jen729w

Can I build an app against this? Do you commit to not changing it again tomorrow?

Which part of my backyard plumber story do you think will be automated?

Either the plumber is replaced, or they aren't. I contest that they aren't.

Wikipedia?

What is it that you are doing with tomatoes today? Store that locally. Sounds like you're cooking. I put those here, then I can find them again really easily.

   10-19 Life admin/
    └── 12 Where I live & how I get around/
       └── 12.44 Meals & recipes

It's almost like sitting at a keyboard and typing is different to driving to a site, lugging a drain snake off the back of a flatbed, asking the owner where their outside tap is, cutting the dead bush away so you can reach that tap, finding the drain, snaking the thing down the drain, monitoring it all, not making a mess in the customer's back yard, packing it all up, and driving away.

I mean they're basically the same job.

I think they won't replace most household plumbers.

You've never contorted yourself under a sink trying to undo a pipe, have you?

People with wrenches aren't being replaced any time in the next 50 years.

The solution to this problem is as old as computers: it's a folder. Just use folders.

`cd` to a folder. Launch `claude`. Do your work. Save scripts and documentation in that folder. `/resume` previous conversations from that folder.

That's it. That's the trick.

Now, having very static, very well-defined folders helps a lot. I'm Johnny.Decimal so I have numbered folders for everything I do. So my process when I want to use my 'process a travel booking from my email to my calendar' script is:

- `jd tripsy`

  - The folder name includes 'tripsy' and this is how I remember it.

  - `jd` just parses my limited tree and `cd`s me to a folder.

  - `jd 21.15` gets me there by number if preferred.

- `claude`

- Say 'hey Claude, there's a new email in my inbox please'.

- Done.

Obvious AI copy on the landing page says to me that you don't care about the product.

I know the thing was written by an AI. That's cool: I do the same myself. They're better at code than we are.

But the product is the thing I'm buying. It's the reason I would consider giving you $50/month rather than getting Claude to knock me up my own version. If I don't think you care about the product enough to explain it in your own words, why on earth would I ever give you money?

(Not a personal attack. I feel this way about every new site and it seems like you're interested in feedback. I wish you all the best.)

Which is why the only people authorised to declare an incident are your incident managers. They follow the published guidelines which dictate what's a Sev 2 and what isn't.

A manager of a service may not declare their own service a Sev 2. If they can, your incident management process is broken.

Fix your relative understanding of how much the average book makes.

Haven't you been served the videos on LinkedIn where people have a thing strapped around their mouth that they can talk into? Apparently it traps the sound so your coworkers can't hear you.

Of course they're mercilessly mocking you instead, but hey.

I heard (via Steve Novella of the Skeptic's Guide) that there was similarly light evidence to support flossing, but mostly because nobody's bothered doing the studies.

The reason for that? It's obvious that flossing is going to help. Obviously brushing away cruft is better than not. So why waste the money on a boring study?

I can see the same for the electric toothbrush argument. Of course it's going to be at least as good if not incrementally better than me see-sawing my way around my mouth.

> the single biggest thing that makes a difference is brushing for the full 2 minutes, and electric toothbrushes can help with that by having a built-in timer

As someone who had toothbrushes for the longest time with a single 2 minute timer, and now has one with 4× 30s timers, I can say that the difference is radical.

I don't understand how I got bored in 2 minutes but I don't get bored in 4×30s, but I do not. Every quadrant gets the full time, and twice a day I do it for the full 2 minutes, no exceptions. That wasn't the case before.

1Password offers a lot more functionality than, say, Apple Passwords.

Apple Passwords will store a user/pass combo for a site. That's it, feature list over.

1Password will let you configure how those things autofill, it provides an `op` CLI that you can use with service account tokens, you can store stuff like photos of your passports and licences, it understands 'Sign in with GitHub', it has per-user access per vault for shared accounts, you can add any number of extra fields to each entry, it integrates directly with Claude, it monitors your passwords against Haveibeenpwned, it'll tell you where you could be using 2FA and aren't … and the list goes on.

> The problem is not smartphones per se. … They let you access information and entertainment wherever you are.

This is the exact problem with smartphones.

Discussed here last month. https://news.ycombinator.com/item?id=49005973

The difference being that the story he's telling is "you'll make money" and he's usually right.

I was thinking about this as I walked earlier.

People who "make predictions" are like psychics. Unless those predications are uncannily good – in which case we call that person an analyst – we can assume that the person is either wilfully lying, or that they're stupid.

