greycol

> I think the fundamental disagreement is you believe that cheaper models can just break the state of the art.

No their argument is that the existence of a worse quality cheaper product will eat the majority of the market for the SOTA company, making the SOTA company less profitable (not necessarily unprofitable) because the number of people who need/want the quality difference supplied by the SOTA product aren't enough to pay to sustain the extra resources required to supply the next iteration of the better product or get a respectable ROI on the current investment.

>SOTA can also stop developing and spending on SOTA and just use a cheaper model

Yes, and they will be forced to (if the market shapes out as above) but the point is that then the company won't be getting the returns that justify the valuation they are talking about in the IPO and will likely never return the amount of money that has already been invested in it even if it is a profitable company. i.e. If I spend $20,000,000 making a lemonade stand that is a successful best in class stand and it nets me $50 profit a day then it doesn't matter how that as far as lemonade stands go that it's a pretty amazing example. I am effectively never going to see that 20 million again let alone a real profit.

They're not arguing that these SOTA companies aren't better, they're arguing that there won't be a return on investment let alone justifing further investment and means everyone will be taking a huge haircut on what is already invested (with the possible exception of early investors who take the bag with the IPO when it inevitably succeeds based on people disagreeing with this evaluation and thinking it a good investment or simply because of index funds being required to buy in)

You can implement policies that tax more than just income. So no not regardless of policy but yes where policies don't exist to reshape a system where capital (i.e. investment in automated production) inherently takes more of the wealth than labour it will inherently lead to an agglomeration of capital.

While a good rule of thumb for the US court system is that someone with more money can screw over a smaller party. That's only up until a point. The most expensive lawsuit in history by legal fees was (adjusted for inflation) $1.15 billion. That was the 1998 Tobacco masters settlement (involving years of litigation by law firms across 46 states). So any award in the billions makes a case arguably justifiable (even more so when the highest litigation cost cases are often that expensive because the litigation fee includes a percentage of winnings).

Companies like autodesk with a ttm revenue of $7.8 billion would probably have fiscally compelling cases (if not legally compelling) to pursue if reverse engineered tools cut into their revenue. That's without getting into class action cases.

No because americans actually commonly call it the Gulf of Mexico (despite US government fiat dictating it should be called the Super Intelligent Gulf and anyone against that being double plus ungood), Japanese people don't commonly call it the East Sea. That's the difference.

That's like saying bank robbers don't actually shoot guns, only people who use guns legally shoot them. It's still preservation. Or even more apt saying archeologists aren't archeologists they're tomb raiders. They may be tomb raiders but they are still archeologists.

You are wrong, you are basically saying "black cats exist therefore cats can't be tabby", there are plenty of preservation initiatives that preserve works in different forms when the original media can not be practically preserved or displayed. Yes best practice is to preserve as close to initial state as possible, that does not mean that when you cannot feasibly preserve the work in it's original state (with digital it's often due to drm) that you just say screw it nothing we can do but chuck it in the dustbin of history.

US case law may not be on your side there. Generally unless you require a login to access the site (and even then you can only really ban for terms of service rather than recover damages unless they hacked around login requirements) judges have found that using scrappers on publicly served data is fair use. It gets murkier with creative content (rather than straight data) in what they can do with it after scrapping but the scrapping itself would still be 'legal' in the they can do it without judicial consequence sense.

It's one of the reasons why sites that served a lot of data used to have APIs or back channels to get that data even if they technically made money by serving ads on pages of that data, because they were going to get scrapped no matter what, so having a legitimate way to get data that behaved nicely with the servers and was nicely readable by requestors was win-win or at least not lose-lose.

Without getting into the much debated philosophy of restoration/display of preserved things and how restoration can often cause degradation to the original piece, consider that just preserving the data in the manner you suggest also requires preserving the original software if you ever want to view it. Which requires not just preserving that original software but also all the external hardware and software that original software relies on. Quite often that's a dead end as it requires preserving an activation/license server that the preservers don't have access to.

It is worth considering that without being able to identify people then only unacceptable behaviour that rises to the level of needing and having an immediate response can actually be punished.

It's the qualitative change than comes with the change in quantity and ease of monitoring (and accessibility of that info) that means being identifiable is so much worse now days.

The reason so many people used the deleted items folder as a filing system is exchange didn't set a quota on that folder and by default would only get emptied by a manual emptying. So back in the days when a user could have a 100mb mailbox quota 'unlimited' space could start looking very attractive. Especially if that quota was 5 years old because the server hadn't grown with the times, since what business wants to spend money on a 'cost center' that won't improve profits.

Then these people didn't change their ways when they were given adequate space.

