My wife is a sign language interpreter, and does a lot of remote work, especially since covid.

To handle a meeting on Teams, sign language interpreters need to pin two video streams - the current speaker, and the deaf client(s).

It is essentially impossible to do this in Teams - they routinely open up a separate Zoom session for interpretation.

You'd think the inability to do this would be an ADA violation...

The biggest complete failure of a US software project. See here.

1982 - 1994. Twelve years of effort, pure waterfall design/test/implement, everything was believed on schedule until the last minute.

What you really want is a dog-like robot with package grippers on its back, and one arm for doorknobs and elevator buttons - something like a Boston Dynamics bot.

Four legs - stable without balancing, so longer battery life.
Low profile - delivery vans could have more than one, in dog-house slots
Can't be mistaken for a human - give it a few cute dog-like mannerisms

I'm a long time NASA contractor. The vibe I got from my management and agency communications was that we could have at least lived with Isaacman - he at least believed NASA did valuable things, things worth preserving and defending. Maybe that's what got him rejected, ultimately.

Now we're back to the drawing board again.

Any bets that finally dropping into flow state will look like mental stress?

Just what we need - a device to detect maximal productivity and interrupt it.

The problem is, by and large jobs aren't taken - they're given by business owners and their MBA-driven middle managers.

If you're a business owner and you have a task that can be done in two ways:

* Human + AI = more productive human, human level salary
* All AI = OK performance, little or no salary

You're going to choose the second one until it becomes clear that "OK performance" is actively alienating your customers.

I predict many spectacular AI failures while this lesson is learned and relearned.

if you're looking for long-term utility and support. I made that mistake with the iPad - the first iPad lost support quickly compared to iPad 2 and following.

Apple's design and testing people are good, but you don't really know what people really want until you have feedback from several thousand users.

To the people who bought Vision Pro - thanks for your input on Vision Pro 2, which we won't see until Apple can solve the obvious Vision Pro problems of weight, power, and cost.

Apple can by default read the contents of your iPhone backups.
I could see them running the moral equivalent of an anti-virus scan over those backups.

You can encrypt your backups, but then you had better not lose your backup keys or you're REALLY screwed.

vast price disparities ($80 per ton for forest projects versus $1,000 for direct air capture)

Plant fast-growing trees on West Virginia mountains
Harvest them
Bury them in local abandoned coal mines
Repeat every 5-10 years

Jobs to replace lost coal mining jobs.
New uses for old coal mines (well, probably not the mountain top removal mines...).

I'm not sure you could grow and bury two tons of wood for $80, but you could definitely do it for $1000.
All that's needed for this to work is a guaranteed price for real not-faking-it carbon sequestration.

This opinion article (Washington Post, gift link so no paywall) says there's 20-year-old Supreme Court case law that says the Government can't treat people differently under the law just because they don't like some of them.

Those cases were under Roberts, and were unanimous where it counted.

Trump is repeatedly on the record saying he uses US law to go after his enemies. A Supreme Court with any spine at all could use this to shut down the worst of his behaviors.

it's logistics.

Apple probably had this kind of pipeline acceleration in the works ever since the election.

Anyone who is a real full-stack, non-vibe-coder developer will be in roughly the same position as COBOL developers today:

"Please don't retire, you're the only one who understands this code enough to modify it."

NASA is odd compared to most federal agencies - the NASA administrator reports directly to the President, not like, say, NOAA which is under the Commerce Department.

IF NOAA closed its DC area headquarters, the stuff it does would be absorbed by Commerce, and the executive chain of command would still work. If NASA closed its Washington DC headquarters, the current administrators would have literally nowhere to go.

Bureaucracy aside, they do perform the important function of deciding what functions are performed at which NASA centers. Without them, Congress would have to do it directly in legislation. Which is how we got some of the current absurdities like Houston Space Center, which takes control of manned launches after they clear the tower (LBJ got Houston as his fee for getting the man-on-the-moon legislation through Congress).

Given that this is Y Combinator we're talking about, all the prompts were probably of the form:

"A system like X, except for Y"

where X = {Uber, Grubhub, Facebook, ...}
and Y is a niche currently-unmonetized domain.

I will believe it when even one of those applications gets any traction in the real world, and doesn't get immediately crashed or owned.

But, do you think the current administration will do anything better?
Or will they just do something just as screwy and inefficient, but with different corporate winners and losers?

The big telcos have outplayed the federal government every time a "broadband for all" program has been passed.

The worst in recent memory was the time the big telcos were given hundreds of millions to roll out broadband for all - they lobbied to get the definition of "broadband" at the time changed so ADSL would meet it, spent just enough to put near-useless ADSL everywhere, and pocketed most of the $$ without actually improving rural internet.

Any bets that right now Starlink is preparing to game whatever new program is created? They won't even have to lobby - Musk will tell Trump and his minions exactly what to put in the bill to maximize Starlink revenue.