elefantastisch

Great.

What I really want is mandatory opt-in to be able to contact me.

Unknown numbers get to send a single standard length SMS, carrier embeds the full legal name and, for businesses, legally registered mailing address, and I can choose whether to accept. If I don't, no other contact from that number is connected to my phone. Ever.

I can also report the message as unsolicited and every company's unsolicited contact stats are made available to regulators.

Grid-powered public transit replacing 80% of trips by personal vehicle.

Every year at Christmas, I get a card from a family member with a short, personal hand-written note and a 1-page, generic printed update on how their year has been, usually with some pictures.

This is what I want from a social network.

I want to be able to keep up some regular connection with people I don't see on a regular basis. I want us to remember each other, know about major life events, and have a convenient way to reconnect more personally when that makes sense.

For people I see (or want to talk to) regularly, I'll just send messages or group messages.

For more of the topic-centered type of internet community I may want in my life... well HN already does that perfectly.

Thank you, kgr, for writing and posting this. I find this a much more digestible summary of the topic than the other recent posting on Statistical Process Control.

This article seems like a good reminder to know what your fundamentals are and to be deliberate about picking them.

But I'm not convinced that this advice can be used to pick your fundamentals, your core business model.

Amazon opted for more choices and lower prices.

Barnes & Noble stuck with faster solutions to problems (get your book today not after a week of shipping as was common when Amazon started), deeper human interactions (you can go into the store and ask someone for recommendations, meet authors, etc.), and increased confidence/trust (you know what you're getting because you can hold it in your hand and read it before you buy it).

Low-cost carriers have moved in on traditional airlines because while it's true that people will never stop caring about added comfort, it turns out they care about lower prices much more.

But Apple built a staggering market cap relying on the assumption that while people care about lower price, they care about great control of your time (just works) and higher social status more.

So it seems like this article provides a good framework for thinking about your business, but doesn't give any answers as to the actual strategy you should use.

Or am I missing something?

Right, this is a values problem, not a technical problem.

Who should get to see Taylor Swift? How much should they pay? How should profits be distributed?

These are philosophical questions. If you assume a particular set of answers to these questions, designing the correct sales process is not difficult. But everyone has a different set of answers.

Judging by the majority opinion in this thread, it seems pretty clear GitHub could have asked and gotten enough people to opt-in to have no problem training their model. They probably would have been thrilled to do it and proud of being included in the training data.

But the preference of the majority does not override the conditions placed by people who prefer not to participate.

We have an even better example of mistaking the numbers for the whole picture. Becoming famous for winning obscene amounts of money greatly increases your chance of being harassed, murdered, or driven to mental breakdown. The expected value of a losing ticket (a chance to fantasize, enjoyment, etc.) is much higher than the expected value of a winning ticket (a lifetime of being hounded for money in increasingly desperate and violent ways).

Definitely this. I have been an Android user since my first smartphone. I've been a Gmail user much longer than that. I still to this day don't have the slightest idea what the distinction between Hangouts, Chat, and Messages (and whatever else I'm forgetting... Wave?) even is. Sometimes my new phone has an app called Chat, sometimes it's Messages, one time it was Hangouts... they all seem to be the same app to me.

So I just use WhatsApp.

Idagio is a great alternative for classical.

Maybe I'm not the only one who resubscribes just for Great British Bake Off...

My understanding is yes, and honestly, this was my biggest sticking point with Obsidian. I'm happy to pay for software. I pay for Dynalist pro (same team for anyone who didn't know), but the separate commercial license on top of paying for sync just felt like a lot. Especially with commercial use being defined so broadly.

Fourteen days to trial just isn't enough to see whether it would be worth the cost to use for my own personal use at work. The Dynalist freemium model where I could use a smaller feature set for a longer time and then decide it was worth upgrading was just much easier.

Erica, would you consider a discount on add-on services for commercial subscribers? A 25% discount would put the price a bit closer to Dynalist pro and make it feel like I was actually getting something for the commercial subscription other than just permission to use it for work. Or maybe a discount for existing Dynalist pro subscribers?

I really love the work you all are doing, here and on Dynalist, just a bit hung up on the cost here.