samsung is the only smartphone manufacturer that still makes phones (though not many) with all the features I want: microSD slot, dual physical sim, side-mounted fingerprint reader, headphone jack, nfc, and regular (long-lasting) security updates
they also have service centers pretty much everywhere in the world, so I can always get my phone fixed (for a reasonable price, as a result of their ubiquity) if and when I inevitably break it
would I also prefer the option to unlock my bootloader? yes. if I'm honest with myself, is it a deal-breaker? sadly, no, I no longer use custom ROMs
They seem to skip some years when bringing updated models to the US for some reason, but Sony Xperia phones check most of these boxes. I have an Xperia 1 V that I use as an app dev test device and as a backup phone and have found it pretty nice. The hardware feels great and their Android build isn’t nearly as junked up as Samsung’s. I’m always surprised they aren’t more popular.
There are no smartphones or pocket computers that tick these boxes anymore, since general-purpose computing is an anathema to the modern, specialized enshittification slop. For a modern device to serve most, if not all, relevant features, it takes a company that is built around principles that go beyond just shareholder satisfaction. You see the dilemma...
AFAIR, the Samsung Galaxy Note9 was the last device that deserved to be called general-purpose pocket computer. EMR stylus, 3.5 mm audio, mSD card slot, USB-C 3.1, good CPU, adequate memory for the time (8 GB), good cameras. If you're willing to forgive the non-removable battery, the only suck was the screen if you were sensitive to PWM, especially with regards to lower flicker frequencies.
Alas, seven years ago Samsung got the itch and divorced from good pocket computer design. The Note9 seems almost like an accident, given Samsung's market policies of today.
I have an xcover 6 pro with dual sim, 3.5 jack, removable battery and micro sd support, it works great (except buying an original battery is not super easy). I know the 7 is out too but I think its reviews were worse on amazon
> samsung is the only smartphone manufacturer that still makes phones (though not many) with all the features I want
Not to mention the built-in EMR stylus. That makes such a difference in using the device, I cannot believe they are not more common. And they are a terrific backup for the not unusual case of a broken screen being unresponsive.
you're just speaking of the Galaxy S line, there are at least four other Galaxy lines, some of which dropped these features only this year and one of which has all of them (XCover), though it looks like this year's release makes you choose between fingerprint and headphone jack (XCover7 vs XCover7 Pro)
it's clear in the thread that he got permission to do so
it's also reasonable to assume he had more information about the state of the location given his access as an employee, particularly given that it was a full two months before he actually retrieved them
It's clear in the thread that a forum user worked for BT. What was unclear was whether the site still belonged to BT and whether the employee was given official or any clearance to retrieve the parts. There was no 'we' language, all 'I', which is unusual at best. For a company of BT's scale one would expect a small team for such a recovery.
I'm curious, where is it clear in the thread that he got permission?
Yes, and people buying random GPUs for ether etc. I'm not a huge fan of what crypto has become but there was something exciting about hacking stuff together at home for it which is currently missing in AI IMO.
Maybe it's not really missing and the APIs for LLMs are just too good and cheap to make homebrew stuff exciting.
It's possible to run models locally, fidget with temp etc
Being able to change other things on the fly like identify weights most used for a prompt and just changing those to see what happens is much harder.
I've tried both LLMS and image generators on my machine locally and while it's gotten in easier it's a long task just setting up. Especially if you run into driver issues.
I don't actually think there was (or needed to be) one...keep in mind they're a non-profit. I think they just wanted to make the internet a safer place, but semi-extraneous (particularly unprofitable) projects sadly need to be cut aggressively with the rising threat of the google antitrust suit, as they may lose most of their income.
That doesn't necessarily change the overall mission of the organization, but definitely does give them more flexibility to offer paid options to help sustain development, should they see an opening in the future.
