• 0 Posts
  • 24 Comments
Joined 1 year ago
cake
Cake day: June 30th, 2023

help-circle




  • This is an area of law governed at the state level. Some states are much better than others. Personally, I have not lived in a state that has a 3rd party hold the money (and I’m not sure if any do that). I did rent in a state where any charges that the landlord claims that they shouldn’t is met with triple damages. So if they keep $200 instead of the $100 of actual cost to repair something you broke, they owe you $300. It really incentivizes landlords to only charge accurately (e.g., not for standard wear and tear), and generally deposits were much lower there than in other states I’ve rented.

    Lots of states also charge interest on any deposit money not immediately given back to the renter.


  • They gave me an itemized receipt where carpet cleaning was the only item on the receipt when I moved out of a place with wood floors. I actually recorded the whole final walk through with the person from the company walking through saying that it was perfectly clean and that I should get my whole deposit back.

    When I complained, they said that it wasn’t carpet cleaning, it was just regular cleaning billed as carpet cleaning. I said I would take it to small claims court, but I never told them about the recording.

    They decided to refund me just enough that the money they kept was equal to the cost of filing a small claims suit.






  • There’s basically a tree of operations that have been applied to a model. At any point, you can go back and edit what you’ve done at a previous step. For example, if you padded a feature out 10 mm, then added more stuff onto that feature, you could still go back and change that padding operation to 15 mm.

    I’m still super new to freecad, and I haven’t done anything too complex in it yet, but my understanding is that some types of those changes can result in the topological naming problem. The way I understand it, when you make a shape, the software numbers all of the segments, vertices, and faces. If later changes are applied to those numbered faces, etc, and you go back and redo the operation that made those faces, etc, in a different order, the numbering will be different, and it will break your model.

    There is a fork of freecad that fixes that whole issue, but the fix hasn’t been implemented yet in the main fork cause it’s pretty foundational to the working of freecad, so there’s a lot of things that can break






  • Thanks for looking through all of this. If I’m understanding right, it seems like Congress is asking NHTSA to do a task that is probably not possible, but they are required to at least go through the motions to try?

    It seems like they just told nhtsa to use technology to fix drunk driving so they can wash their hands of the situation and claim they tried to do something, but nhtsa couldn’t figure it out. Why didn’t they tell the NIH to eliminate the cancer while they were at it.

    I do believe the technology to detect BAC is too erroneous to inflict on innocent drivers, and technology that could detect impairment through driving characteristics, while possible for individual drivers could never work on a population level. There’s going to be a lot of overlap between impaired drivers and just naturally bad drivers.


  • Im not 100% sure how the batteries are constructed from all of the cells (and I know it depends on the model), but the re-using process can work a lot better than just pulling out the battery and popping it down. EV batteries are in the range of hundreds of volts, but the cells themselves are about 4 volts. It’s my understanding that the battery as a whole doesn’t uniformly degrade, but you might have individual cells that degrade. If 1 cell in a chain of 10 goes bad, that chain can be made off limits to the battery, so you still technically have 9/10 cells that work fine.

    The way a lot of people reuse/recycle/refurb (not sure what the right terminology is in this instance) EV batteries is to test each of the cells themselves, and get rid of the duds, and keep the decent cells. Tesla, for example, all used to use 18650 cells (and I think some models still do), which is the same exact cell that’s in decent name-brand cordless power tools.

    When you aren’t required to keep weight, space, and extra circuitry to a minimum, you can really design a system that squeezes every last drop of usefulness out of those cells before they need to be melted down and remade.


  • Your title here isn’t technically wrong, but it is misleading. It should be "white AI faces judged to be more real than white human faces. The way you phased it insinuates that white AI faces are judged to be more real than a diverse mix of human faces.

    For anyone who doesn’t want to read the paper, they basically took an 60 white men and 60 white women, and showed them a whole bunch of white faces, half of which were generated by AI. It turns out that AI faces were rated as more human-like than actual humans, and they had some hypothesis why. Principally that AI, by its nature, generates images close to “average”, while real people tend to have features that are not “average”. The reason the study focused on white people is that most AI have been trained on white faces, so AI tends to do better with white faces.


  • I noticed the same thing. It seems like there’s a lot of background battery usage. I switched it from “optimized” or “unrestricted”, I don’t remember which, to “restricted”. I don’t use Firefox for any reason I’d want notifications, so I don’t need it running in the background at all.

    The one downside is that I have the search widget on my phone set to open things up in a private window, which is nice to avoid ending up with 1000 tabs open, but private windows now are not as persistent. Like if I looked up a recipe in a private window, a few minutes later, that window will be gone. I might have to switch the power setting to “optimized” and find a way to reduce background battery usage from within the app