cadamsau

My comment was about how tolerance affects the way homelessness is viewed and addressed, nothing more nothing less.

Also in UK and AU English (and likely others) it’s the word for trashcan.

Pray your office is free of folks who think bin puns are funny & original.

Sounds great!

My hope is 2030 sees more organizations that are each smaller. It implies more services provided by the economy thanks to greater specialization. For example today, SaaS tools exist for so many things you’d have had to hire for in the past. Thanks to SaaS you can get started really easily - take payments, get a storefront, send email newsletters..

If the trend of increasing specialization crosses over and starts to include services that require intelligence, it will be a net good on a scale we can’t imagine.

The megaliths of today also have more market power than optimal, hence so much anti-consumer shenanigans.

Bring on the layoffs!

Small suggestion - you can edit U.S. to US and gain 2 characters.

Soooo it does what it’s supposed to?

What I’ve observed is it becomes part of the job of the office admin person. So not zero headcount maybe 0.1 or 0.2 but that’s pretty good if the SaaS bill is another 0.2 headcount.

For everyone who thinks H-1B and its ilk make it possible to hire foreign workers at below what citizens are paid - have a look on your favorite search engine for:

“Prevailing wage condition”

It’s a requirement that’s part of the Labor Condition Application wherein based on the location you’ll work (the “Metropolitan Service Area” or MSA) to be granted a visa your employer must prove they’ll pay you above publicly available and published minimum wages for each job title.

These wages are public. If you have a problem with what they’re permitted to pay H-1B workers, the published prevailing wages are what you have a problem with. Spoiler though: they’re actually pretty accurate.

Here is one example of how hard it is to underpay workers on visas: during the pandemic, workers on visas were not legally allowed to be furloughed, because they would run the risk of not meeting the prevailing wage that year, putting the employer out of compliance with the LCA and subject to fines in the event of an audit. So what happened in practice was negotiated unpaid leave or in most cases the US gov covered wages via programs like PPP.

Now this is all out the window if the published prevailing wage for a given occupation is too low or the employer somehow sneaks one by the consular officials approving petitions - by selecting a title too junior for the applicant’s years of experience, for example. There will always be anecdata that makes this seem like a huge problem so be wary because one story does not a trend make. As mentioned above, by and large the prevailing wages are pretty on point with what American citizens are paid.

The reality of the way this system works is it’s WAY more technical than fearmongerers would have you believe. Visa holders are very much NOT undercutting anyone and the H-1B is not a completely broken system - even though the lottery and the fraudulent applications cause hell for applicants and employers it does basically do what it’s intended to do. So besides the exploitative situation these changes seem to proactively address, it mostly works and alongside the O-1 and a few other visa categories, has played a key role in the US’ ongoing supremacy in AI and many other industries.

Source: Australian citizen spent over 6 years working in SF on an E-3 visa which is very similar to H-1B.

You’re only thinking in the first order.

In reality no-win scenarios are avoided by having so many cameras and such an overwhelming amount of data that Waymo would win in court instead.

This is good for all as the incentive for maximum data collection as CYA also means maximum velocity of product improvement.

Claude and pasting code back and forth for simple things. But I’d like to try Claude with a few MCP servers so it can directly modify a repo.

But lately Cursor. It’s just so effortless.

Seeing something once doesn’t make it a real phenomenon.

Besides, if that is common, Waymo will have data on it and if it’s common enough, they’ll fix it.

I’ve sat in Waymos on 2 occasions, each on a separate ride, in which the car avoided collisions that would have happened due to bad human drivers. Your fear is worth examining closely as it’ll hold you back from enjoying this amazing technology.

My non physical desk job leaves me with a fresh body after my very much non 12 hour workday.

Because I have the physical energy to, I take that body to the ocean for a swim, to the park for a cycle, and to anywhere with a nice view for a run. Depending on the day. Oh and the gym is a pleasant 12 minute walk from home, so I do that a few times a week as well.

Sorry to burst your bubble OP.

One hundred per cent this.

LLM pair programming is unbelievably fun, satisfying, and productive. Why type out the code when you can instead watch it being typed while thinking of and typing out/speaking the next thing you want.