>> I can not think of anything a non-CS person...

Install an app. For apple you need to try test flight to find out too many people use the app so you need to be able to install in developer mode.

I can understand the argument that you should have some technical competency to evaluate that these programs are safe to install, but neither accounts for a technically competent friend/family member suggesting/helping you install an app. Apple requires expertise to even keep an app that doesn't pay to be on the app store installed longer than a week.

People joke about a fire drills at work or their neighbour in the apartment building burning the toast and setting off the alarm. We don't equate that to people laughing about dozens of people burning to death (even if that's a possible reality of a fire) because to most of us it's not conceptualized like that.

This is a teacher in the US who would have had to sit through live shooting drills and never believed a live shooting would happen at her school (How much would you need to be paid to work somewhere where you seriously believed you had a 1 in 100 chance of being shot in the next 5 years, it wouldn't be a teachers salary for most of us).

No because, some companies know they'll need ram, but they may have some leeway on timing (i.e. doing more to prolong the life of systems they'd typically replace). That means to guarantee ram they need to enter an agreement to buy the ram now. By setting the expectation that these prices continue or get worse companies that may have put off agreeing to these terms locking in prices based on demand rather than something closer to the cost of production. The agreements that these companies are being forced to make are upfront payment and then a top up to meet current market prices before delivery, there is no downside for these ram companies whether the price rises or falls (at worst they end up with supply that was already payed for by a bust company at a healthy margin and a glut in the market that would arrive anyway)

Competitors are going to come online anyway because of the threat of chip embargoes against China that mean they are investing in production regardless of what the outside world is doing.

The broader tariffs are the worse they are. Targeted tariffs can have a function but if targeted at a country that function isn't an economic good for either country it's only good as a political tool. You could argue that industries that you are the second cheapest in compared to that country could reap some benefits from supplying local markets (but not if there are other countries cheaper than your own that aren't also tariffed that can supply more of the demand) and then a targeted tariff may make some (long term) economic sense.

Targeted tariffs can have more nuanced effects. You'll notice that Canadian politicians are talking about targeting red states with their tariffs. These tariffs will also economically hurt Canada, but you won't see a retaliatory tariff on things that hurt Canada disproportionately which would be the case with universal tariffs. You can expect them to pick luxury industries (like they did with bourbon) where preferences can be changed or industrial/commercial ones where (preferably) local or other suppliers are near parity with the US rather than say gasoline.

Which is ridiculous on it's face, imagine claiming that stopping someone's income until you get around to reviewing it doesn't have a significant effect on someone.

The real issue is if uber comes away from this thinking they can do a rubber stamp review, because the point of the law was clearly to stop unjustifiable decisions automatically being made(protecting both parties), as uber will still be on the hook for unjustified suspensions but redress will need to go through the court making it worse for all involved.

Anna's Archive is a controversial example with that as a stated goal. But more to the point most of the people doing it aren't seeking to profit from digitizing these books so the 'monetary value' of them is not the yard stick or motivator. A higher monetary value could actually make it harder for these projects as they can rely on donations of books or even waiting for someone who wants a book they bought so they can spend that on the next one (which is less useful if the next one on your list was bought to destroy by the same person who bought the book you finished scanning).

They're often books tied to a region, i.e. stories about a local fisherman's life and how they overcame adapting a boat to local feedstock, or the engineering that went into building the local train station. That's to say they can have little value to those outside the area but are still a rich read for locals, tourists, and those who move to the region or have a love for a place. I expect we'll see a concerted effort to organise to get these kind of book sent to regional mueseums for scanning now that people are aware they are actively being destroyed and they realise the timeline isn't how long the book can last.

Which is what has happened over time now that humanity has the ability to digitize them. The problem arises when we are purposefully picking books where we know those scans aren't yet available and purposefully destroying them. There'd be less outrage if any of these companies made statements about guaranteeing the survival of the scans they are taking and the release or planned guaridanship in the case they can no longer maintain the data.

Engineering considerations in favour of this setup:

The cost of that 30% oil loss is less than engineering/site selection of a system that can be fully drained.

The cost of a more robust system that isn't as damaged by use is higher over it's lifetime assuming the *Emergency* use case than just creating/prospecting replacement damageable reserves if necessary.

If someone wanted a small reserve for 'market stabilizing' short term and repetitive small fill/drains they could build it for a lower cost than making the emergency reserve resilient enough to serve this purpose.

Can't claim these are true in this case but also remember that norms were much more enforced at this time and voter could be expected to punish bad politicians (This was a couple of years after watergate) so expecting political parties to hold short sited/corrupt politicians to account when they were doing something visible may have influenced what they expected to happen to a "get me reelected on lower gas prices damn the consequences politician". So the main concern could easily be the stated 1 or 2 emergency drawdowns in the next century and we'll over engineer it so it can handle 5.