This is more or less taken directly from Thunderbird's website (which I think is a fair comparison): "Thunderbird operates in a separate, for-profit subsidiary of the Mozilla Foundation. This structure gives us the flexibility to offer optional paid services to sustain Thunderbird’s development far into the future."
my contractors provide me paid versions and it's my choice which one to use. Most of the time I use them as a sanity check and autocomplete. Anytime I get to anything complex like writing rust - every AI is useless to different levels
PS. Maybe I am a bad example, since I don't even track with model was where. I tick different boxes and see which one works better. In general, not a good experience
I agree, but it's fairly common practise. I think a "trust, but verify" approach should be used and jail board members for attempts to fool the regulators (c.f. VW emissions fraud)
The problem is that the manufacturer of course has a very strong incentive to stand with themselves. People don't advocate against themselves, and companies are no different.
When J&J found out about potential asbestos contamination in their baby powder in the 70s, they managed to convince the FDA that they would research and handle it. It took until 2020 for it to come to light that they did not do that, and that, in fact, their baby powder was contaminated.
They ran multiple studies, and some of them even showed that the amount of asbestos in their product was dangerous. But those studies never saw the light of day, and the company acted in a self-preserving manner. It's a similar story with 3M and byproducts of Teflon.
But, federal or state agencies have no alliance to a company's bottom line. They don't have the same incentives to lie or downplay. So, I think, it only makes sense that they should be responsible for testing directly, not just supervising.
I also think we need to adopt some legislation so that we must test products before we release them. You may be shocked to know you're allowed to release new chemical products without proving their safety, and you can even release new food products without proving their safety. Most of these products end up being okay in the long run, but some we have to retroactively ban. It would be easier for everyone if we begin in a banned state.
I do agree with that. It's a complicated trade-off between spending resources to police companies Vs doing all the testing.
Companies will usually be doing a lot of internal testing on new products, so they'll have a lot of the necessary tech and processes already in place. The trick is to ensure that faking results is penalised enough to make it not worth the risk. Most of the time it's cheaper to trust companies and then focus on the safety stats, though that fails with your examples of non-obvious issues.
The point is you either willingly spend the resources to independently test things for safety or companies WILL kill you to save a dollar.
This has been proven time and time again, big and small. We have the FDA entirely because a small company made "medicine" by buying medicine powder and putting it in a solvent that was acutely toxic that they didn't even think to give to an animal or something first. Literally just sold a random liquid to people as "medicine" and it killed a hundred people in utter agony.
Leaded gas was never a technical requirement. We could have been using Ethanol as our anti-knock/octane booster since the very beginning and never poison anyone but quite literally, the company chose instead to put Tetraethyl lead into gas because they could patent it, despite it being literally poison.
Same with PFOAs from 3M, who knew it was lethally toxic to mammals at fairly low doses decades and decades ago, but made no attempt to tell anyone or notify anyone or even reduce how much they were pouring it into waterways upstream of small communities. When they finally got sued by a lawyer who dug this all up in discovery, they finally said "Okay we will replace it" and replaced it with a nearly identical chemical that is just as toxic to mammals that they hope will break down easier in the environment, but how long will that take to demonstrate is false, and what are they going to switch to then?
Nixon made the EPA because companies would rather everyone die than change literally anything they've developed about their process or product, because they don't care about people dying. This idea of "bad press" or people will just stop using your products has demonstrably failed.
So suck it up, pay some taxes, test things for safety, and stop letting people die for such minuscule boosts in private company profits.
if that were truly (sustainably) possible, I'd support it, but imo that'd just be signing ff's death warrant
side note, thunderbird is already independent and democratically-managed by the community (as of a few years ago). the way I understand it is that they effectively just use mozilla's resources for legal, logistical stuff
Neither of these represent the cost and support footprint to maintain and develop a fully featured browser because as it stands neither of them are fully featured, complete browsers.
>cost and support footprint to maintain and develop a fully featured browser
True because the real cost to support a "a fully featured browser" is at least a 1/4 billion dollars....because OpenSource needs to make money, not for the Dev's but the MBA's ;)
they also have service centers pretty much everywhere in the world, so I can always get my phone fixed (for a reasonable price, as a result of their ubiquity) if and when I inevitably break it
would I also prefer the option to unlock my bootloader? yes. if I'm honest with myself, is it a deal-breaker? sadly, no, I no longer use custom ROMs