crabmusket

The way I see it, databases are essential complexity, but global variables are accidental complexity.

It may be the case that "the systems I’ve worked on that haven’t involved a database [are] usually much cleaner and more elegant" - so why not propose that you remove the database from those other systems? You can't - the database, when you have one, is essential to the system's purpose.

What we often deride as "just crud" is indeed mediating between a client and a database. The app's job is making specific use cases effective and ergonomic.

> This must have been intended as hyperbole.

I sure hope so. The idea that if you want to work in [profession x] you have no choice but to pay the [megacorp y] tax is obscene, and should be fought against even as we acknowledge that it is often the reality.

We've come a long way since the days of Brandeis and the trust busters.

The author wrote more about Word compatibility in their 2022 post https://decoded.legal/blog/2022/10/running-a-law-firm-on-lin...

> I had wondered if I would need to switch to Word for some documents but, so far, no.

> I don’t know how many “Word documents” I’ve worked on in the last year - low hundreds, at a guess - but there has been no problem.

> Even a document with hundreds of tracked changes and comments, and multiple contributors across multiple organisations - an unusual situation for me, anyway - was fine.

> (Admittedly, it took me a while how to deal with LibreOffice Writer’s tracked changes function and comments, but that was simply user error.)

For anyone else who did a double take, this is different to Orkes Conductor which was the subject of a recent Jepsen analysis.

I'm seeing this claim all over this sub-thread and I will stick up for JS here.

While yes, PHP begun as a collection of macros and was never intended to be a programming language, JS was made by someone with intention. To quote from wiki,

> [Eich] originally joined intending to put Scheme "in the browser", but his Netscape managers insisted that the language's syntax resemble that of Java. As a result, Eich devised a language that had much of the functionality of Scheme, the object-orientation of Self, and the syntax of Java.

I've seen quotes before from Eich himself I believe that he was combining these influences at the time - Scheme's closures, Self's prototypes. It didn't "just happen".

That's fair, though I was somewhat responding to the short list of topics from the article and discussion in this thread.

I still don't think minimalism inherently demands what you're suggesting. If you don't find social media has any beneficial uses for you, it makes sense to cut it out of your life completely. Someone else might find they need some specific things from it, and decide to keep some engagement - on their own carefully dictated terms.

From 404's reporting,

> “Due to insufficient velocity,” Proton is not on track to achieve its annual recurring revenue target by the end of the year, one internal document obtained by 404 Media reads. Proton’s year-over-year growth is also “far away from what the market today considers to be good,” and, “By all metrics that matter, we’re an underperforming company.”

So, the "market" is demanding it?

POSIWID is dumb, but also, why are you ignoring that it reads posts from multiple networks? Why isn't that its purpose?

Actually if we take the phrase to refer to Cal Newport's book, no. What you're describing is digital abstinence. From the article,

> Explore ways of greatly decreasing or scoping your social media use ... Get the information benefits of the platform without getting sucked in to the vortex

This is more or less what Newport is recommending. Find ways to get the benefit of digital technology while refusing the manipulative aspects and other downsides.

EDIT: oops, TFA doesn't actually mention the book, I thought it did. I think this point still stands: "minimal" does not mean "none".

What did you find unreasonable about the footer?

> because git is an interface that agents are experts at using

I'd actually argue that this is the wrong way around. Being able to git diff legislation is very good, but mainly for humans, who need tools to make work tractable.

An agent will use whatever interface is available, including no interference, won't give up due to the task being too onerous, and can be scaled by paying for more tokens. It could simply brute force diff the raw text, writing bespoke Python and regexes, etc.

LLMs make brute forcing intellect possible. It's humans who need finely crafted tools.

Does this imply a forthcoming wet lab boom comparable to the data centre boom?

> Life throws obstacles at everyone.

This is sensible and pragmatic, but I think you're talking past the other commenter.

Obstacles presented by, for example, your boss, or your company's competitors, or your political system, are not the same as obstacles like heartbreak, natural disasters, emotions, illnesses we cannot yet cure.

> Learning how to dodge and weave is essential to survival.

A lot of people perceive that elites are throwing dodgeballs at them in order to increase their own profits. Saying "them's the breaks, it's just the way it is" appears to be deflecting criticism from those elites and that system.

It probably depends on whether you think we could actually feed everyone, today, if we chose to, or if you think e.g. human nature prevents that from being possible.

Would you reject the characterisation of model weights as a black box then?

I suppose if LLVM's source code has a similar size in bytes then it would be similarly difficult to work with. However, it would still remain source code readable by a human, not binary data with no semantic content when examined plainly.

I mean debugging within the model, not the harness. Harnesses are software like any other.

The model itself is more or less a black box - if tokens come out that you don't like, it seems like there's no equivalent to "this LLVM release fixed a code generation issue on platform X relating to situation Y".

