indymike

I chuckle a little bit on this one - SD cards are supposed to be "Secure Digital" cards... but that security was always about making it secure for media publishers to distribute DRMed content and not really about consumer security.

The Claude subscription does not include Claude API usage.

Claude behaves in Claude Code somewhat differently vs other coding harnesses (sometimes it's a remarkable improvement) connected directly via api. Also, it cost more than the subscription. There's quite a bit of baked in tooling and assumptions in Claude code that may be good for you, or unhelpful based on what you are doing.

> That means Apple will begin imposing all sorts of UI dreck on all of us, whether we want to use our screens as touch surfaces or not.

Since we can run ios apps on mac now, touch makes sense as the mouse experience is very meh for some apps.

Also, I would have loved a touch screen mac when my job involved building mobile apps.

Finally, I suspect apple will execute UI well on touch, they get to learn from everyone's early mistakes... Vista especially.

I'd like it a lot if Google would enable antigravity to use the latest greatest models faster, and also give me more choices. I really like the product, but I don't feel very ultra even though I'm paying for it.

I was there :-) My business partner at pycon observed that Docker looked like a complicated way to fix easy_install (at the time it looked like to official fix would be to adopt pip). I thought it had some potential :-) but definitely didn't expect it to take over the way it did.

Back in the day, the 8087 and 80287 were supported by quite a bit of software (spreadsheets, statistics, etc...). First dev job I had (was 15 at the time), I had to recalc our accountants 1-2-3 models because I had a 80287 in my AT clone... and that resulted in a "hell yes" a year later when I asked for a 386 + 387. I put on the write up on why I needed it was a) it would compile faster and when needed I b) could recalc financial models faster. That model took maybe 5 minutes with the 287 and maybe 40 sec on the 386+387 setup... I suspect the generous 4mb of RAM was more important... I never really understood why people wouldn't shell out the extra couple hundred bucks on the math coprocessors, they really sped things up provided that software supported them.

There's a lot that I think you missed in your reading:

If you read the preambles to each chapter and the glossary in the back of Dune, you'll find that the Butlerian Jihad was a violent war that lasted 100 years.

If you read the preambles to the chapters in Dune, you'll find the Butlerian Jihad's problems were explained rather succinctly: people turned over their thinking to machines and were ruled by the master of the machines, and thinking machines created poverty for workers.

The entire society, the great houses, the spacing guild, the Orange Catholics, Bene Geserit, and Mentats, all who emerged to replace functions of thinking machine and all three organizations had very real motivations to ensure that thinking machines would never rose to prominence again (and the Bene Geserit were quite surgically violet when needed) and it was clear that doing for an individual was a capital crime, and for a ruling house, would invite at the least the Spacing Guild cutting your world off, and transporting armies to destroy you at a deep, deep discount. The entirety of society was aligned against thinking machines, and this is what made Dune so different, and for me, fascinating as I grew up.

Spend your time on things that matter to you, and especially with people that you like and people that you love. All we really have is our memories anyway.

I think there was a lot of resentment here because of how easily it appeared she was raising money and how quickly the company grew at first without having a working product and a murmur from experts that their miniaturization play may not be able to work at all.

I don't want that world. I want a way to make that world not work.

I wasn't talking about cars alone.

I'd like to have a radio kill switch be required on all devices. The reason is that eventually, things that emit RF energy could be targeted by drones. That scares me.

His death was far too early. It's hard to explain how much influence Cringely had over the years, how many minds were influenced by him - and how much his humor has been assimilated into the tech industry. He will be missed.

> Yeah, I’m surprised at the negativity here.

I get design/product people sometime that get all caught up in "we need to simplify" or there's not enough whitespace. Every time, the design/product person is not a user of the product, and the result is you ship their improvements, get blistering angry feedback (last time I even got the most prolific user and recommeder of our product calling me on my cell at 11pm) and revert.

> their version of tech is lots of padding, whitespace, diffused colouring and not information density and consistency

Exactly this. I think is speaks more to how frequently the designer uses the product versus actual users, and they don't understand the difference between UX for casual/infrequent users and people who spend all day in the product.

> it bothers OP less that they take a 15% tax than the fact that google provides terrible support for their own play store.

The terrible support is really deeper than support. You literally stopped doing business here, because the lack of a feedback channel that allows Google to improve their business (i.e. realize they need a way to quickly correct stupid automated decisions) been Google's worst enemy and best friend. By ignoring support & feedback, the friction for internal decision making is low (no pesky customer demands, the product manager is king, management gets their way), but it leads to I think a lot of Google's products failing because they do not really understand their users, and the why behind the what they use.

