LLM-generated coding full-time.

A lot of corporate speak is developed in the US and then companies all over the world spread it around. Often the adoption happens without deep understanding of the concept, and without adaptation to local realities. And thus it feels much more unnatural.

> Apple very rarely admits mistakes

Excep every time they do a big redesign like this. This happened when they moved away from skeuomorphism in iOS7(?) and then backpedalled hard in the following revision because of negative user feedback. Similar thing happened when they presented the reinvented Safari (I do't think that one even survived through betas). And it is happening now.

Something that not many people consider. It is almost certain that C++ "saved" C by existing. Without C++ there would be an enormous pressure to add more features to C itself. One reason why C committee could get away without adding much over the years was that they could nod towards C++ and say "that's their job, not ours". And if later didn't exist who knows what kind of language C would become. Classes? Templates? Lambdas? We can only speculate.

To be fair, so are many other UIs. Windows 95-style boxy buttons and bevels make the content look organized. Every possible action gets its button that looks like a button. You often see the total set of available actions by looking at a toolbar. You don't need to second-guess whether some piece of content itself is clickable / editable or not.

Also, everything has excessive padding now. Modern Windows control panel UIs often feel like a multicolumn wall of text with lots of empty space and a few switches dropped in, and to fit the same amount if options as the older UI they had to either hide some toggles because "known needs them anymore" or introduce extra intermediary navigation steps. As a result the new Control Panel feels bloated and less useful.

ChatGPT is a word now. People may use Perplexity, or Google, or Grok to ask questions online. And later they tell you "ChatGPT told me this". It's a new "I googled in Yahoo".

AI code is uncopyrightable. If you want to be hygienic about it you would let AI commit code as is, and if you don't like the code and decide to change it manually this should be a separate human-only commit. Mixed human-AI commits add ambiguity about copyrightability of that code. So its should be either:

    Author: Some AI model, Committer: You Friendly Dev

or

    Author and Committer: Your Friendly Dev

"Co-authored by AI model" is a nonsense that AI labs are pushing everywhere because they definitely want the copyright rules to change, so that while their armies of employees feed LLM prompts every day for the past 4 years they can still claim that the IP is theirs an theirs only. "See all Co-Authored commits, your honor? There's no way to split them apart, so they should all be ours!"

The laws will likely change in the US thanks to lobby, and maybe even change in Europe, too, where they may compromise about it to "not be left behind" and will present them under "Digital Sovereignty" umbrella.

Overall, this "Co-Authored" bit is yet another Trojan horse.

I'm surprised they even had them in a first place. Doesn't Microsoft have a deep partnership with OpenAI? Aren't all Copilot things powered by various GPT models? I would assume the two companies have barter agreements of sorts.

Tokens are current era' "lines of code per month"

https://www.folklore.org/Negative_2000_Lines_Of_Code.html

They lay people off and look good in front of investors. Then they hire people, talk about "growth", and once again look good in front of investors.

This would never fly if stock market was rational. But it never is.

I remember when Mozilla released OdinMonkey that was hyper-specialized for asm.js code, the Chrome / V8 team instead worked on general-purpose optimizations in their JIT that would run normal JavaScript faster but also would help asm.js. The difference in speed was 2-4x in favor of Firefox, and they hyped it a lot :D

Nowadays most browser JavaScript VMs converged to very similar designs and optimizations, so even without Odin asm.js code would run pretty fast anyway.

> Alchemy

Yep, that's the name. There was a brief period in late 2000s when Adobe was pushing hard to make Flash embedded into web ecosystem. They made Air as a way to package Web or Flash code into a desktop app. Essentially it was Electron-before-Electron. Alchemy was a part of this grand plan to be able to integrate existing native libraries with Flash code. The plan was like you said to compile to Flash bytecode, and AFAIK it never went further than a tech demo.

This whole ecosystem turned into a slow train-wreck over approximately 5-year period. Adobe really saw themselves as future stewards of web technology. They donated their ActionScript VM and a JIT to Mozilla and hoped that Firefox would become the first browser with fast JavaScript engine. Google developed Chrome and V8 in secret and managed to release their fast browser early. Microsoft and Yahoo sabotaged adoption of ActionScript dialect as at the next JavaScript. And at the same time Apple went fully anti-plugins, and with the rise of iPhone both Flash and silverlight died off.

