Looks like this was timed to coincide with Google adding search grounding data to Gemini API: https://news.ycombinator.com/item?id=42008834 // https://developers.googleblog.com/en/gemini-api-and-ai-studi...

They are taking a page out of Microsoft’s strategy of clouding out all sunlight.

My first memory of something like this was the rushed Google Wave announcement occurring on the same day as a Bing rebranding launch (if memory serves, the domain registration timestamps for the Google domains were also highly coincidental but I've long since forgotten with what)

I know there is a HN guideline about not commenting on stuff like this, but in this case the text was so large it prevented me from wanting to read the article at all. Just uncomfortably large.

I have a 49" ultrawide and if I maximise the window, it is borderline unreadable.

https://imgur.com/DGWQDpI

What’s the technical difference in how this works vs. previous face swapping tech (like Snapchat filters)?

Yes, I played around with Skyvern and it works really well!

https://www.skyvern.com/

https://github.com/Skyvern-AI/Skyvern

I created a bit.ly link that points to http://yourpiholehost/admin/api.php?disable=300&auth=api_key

Bookmark it for myself and other people in the house and then turn it off for five mins whenever there is an issue.

Would be fun to hook this up to one of those IoT Amazon buttons.

Home Assistant can definitely do this. I just now successfully got my HA connected to an Aqara Zigbee button (which, hilariously, only briefly functioned when I was using it with Aqara's garbage hub).

Or say “hey google disable ad blocking” and triggering it. Then google learning of this and doing it all the time behind your back

I think Paypal's micropayment option[1] doesn't have a ton of competition, but at 4.99% + 9 cents you will do a lot better than typical stripe rate.

[1] https://www.paypal.com/us/webapps/mpp/merchant-fees

> Usually about 75 per cent of all gym memberships are taken out in the month of January.

That's a pretty wild claim to include without a source...

I googled it and found a source that said 12% of signups in January compared to average 8% the rest of the year which sounds a lot more realistic. [1]

[1] https://www.ihrsa.org/improve-your-club/how-to-keep-resoluti...

Could it mean 75% of a given year’s memberships are active in January? Probably an overly generous interpretation…

this article is from 2015, maybe people wised up. I remember I once joined a gym with a 3 month prepay and never went even once. After the 3 months were up they called me asking if I want to renew lol. Safe to say I never since payed for more than a month

While every year comes with a January 75% sounds like a fish story.

If one was to be loose with the application of statistics, one could say that January has 50% more signups than the average for other months.

This could reasonably be Chinese-whispered to become that 75% statistic.

I am not following how playing in a band for money has the same business plan as this app

I was being a bit tongue in cheek. However this was relatable, especially to the plans of the original musicians:

"There’s no clear monetization strategy — yet. [...] they’re focusing on growing [the band] and developing complementary new [songs|social media].""

The paper[1] says this in the conclusion:

> [Llama 2] models have demonstrated their competitiveness with existing open-source chat models, as well as competency that is equivalent to some proprietary models on evaluation sets we examined, although they still lag behind other models like GPT-4.

It also seems like they used GPT-4 to measure the quality of responses which says something as well.

[1] https://ai.meta.com/research/publications/llama-2-open-found...