AI certainly is the shiny new hammer, and it is tempting to see the world as nails.
Traditional methods might not be perfect, but they also easily fit in the memory of even low power devices. Perhaps it isn't a problem worth burning a dollar of tokens for every spelling mistake.
Haven't they hemorrhaged a lot of the founding talent in the years since? Now it is full of ex-Meta ad-tech people trying to find a way to make it actually turn a profit.
It was pretty obvious he was wanting to enter a conflict with someone, and was mostly held back in his first term by the actual professionals in his cabinet at that time. But the guy wanted a military parade with tanks rolling through DC on his birthday, wanted to nuke a hurricane, and forcibly annex Greenland. It isn't really surprising that once he replaced the sane people with sycophants, he would start something.
Claude folks proudly claim to have Claude effectively writing itself. The CEO claims it will read an issue and automatically write a fix, tests, commit and submit a PR for it.
I would probably say that both the city and the cop should, independently, be liable. Given the position of authority the city provides, it is ultimately responsible to hire and properly train people who will use that authority well, while the individual is also responsible for their own actions.
From what I remember of a presentation they had on how and why the made Go, this is no coincidence. They had a lot of Python glue code at Google, but had issues running it in production due to mismatched library dependencies, typing bugs, etc. So they made Go to be easy to adopt their Python code to (and especially get the people writing that code to switch), while addressing the specific production issues they faced.
I wonder if we just don't have many of these types of satellites in a polar orbit, since we don't have as big a need for that type of imagery for the poles?
On the other hand, I had the misfortune of having a hardware failure on one of my Hetzner servers. They got a replacement harddrive in fairly quickly, but still complete data loss on that server, so I had to rebuild it from scratch.
This was extra painful, because I wasn't using one of the OS that is blessed by Hetzner, so it requires a remote install. Remote installs require a system that can run their Java web plugin, and that have a stable and fast enough connection to not time out. The only way I have reliably gotten them to work is by having an ancient Linux VM that was also running in Hetzner, and had the oldest Firefox version I could find that still supported Java in the browser.
My fault for trying to use what they provide in a way that is outside their intended use, and props to them for letting me do it anyway.
> means you can stop paying $1000+/month to someone who is already a millionaire, that's still a savings even if it adds $20 in overhead.
Only if these hypothetical millionaires you are stopping make up more than 1/50 of the people you are means-testing. You are not only paying for those who fail the means-test, but for all those who are passing it.
The looks are what they are, but if that wide shelf will mean the phone doesn't wobble when placed on a table, that is a good usability improvement. One of the small things I missed when moving from a Pixel 6 to an iPhone was the ability to not feel compelled to pick it up in order to use it.
I largely agree, but when we hold phones it is generally by the side without the camera. That means that this phone will feel smaller in the hand, which could be a very effective marketing gimmick to upsell people from the base iPhone.
It isn't as bad as some practices, for sure. The question is how likely are the 'upgrades' actually upgrading anything for the user? Will the extra camera on the Pro be $100 of utility for the user over the lifetime of the device? Or are they using the uncertainty that the user _might_ get a use out of that camera to push to a higher model.
It seems mostly an exercise in price discrimination. You always have a slightly higher price point, and some extra functionality to justify it, and the customer will likely push themselves up to the maximum they are willing to spend instead of settling on the cheapest option that meets their needs.
The confusing choices are deliberate way to exploit psychology of potential buyers into up-selling themselves. The idea is to entice them by the more reasonable base price, but use the uncertainty on if it will really meet their needs to push them up a ladder of upgrades.
Maybe the 16e sounds good at $599. But, it might be a bit underpowered, so maybe you should just upgrade to the 15 at $699. Then it is only $100 more to just go for the 16 (or 15 Plus), so might as well right? But maybe you want a bigger screen or twice the storage, which are both another $100. Then for another $100, you can get the nicer materials or the extra camera, etc for the 16 Pro...
This is a marketing strategy you see in a lot of the phone market, and has proven to be successful at pushing customers into the higher-margin devices.
Of course, but that was my point. Even in traditional media, exceptions are made for factors outside the control of the publisher. User-generated content is, by definition, outside the control of the platform so there should be at least some exceptions made for it. Some mix of filtering, moderation, and content flagging are its form of "reasonable attempts" to moderate the platform.
It is the end result if you expect the same level of liability as a newspaper or magazine. Every single thing you see in one of those was deliberately put there by a person (well... at least it used to be). If an agent of the print publication deliberately put something in the publication, then the liability falls on the publication and/or that person.
Social media is not the same. The content being posted is not vetted by any agent of the platform, so the liability at least in part falls of the person who posted it. You could argue that the platform should share some liability that is waved as long as they at least try "hard enough" to police their platform, with whatever definition of "hard enough" is chosen. But no automated filter will be perfect, so if you demand the same level of liability as a print publication you are effectively outlawing social media entirely.
"Algorithmic boosting" is not (always?) the same as an editorial slant. Promoting the post with the longest title would be an "algorithmic boost", but clearly not editorial in any way. The most common forms of algorithms are just a function of the number of times people viewed it weighted against the age of the post; there is still no editorial slant there. Even recommendations algorithms like YouTube are mostly the same, with an additional weight based on how likely others who watched the same things as you were to view that video.
This isn't entirely true in all cases. Consider something like a live broadcast of a sporting event. If some streaker runs naked across the field, are the stations held to account?
That is, in a way, similar to the problem of user-generated content. There is a limit to how much control a social media company will be able to have over the actions of its users. Unless you replace the system entirely with one where all posts are manually approved by a person before they go up, you will need to have at least some reduced liability for the platform owner.
The switchup of 'worst' subject in high school to college seems so striking to me. At best, I could see it coming from an over-fitting of data. At worst, it was a test designed intentionally to fail anyone without the answer key. Not even 'playing in to stereotypes', but 'what combination of answers did no one choose, so we can block everyone?'
Traditional methods might not be perfect, but they also easily fit in the memory of even low power devices. Perhaps it isn't a problem worth burning a dollar of tokens for every spelling mistake.