Years later Java folks tried to build their own version of Alchemy as part of GraalVM. The project was called Sulong and was using Graal to execute and JIT LLMV bitcode. TruffleRuby was supposed to be a primary early beneficiary to be able to compile and run Ruby native extensions. This was during the period of a race between several JIT solutions in hopes to become "the next Ruby", and Truffle team (along with IBM's OMR) lost the race first to MJIT and then to YJIT. Graal itself seems to loosing steam, because their multi-language VM never got enough adoption among Java, Node, or Ruby people, and the VM itself tended to use too much RAM in era when RAM became premium in the cloud.

At least some of the projects in these industries now specify strict no-AI-use policies in contracts. I participate in a few of these, and it’s becoming a bit of a pain, because all dev tool vendors insist on adding AI features, and if there’s no way to turn them off completely we have to migrate away.

However, the temptation of productivity gains are strong, and few of the customers look into relaxing these rules.

Let’s be honest. If anyone would be building a brand new company in Poland or any other country with the intention of raising capital or IPOing they would still incorporate in the US or a handful of other countries. So any successful Polish company would count as American/ German / Singaporean / etc anyway.

A lot of SMBs use Instagram to connect to their clients, so Instagram build-in messenger is a default option for a lot of people (especially women) in many parts of the world.

Some places have regional messengers that are very entrenched, like Line in Japan or KakaoTalk in Korea.

WhatsApp is a default option in a large number of countries including most of Middle East, parts of Europe, Brazil, most of Africa, Southern Asia. To me it is surprising, too, because out of all messaging options WhatsApp seems like the least developed and least ergonomic.

And yes, this does mean that most people share whatever data Big Tech wants. They use Meta to talk to each other, auto-upload their photos to Google, click "accept" to every cookie banner so that thousands of no-name companies around the world know where they are and what they are doing at all times.

> It's a social network that became socially acceptable to browse at work.

YMMV. I’ve heard a few stories where opened LinkedIn at work was treated as a massive red flag: “this person looks elsewhere, they are not committed to the company anymore”.

This would probably push some high-end audio professionals away from Logic. One of the niches Mac Pro has been popular is audio production. And with cheesegrader the ability to slot in many-many different audio interfaces into a box instead of dangling out to various PCIe enclosures has been a big win.

Here's a good video how it looks like: https://www.youtube.com/watch?v=kIQINCWMd6I&list=PLi2i2YhL6o... (at 1:40 Neil Parfitt shows Mac audio setup his before and after).

Very good, actually. But you have to nudge them slightly. Tell them you prefer the modern version of the language, with gradual typing† and function signatures, and you'll get very good results. Perl interpreter comes standard on modern OSes and due to permissive licensing and impeccable backwards compatibility you can always assume you deal with very modern versions of Perl.

I write Perl scripts that are 10-100 lines of code, and at this size Perl is a Strictly Better Bash: better syntax, some type checking, better text support, and still effortless calls to external processes: essentially you put a command with arguments in backticks, and you get it's output. Ruby can do it too, but not all systems have it. Python is another obvious choice but calling external commands in it is annoying. I also use Perl for some one-liners as a better `sed` for text replacements.

† Perl nowadays have TypeScript-style type checking for function parameters. So, while the syntax is wild sometimes, the language is much better than it used to be.

That's Reveal.js / Slides.com format. It became very popular in 2010s. The idea behind the 2-d navigation is that you can use left-to-right to move between chapters, and move down to dive into a specific chapter. This allows you to skip chapters due to time constraints. Or hide gnarly details about something so that these specific slides do not break the flow of presentation but still having them available for the audience online. Or, having slides announcing demos, but if demos do not work the down slide would have a video demonstrating how the demo is supposed to work. Many possibilities like this. Also the slides are produces using Markdown, so the format was appealing to many authors.

However, doing chapters well turned out to be tricky. Ideally you want them to be of similar size and have 3 to 7 of them in the talk, but many presentations aren't structured like this. The rise of Slideshare and SpeakerDeck for sharing slides in mid 2010s caused this 2-d navigation to go out of favor: those services only support linear static slides. This is also a reason why people use fewer animations in slides nowadays and why tools like Prezi didn't catch on (that was another presentation tool with non-standard navigation that went out of favor very quickly).

Many people still use Reveal.js to make their slides but they stick to left-to-right nav only.