For those who enjoy typing, you could try to get a job dictating letters for lawyers, but something tells me that’s on the way out too.

Careful not to shout too loud about it or they’ll start training for it!

Latest Claude does a suspiciously good job…

> jotted down

As an MCP fan, am curious! Care to share. Link maybe? Or even just a comment would be great!

Maybe RIF-balancing is a better term.

Fascinating that 2-3 probes per request is a sweet spot, intuitively it seems like a lot of overhead.

Whenever a “good enough” prototype can be created just from a prompt, it becomes possible to scrap it every time it’s not quite right and regenerate, instead of editing the code. Working this way, the prompt is the “source code” and the code a kind of compiled artifact.

Currently the generated prototype usually needs tweaks and that’s if it even works. But when it does work, it’s like the model is reading your mind.

In the future as models improve at coding, they will anticipate the tweaks that make sense, less of the prompt will need to be specified & there’ll be less polish work after you get the generated artifact, and you can work at an even higher level of abstraction and thought. Domain experts can create even bigger, cooler things without spending’s years getting software engineering skills.

Assemblers and compilers came along very early in our industry’s history. If you run the thought experiment that that’s where we are at with prompted software creation, it will be a wild and exciting future. More people creating more stuff means a tremendous amount of amazing creations to enjoy.

Thankyou @gd3kr this is a really handy tool! Will use it next time I need 3D models.

Just made a porcelain elephant to test it, it came out so cute!

In terms of the discussion detailed in the article let’s wait and see. It has to be a vastly superior alternative to make adding it worthwhile.

On the more fun side, when I read the headline I instantly thought of making a side project that lets you watch a bunch of language models in a chat room, infinitely bikeshedding among themselves about the merits of this decision.

Kind of like https://degenerative-sitcom.online/ but for bikeshedding :)

I don’t think it’s fair to look at artefacts of a past era and compare them to today.

The only things preserved are the ones deemed worthy of preservation. That leads to a skewed perspective.

While I agree with what you’re saying in the first order, what would actually happen is pretty predictable - an even more extreme version of what we’re seeing with scams, lost wallets, pump & dump schemes, fake ICOs and people taking advantage of each other with impunity.

Cryptos are simple but cryptoCURRENCY - where the rubber hits the road and the numbers on the internet become useful - is anything but.

Missing from the complaint is “too much” churn. But then it’s boring and obvious. Too much employee churn is lethal to any company.

Not a huge insight.

But if we could somehow arrive at the answer to how much is too much, it’d be something that can be optimized for. And we’d live in a fascinatingly different world.

It’s simpler to say no than to answer the next questions of “how much better”. Because no matter the answer, I promise someone somewhere won’t like it.

“I writ a word salad and it is very scary, also please send me lots of monies kthx”

What a place the internet is.

True, but that’s made possible by the aggregator (the publisher who’s printed the magazine)

This dream of eliminating that middle-person is still a dream.

Justice is subjective. What is a crime is subjective.

Hence we have laws - an eternal, never perfect project to find an agreed definition of justice.

Abandoning centuries of precedent of law happens in some places from time to time and they’re not places you’d live by choice.

Nature abhors a vacuum. If not this dude, would’ve been someone else.

As much as I hate to point it out, it’s not so simple.

He had a legally enforceable mandate to maximize shareholder value. There’s some wiggle room in how that’s accomplished, but not as much as it seems from the outside.

That does make him, much as I hate to say it, a small cog.

Put a David Mayer question on your company’s job application form and boom!

You just filtered out every lazy job applicant who uses AI tools without proofreading their output.

There’s thousands of different tax regimes in the US - dozens to hundreds in every state, thanks to city taxes and the like. All businesses would need to advertise the right price everywhere. Doesn’t scale.

It’d be neat to see a campaign to popularize a term like “final price” or “all-in price”. It’d encourage the creation of products that always show the right price even as taxes change. And consumers would be empowered.

Drivers will always want better pay & conditions. They lack the power to make demands, and bleeding out all over the media won't change a damn thing. They need to unionize.