Ed's predictions have so far failed to come true. I don't think he's stupid (but I might be wrong). That leaves me feeling like he's the psychic who knows exactly what they're doing. They're telling a gullible audience a story they want to hear because it's a nice way to earn a living.

Comparing him to the CEOs of companies who produce incredibly useful products used by millions is facile nonsense.

It seems you missed this part of my very short comment.

> That's fine. Do what you do.

But this doesn't excuse Ed making a living from awful predictions, surely?

If you realise you're bad at something it seems immoral to keep charging people for it.

This guy – whose RSS feed I subscribe to – makes a living [0] by being anti-AI.

That's fine. Do what you do. But don't read this article as any sort of science. It's massively opinionated rage-bait.

[0]: https://www.patreon.com/davidgerard

Mind blown. Thanks. I'm British and I've had a debit card since I was 16 which is a loooooong time ago now, let me tell you.

I run an online store selling e-stuff. I offer Stripe by default, and a 'use PayPal' link on the side.

It accounts for about 20% of my sales. I was about to write 'overwhelmingly Europeans' and I thought I better check. Nope.

    Row Labels      Count of Country
    ----------      ----------------
    United States   54
    Germany         12
    Australia       10
    United Kingdom  7
    Netherlands     7

But you have a Visa/Mastercard, surely? A thing with 16 digits that you can plug into the Stripe checkout screen?

I have, but never use, a 'credit card'. The thing I use every day is a 'debit card', where the money comes straight from my account. But it still has 16 digits and Visa written on the front.

> would be 100ms slower, who cares

It's almost like you didn't read the article.

Am in Japan now and can confirm: you got a broken one. It's magically fast, everywhere. I prefer it even at the 7-eleven. Why wait the interminable 2 seconds for your Visa 'cardo touchy' to work when Suica is literally instant.

Reading this article it makes sense why it feels too fast to be real, especially when using a powered phone. The things are already communicating as my hand is moving towards the reader. By the time it gets there, it's happened.

Stunningly good.

I'm Johnny.Decimal. [0]

I use an Obsidian vault and my system. "It's at 23.16", I tell Claude. From that it can instantly find my notes and files.

No extra software needed. Just some basic structure. Claude loves it. [1]

Oh yeah, and to keep conversation context contained so you can /resume and so on, just

    $ cd 23.16
    $ claude

[0]: https://johnnydecimal.com

[1]: https://www.youtube.com/watch?v=mZAT0Ft--wE

This is the way. Mine looks like `ctx: 0k / Fable 5 high`. When ctx ⪆ 200k, time to start a new session.

```statusline-command.sh

    #!/bin/sh
    input=$(cat)

    model=$(echo "$input" | jq -r '.model.display_name // empty' | sed 's/ context)/)/')
    effort=$(echo "$input" | jq -r '.effort.level // empty')
    used=$(echo "$input" | jq -r '.context_window.total_input_tokens // empty')
    out=""

    [ -n "$used" ] && out="ctx: $(( used / 1000 ))k"
    [ -n "$model" ] && out="$out / $model"
    [ -n "$effort" ] && out="$out $effort"

    # bright yellow + bold
    printf '\033[1;93m%s\033[0m' "$out"

```

You're at work. You need to produce an artefact – a project proposal, say. 100-odd lines of Gantt, costs, that your BD team will send to the client.

You get Copilot to do this for you.

Later at the review meeting, errors are found. A bunch of stuff was missed, the costs were off.

Whose fault is it? Who looks stupid? If it's bad enough, who gets fired?

You'd rather the banal corporate beige in every other handbook? At least this reads like someone means it.

I 100% agree with the sentiment. If I had staff, it's the benchmark I'd apply (even if I did word it a touch differently).

Here, let me paste you, verbatim, an ex-colleague's LinkedIn description. This might be more up your alley.

    Over 25 years’ experience employing strong business and architectural skills in orchestrating and delivering innovative solutions that provide quantifiable business value to medium and large enterprises.

    I have strong technical leadership, a broad background in technology and a deep understanding of business drivers and challenges across multiple sectors which enables me to assist organisations in removing technical blockers and delivering business outcomes. Employing enterprise architecture principles, I am able to balance strategic thinking with pragmatic operational delivery to ensure life-cycle optimisation, organisational change and IT governance requirements are addressed.

> I wonder if Opus could be prompted to respond in simpler prose via agents.md

God knows I've tried. I've got a variant of the ASD-STE100 trick which does the job, mostly, at the start … but get to about 100k of context and it goes out the window.

The model's personality is too strong for simple suggestion, alas.