The idea is interesting but in looking at methods and github I feel the tooling leaves me wanting. I mean you are trusting LLMs here to establish, vet, and detect your various thresholds that were then used to train the classifier. I'd rather see this sort of thing done deterministically with actual code vs lossy human english prompts and a dependency on token spend to a single third party (who will probably pull the underlying model used in what a few short years probably) to replicate the results or try and use different training data.
Mother nature has no perspective. There are no improvements either, only things that are and are not. Everything is dependent on local context at the present point in time and framing whether it is "good" or "bad" (e.g. one could frame our intelligence as good or bad given how it resulted in anthropocentric climate change that may doom us in time, while 'dumber' species have existed for millions of years without wiping themselves out).
What works today may not work tomorrow. What didn't work yesterday may work today. Evolution is not linear. It tends to happen in bursts almost like a dam busting loose. A constant cycling between periods of genetic diversity (i.e. big resetting just happened, no winners taking all quite yet), and what is known as allele fixation (a winner emerged, oh god no...).
>What works today may not work tomorrow. What didn't work yesterday may work today.
All the more reason for variety. Which stops happening when only one set of genes is mixing. (Assuming that set of eternal genes would have any motivation to mix at all.) If you want a higher chance at producing the genes most resilient in a given context, aging and death are a necessary part of that process if you're stuck with sexual reproduction. You need motivation, and you need maximum mixing.
Only if it works. Evolution works like this with local fitness maxima and wells of stability favored by entropy. You have to luck out with the right combination of mutations to not only get out of that well but fall into one of lower entropy. Imagine the size of the search space required for something like this. This is why evolution is "slow" in species that live as long as us and quite rapid in things that grow logarithmically by the hour.
There seems to be a balance point with the number of Chesterson's fences. Too many and you are penned in or routed on one vector only. Too few and you are blind.
We have the ability to convert to becoming immortal. It actually happens to some of us. It sucks though. We call this cancer. Turns out the immortal cells become selfish. But you can scrape some off yourself, throw them in a petri dish, supplement growth media, and they will divide forever.
Flagship phones are so damn expensive for what you are getting. I just buy whatever is the current iphone SE which is usually around $300 for the top storage sku when I show up to buy it a couple years after release. At my current burn rate $1300 would be like 20 years spend on phones for me.
What stopped me from buying flagship phones is the repair costs for phones are far too expensive in relationship to the value of the phone. Results will vary, but waterproofing usually diminishes after a repair.
I am not the right market for flagship phones because I don't have thousands of dollars to put towards disposable income, but I would expect a premium phone to come with a premium warranty and repair service... Do they typically not?
Eh... get protection plan, or trade up every year for big discount on the next flagship. Or with AT&T plans you are getting a "free" phone fairly often if you don't intend to leave the carrier for a few years (most people wont). The parent comment you replied to makes some assumptions that you have to spend full price all at once every year which is false. Also, there's almost always a deep discount on Pixel phones around most holidays.
And if you truly are ultra budget $300 phone guy then sure this doesn't apply to you.
My current phone is definitely flagship for the time, ~$850, and I will be holding onto it as long as I can because flagships, with headphone jacks, available in the US, are something that just does not exist anymore, and I'm not spending $1500+ to import the cheapest Xperia only to have it barely work on my cell network.
That's also what I buy with Apple now. When I was using Android I'd buy a cheap "Moto G" for maybe $150 brand new. Never understood dropping a grand or more on a mobile phone.
Cameras of flagships are still significantly better than a $150 Moto G. If you are using a smartphone in place of a camera and like good-quality photos, it's worth the money. That said, flagships can last many years as well.
I update pretty frequently, but all the older devices get used by family members, etc.
The camera is probably the least important thing on the phone for me, but yeah if photos are important to you then that might rule out the cheaper tiers of phones.
Yeah, people have different needs. One of the nice things with GrapheneOS on Pixels is that you can even pick up the latest Pixel 10a for ~400 Euro (and older models even cheaper) and it will still have what Google calls a flagship chip, but more importantly all the nice security stuff like the Titan M2. For a mid-range phone it also has pretty ok cameras.
I hope that a few years down the line there will also be great midrange Motorola phones with GrapheneOS support. But they have to start somewhere.
I'm glad to hear I'm not the only one who thinks smartphone prices are just absolutely crazypants right now. I can't envision paying more for a phone than I would for a (gently used) workstation-class laptop.
Over decades of buying Android devices, I have yet to pay more than $300 for one. The one I'm using was actually free in exchange for buying a year of T-Mobile prepaid credits. No, it never was the fastest thing on the planet but it makes calls and runs the apps I want.
I keep my phone until the security updates stop. I buy top of the line phones because cheaper one suck. I tried a cheap phone and it used to lock up all the time. These phones are crappier than my obsolete top of the line phone. There has got to be a middle ground between 1K flagship phones and cheap sub-$200 phones, but I swear I can't find it.
I just keep buying phones with way more features than I desire just so that having too many apps open at once or going to the wrong Javascript laden web page doesn't slow my phone to a crawl. I just tell myself that at least I only replace my phone every 5 to 7 years. If it wasn't for work, I'd keep them longer, but the OS has to be patched for work stuff to be on it.
> There has got to be a middle ground between 1K flagship phones and cheap sub-$200 phones, but I swear I can't find it.
The Pixel phones are precisely that, and they are the only phones supported by Graphene for now. The a-series is a particularly good value and I am posting from one.
Also using Moto G-series phones for a decade or so now. Seems I have more junk (gambling apps, linked in, etc.) to scrub off them each time though.
I could see myself splashing out on a more expensive phone if it were less locked down - so probably never. However I don't play games and as a Gen-Xer my laptop is where the rest of my digital life lives; it makes sense that those for whom the phone is their primary device would prioritise it more (especially with somewhat predatory financing).
The only reason why I've decided to pay 500+ (for the first time in my life) for a flagship device was for a better camera and storage, as I wanted to have good quality photos of my kids.
That said, I will keep on replacing the battery until the circuit board itself turns into ash
Sadly there are no more iPhone SE models so this strategy won't work forever. Eventually iOS 18 will stop getting security updates and you'll have to use the increasingly worse newer versions of iOS.
Nah - Music, TV, Fitness etc all included. All shared with the family (and extended family) so it doesn't have the friction of piecemealing it all together, which I value more than anything.
reply