I'm not sure being a monopoly alone explains Google's behavior here, but it could be as simple as that: who cares about one small customer... it's not like they have a viable alternative.

The biggest problem with AI isn't being able to do things. You can do them. The question is "should you?" I had a peer dev this week put in a commit that was 1.2mb (this despite an AGENTS.md saying not to do exactly this) of plan documents and ~20 lines of code. The conversation on the PR review was probably one of the best learning experiences for both of us. It was all about the ramifications of a book's worth of AI pontificating about what it would do and what succeeded and what failed. And we were able to validate that that plan document had huge side effects as we prompted with it in the repo. As we read the document we realized that it was just 1.2MB of smart sounding gibberish and maybe two bullet points of signal.

Forcing customers to talk to sales is actually a failure mode.

I honestly never really thought much about this, but there is a level of awesome to it especially when it makes zero sense to have 50 words of text accompanied 6.5MB of markup, scripts and styles. The experiment went well... and it worked extremely well.

MCP is one of those things that is "too good enough".

Best idea in the article, and I'm going to try it out later today:

We should start to standardize how agents use HTTP APIs directly. For example, agent clients could attach headers to identify themselves as agents, and servers could automatically send them response data as Markdown or text instead of HTML or verbose JSON.

I'm a little confused: LLMs were invented in 2018.

I find it even harder to trust elected leaders from any party. At least Sam and Dario are aligned with a value set that is understood and clear, whereas political leaders values changes as do the polls their livelihood depends on changes.

> It can't be discoverable if it doesn't exist ...

This cuts both ways. I've seen plenty of litigation go south because one side had evidence and the other side had nothing because they deleted/shredded/lost the proof.

This is accurate.

> Even if it was a mistake or "mistake", the part where the agent can exploit things across multiple levels like that, isn't just marketing.

When we say "safety" people do not think we are protecting them from accidental automated crime at scale being committed on their behalf.

Or worse, take money to steer you to a specific airline.

Just disable wifi. Occasionally factory reset, hook up to wifi for firmware update, then disable again.

> the only option is to cave

Or sue, or organize and protest, or win the next election. Not every government problem warrants revolution, and not needing to revolt is not a failing of the people.

> Americans/Westerners (left, right, center) are way too comfortable for a revolution

A better read is that the situation for most Americans and "Westerners" whatever that means, is that things are good enough that the people do not see value in revolution, or they are able to adjust the government's course enough through elections, redress, courts and other political feedback mechanisms.

> This is unsurprising to me, since people are often quite resistant to learning.

This is an attitude I had when I was 19 and had spent a bunch of time and money going to school. I had a life event that made me leave school and take on a huge debt that had to be paid. Most people learn - they just learn in context of their necessity and their own passions. For other things, they lean on the community around them (doctors, lawyers, accountants, software engineers etc...)

> AI seems to be a good tool for people who are already good at learning and critical thinking to use to accelerate their learning.

I'm watching the huddled masses start to figure it out. AI is going to be to arrogant intellectualism what the gun was to the knight. Now anyone who can operate a mobile phone (and literacy is not required) has access to a shocking level of expertise and contextualized knowledge. And as AI improves this will be more and more the case.

> constantly missing words due to a lacking vocabulary

First prompts of non developer employees were "what does ___ mean?". Real time vocabulary is a thing now.

> “well, maybe literacy is actually not so important,” is so utterly ridiculous that I hope it’s satire and I’m just missing the joke.

It's not a joke. AI is going to reshape society. It's just started. Back to the gun analogy - the knights are just now starting to realize that proficiency with a sword and 70kg of armor are no match for a peasant with a pistol. Likewise, I suspect that Joe six-pack with AI in his pocket can do knowledge work at a "way good enough" level. I don't want to accept it, but that seems to be reality.

> It kind of seems like an insane TLD structure to begin with, right?

It's been around for years. I seem to remember this issue coming up around 2001 where originally .name was for third level registration (i.e. john.doe.name) and changed to second level it a few years later and caused some problems... https://publicsuffix.org/ talks about it in light of architectural limitations of domain names.

> can Joe set a cookie on all of .smith.name?

That can happen. I seem to remember ancient browsers made it so .name (and other non-generic TLDs) required three periods. I think country code domains and new generic TLDS caused the browsers to change it.

It's pretty screwed up, but a lot of the people with .name domains have had them for a very long time. Sad to see them all lose their identity online that way.