They had basically been planning for this shortfall for ever as a likely outcome of sanctions/blockades during an invasion of Taiwan.

Curtailing imports so much is a smart soft power move in not only does it ingratiate countries that would suffer most from the price spike but it could soften the resolve of Taiwan in showing that one of the largest penalties that would result from the invasion isn't nearly as punishing which in turn makes Taiwan's porcupine strategy less viable.

Along with US own goals like the thoroughly bungled and unplanned invasion, tariffs, corruption (both business crimes and more moral crimes like pedophilia or disregard for following legal process) it's made most of the west rate China above USA in favourability [latest Pew research poll].

The first sale is the hardest, then you rely on word of mouth. Alternatively if the buyer is doing their research online before deciding to purchase they may ask for pictures of the product and like what they see.

You are much more likely to gain weight if you eat biscuits until you feel full versus eating lettuce to feel full. I.e. if you eat the same volume of BLT vs bacon sandwhich without the lettuce and tommato you will have consumed less calories with the one that is less calorie dense.

Which is a benefit in itself, it adds texture and is inoffensively flavoured (hell on the right day or if I haven't drank enough water yet it can taste great). All without adding many calories to the meal. It's also a much cheaper alternative than ozempic for those of us who don't have the consistent willpower to not treat ourselves to more food when we're hungry even if we know we have all the calories we need.

We've discovered that ambulance drivers are only driving people to hospital for 5% of the week. As such we've cut the 20 staff that used to give 120% coverage to 5. The 4 days of coverage is still more than 8x more than needed to cover working 5% of the week, so obviously we don't expect any deaths on the days we don't have any ambulance drivers working.

In the unlikely event excess ambulance drivers are needed we'll temporarily hire them on an ad hoc basis at double the normal rate. They will obviously happily jump on this opportunity, working less than 5% of the year at double pay is much better for just those hours than being paid 60% of what you'd get as an ambulance driver in another career all year round and actually being able to afford your mortgage.

But if a single person "can't recreate verbatim at length an entire work anyway" why does copyright law treat playing music to guests in your house differently to playing it to a classroom? I'm not saying that copyright law isn't stupid I'm saying that these distinctions make a difference and the fact that you can't copy humans is a pertinent distinction. LLMs would be treated very differently if like humans you had to spend the training costs of each model for each running instance of that model, the fact that models can be copied is a core feature of these models and makes them distinct from showing 'training data' to a human.

Or emptying your bottle of water when the TSA found it on the XRAY. Unlike the pepper he's actually allowed to have that data on his phone both sides of the border.

There are several distinct differences, I think the most pertinent one is that you can create seperate instances of the LLM that ingested that media without reingesting the media (i.e. you can copy the LLM but you can't copy a human). Copright law is screwed up for multiple reasons but it makes sense to treat LLM as different than a human consuming the media, especially if the LLM is created for profit (e.g. movie owner selling tickets to a dvd he bought in the store vs a home viewing is treated differently).

VPN is also common in those discussions, it's more setup on the client side but can win on security. Of course there'd be a lot of people in trouble if you couldn't trust tailscale to be better than a beginner selfhoster on security so it's still a good goto option.

While in jest, It does remind me of the time Dolly Parton entered a Dolly Parton look alike contest and lost.

It's always listed in the same spot on all the store page for games. Now why they haven't allowed us to filter games by DRM (like they have f2p) is another question. I'd love if they allowed this and published stats because I like most people assume I'm in the majority and would love to see that you're cutting off access to X% of your market if you include rootkits/aggressive drm.

There was only one category that it made sense to refuse the covid mrna vaccine in hindsight (excluding specific medical conditions), that was if you were over 10 and under 25 AND gave yourself greater than ~1,000 to 1 odds of not catching covid. That's based on the fact that the chance of a young healthy individual having a severe negative outcome from the vaccine was that much less likely than a sever negative reaction from catching covid. So yes in a vacuum you can say that having the mrna covid vaccine when you don't need it is a negative but "don't need' is specifically if you have low odds of catching the disease not low odds of having a bad outcome from the disease. I'd say a preeminently foolish bet by the time the vaccine was actually available.

I suppose you could also argue other categories based on higher or lower ages and who gave themselves lower chances of ever catching covid could technically also fall into that group. You could also argue moral reasons not to get the vaccine but I'd say in any situation where you still expected to get healthcare that detracted from the healthcare of others (i.e when the hospitals were at capacity and your treatment could kill another) would be morally bankrupt.