I'm not sure why people are so motivated to say these are the same thing when they are clearly not.

And yes, rigorous approaches can be taken to systems built on and with LLMs, as you say. That doesn't make them the same as compilers.

Could you go into more detail about how you'd debug LLM output?

I don't really care about what "most engineers" do. I did phrase it lazily but I am trying to point out something fundamental about LLMs and compilers.

You can debug a compiler. If an LLM writes code you don't like, how would you debug that? You wouldn't. You'd just ask again and hope it does a better job.

> Deno 1.0 shipped in 2020 on the original terms, & then spent the next four years walking them back one piece at a time. First the deps.ts convention, one file that re-exported every URL: a manifest in all but name. Then lock.json, an integrity hash per URL: a lockfile. Then import maps, so you could write bare specifiers again.

I object to two of these categorisations.

deps.ts was a convention, not a built-in runtime feature. If you want to follow the convention, that's fine. It's just a regular old typescript file!

Import maps were not Deno walking back, but implementing a web platform feature. Parity with browser JS was a standout feature of Deno from the beginning. This was walking forwards if anything.

However, broadly, the story is indeed one of Deno moving away from its opinionated roots towards pragmatism (drop-in replacement for Node).

> who am I to say if someone else is willing to pay for it?

You are a moral actor capable of rational thought. The very image of God themself, if you like. Don't shrug off your own agency.

Something that comes up frequently on their podcast - often in passing, I'm not sure I remember a specific episode to refer you to - is the woes of trying to support/debug commodity hardware. Think waiting for Dell to get back to your support ticket about a BIOS issue. Or being told "well you're the only person with this problem, so it must be your fault sorry!" This seems to come especially from Bryan's experience at Joyent. The way they describe "Oxide value" is being able to own as much of the stack as possible so that they can debug things down to the lowest level possible.

> Nothing in Blindsight says "there can't be a God"

I feel this is missing the point that OP is trying to get at. I wish I could cite Blindsight more accurately, but it's been a long time since I read it.

I don't think OP is suggesting the characters in the book sit down and reminisce about the time scientists proved there's no god after all, ha ha weren't all those believers silly all those years.

But Blindsight I found to be deeply nihilistic about the place of "humans as we know them". Whereas religions affirm that humans as we know them are somehow cosmically special due to their relationship with God, Blindsight posits they're just a weird blip of biology destined to be outcompeted.

I wouldn't necessarily call it atheistic so much as ahumanistic. It seems to focus on survival, power and force, and feels like it denies, rather than affirms, any sense of our own worth.

I may be overstating the case; again it's been a long time since I read it. I remember having, as some other posters have a said, quite a depressed feeling after having finished it. I didn't hate the book, but I didn't feel like I'd learned anything that would improve my life. Other than having had the chance to grapple with challenging ideas- which is better than having read something with no intellectual stimulus!

Yes, he's a prolific writer and cultural observer. I think he's good at identifying and naming concepts which I find somewhat useful, like the one I mentioned.

> For instance, a future job site worker equipped with smart glasses and an earpiece might receive continuous direction via an AI interpreting and providing feedback on their actions.

I see this happening very soon- already, in some cases, for some applications. And I worry that it will be another way for capital to deskill and disempower labour; see Doctorow's "chickenized reverse centaurs" with Amazon drivers and warehouse workers being the classic example.

What was the transition like? How did this article's description of union electrical work as like the "McKinsey partner" of the trades world strike you?

I agree with your framing of agentic AI systems as similar to corporations. That's interesting!

But what in TFA suggested to you a specific and wrong-headed regulatory approach? As I read it, the post calls to investigate the AI labs, to increase transparency of their operations. The linked NYT piece says:

> Before any intervention, Congress should lead a public fact-finding mission that reveals the unvarnished details of A.I. development.

Isn't this the first step to holding any corporation accountable, whether it is made of people or software?

> There are some channels on Youtube where someone walks through neighborhoods in Tokyo

Shout out to Rambalac whose videos I enjoy a lot for this. No words, no music, just walking. An occasional vending machine stop.

> and/or whatever the (in this case Australian) equivalent of a CIA assassination squad is*, sent after you

Sounds like a great concept for the next season of Danger 5.

Minimalist phone, maximalist website.

Oh yeah totally. I guess seeing your SQL queries embedded inside the board doc, I immediately wondered if it would make sense to embed a GGSQL program the same way, but in the charts section instead of the queries section!

I guess at the moment GGSQL itself wants to be directly connected to the source database, whereas you've got a split between queries and the charts which display the results.

It would be really cool to have SQL "all the way down" to the chart definitions as well :)

What do you think about GGSQL, or is that not entirely aligned with what you want? They also produce vega-lite JSON as output.

https://ggsql.org/