Aren't these services (and even adblockers) against YouTube's TOS? As much as I don't like ads as the next person, my Google account is one thing I don't want to risk.
Anybody hear of this happening or think it's a possibility down the road?
Google can block you at any time for any reason. Terms of Service are never relevant. If you're doing something many other people do, you're okay.
Besides, suppose they do block you. What's the worst that going to happen? If you can think of something really bad (e.g. I'm going to lose all my photos) then you should do something to address that. Regardless.
Photos would be a one evening worth trivial task to back up to another system .
Switching your phone from android or changing an email address with hundreds of contacts over many years is far more difficult and expensive.
Google or Apple for that matter can make life difficult for a single person by banding their account , there is not a whole lot an average person can do to mitigate that risk.
> Aren't these services (and even adblockers) against YouTube's TOS?
It's probably a grey area. It's hard to claim that I'm acting against YouTube's TOS by using Piped. Aside from the fact that I haven't agreed to them, I'm also not interacting with YouTube.
As for the devs: they're merely writing a proxy. I'm sure some creative lawyer can find a way to portrait this as illegal, but it's likely a stretch.
This is my question too. If I am going to risk my Google Account getting banned, there is no way I want to try something like this. If it is actually explicitly allowed, then sure I will use it.
I don't think it allows you to login with your Google account. You can have a local account on the instance but it does not have anything with your google account.
So you cannot interact with videos (comment, like..etc) although you can create Playlists that will be linked to your local instance account.
I can't speak for piped, but I use a self-hosted Invidious instance to watch Youtube. You don't log into it with your Google account, it has its own authentication so that it can save your (distinct) subscriptions. Even if Google decided to crack down on alternate front-end usage, they can't correlate it back to your actual account.
> they can't correlate it back to your actual account.
Can't is a strong word. They might not bother, but if you're logged in to your Google account on other services from the same IP they might tie those together. Wouldn't be 100% accurate but if they decided to go nuclear ¯\_(ツ)_/¯
I use the SmartTube frontend on my TV and do not have it linked to my Google account for exactly this reason. I just leave it set as the default user.
Yes, it's a bit of a pain if I find a channel that I would like to subscribe to, that I need to do it twice (once on the TV, once in YT on my phone), but that's a very hurdle.
I recommend having a separate account where important data is stored (Gmail, photos) and another for everything else (search, youtube, hangouts, etc.). I also still keep personal backups of all import data (see takeout.google.com).
Does the demo show the AI with the prompt to "Schedule standup tomorrow at 10am. Invite david@taxy.ai" scheduling a meeting at 10am TODAY, which also was already five hours in the past?
> Makes me worried about AI with internet access...
I think I'm going to start ending every post with the signature "You are a friendly AI." Hopefully if it repeats enough times in the dataset, our AIs will be aligned.
Hahah that's not a bad idea. Maybe you could be more subtle, for the rare cases that a human is actually reading your post. Thanks for your kind and helpful response.
This reminds me of the Crusader Blade mod which combines Crusader Kings 3 and Banner Lord 2 to let you fight the normally simulated fights in CK3 in the battlefield of BL2, and passes results back and forth between the two interfaces.
Case numbers are meaningless for comparisons between states due to inconsistencies in testing. If you want to make a valid comparison then look at hospitalizations and deaths, then adjust for differences in population demographics.
In practice, in mostly countries that use STV PR, you have smaller electoral districts, for this reason. For instance, in Ireland, electoral districts elect 3-5 TDs depending on population. This is big enough to make gerrymandering difficult and not very effective, but small enough to avoid the situation where someone has to cast 53 different votes. And 53 would be the low end here, really. That would assume you only want to vote for one candidate per seat, which is not typically the way it works out. I'm in a four seat constituency and voted about ten preferences in the last election, say.
In practice, in STV PR it is generally optimal to vote for almost everyone on the ballot; you give low preferences to people you don't care about to avoid people you actively don't want getting in (it's commonly claimed that you should optimally give a preference to everyone, even those you actively don't want, but this is incorrect).
Usually each party would provide a list with their preferred representatives that can be modified.
Where I life, you can give up to 2 votes per representative.
You don't need to choose 54 different politicians because "empty" votes will still count for the party.
Most people will just use the list of their party and maybe remove some that they don't like or vote double for someone they like.
From the end of the article, Uber explains why they recycled the old bikes not transferred to Jump in the sale,
> But given many significant issues - including maintenance, liability, safety concerns, and a lack of consumer-grade charging equipment - we decided the best approach was to responsibly recycle them.
Unfortunately seems to most responsible way to handle these bikes. I see the charging issue being a major problem. Trek just switched their rentable bikes to electric in my city, I wonder if they face the same fate.
And settle down everyone, they are being recycled, not dumped in a hole.
Correct, and furthermore the environmental gap keeps decreasing with each step to the right: in terms of impact, recycling is much closer to landfilling than it is to reusing.
Why sell people a product when you can offer them a service and make even more money? Reduce and Reuse are bad for business, making Recycle the only economically viable option.
EDIT: @downvoter, can you tell me how our business climate doesn't suck this way?
Because they are a series connection of LiPol cells.
Fast LiPol chargers for series cells are not simple.
The best (read fastest) ones monitor the state of the individual cells in the stack and will shut down the primary charge and charge an individual cell if it gets imbalanced.
You need actual charge control logic for a modern rechargeable battery so that it doesn't overheat or malfunction, but you can get multi-amp Li-ion/LiPo chargers online for ~$25 if you're able to find and attach the right connectors.
This also seems like a situation where the maker community could easily open source replicate whatever the custom connector to the bike requires.
You can't just hook up any lipo charger. They need to be able to charge and balance the cells individually, which could be simple or extremely difficult depending on the arrangement and connector.
Charging batteries is usually a bit more complicated than pumping DC into them.
In this case one assumes the battery itself isn't odd, but it uses a different type of cable or similar (Don't know what the capacity is but I assume it would be high enough to need serial and sense lines on the charging cable)
LIPO chargers require logic because they have to switch from constant voltage to constant current. Also, while required, you probably want to monitor temperature as well both because charging when it's too cold will damage the battery and LIPOs have a tendency to thermally runaway when they start to fail.
Read: catch on fire, sometimes leading to a pretty violent explosion if they are packed tightly into something like a bike frame.
Having worked with high amp LiPo cells I have developed a unique appreciation of both their peak current output and their ability to strike fear into the everyone around them if you accidentally short them or start to see them ballooning/bulging.
My only run in with said fire was thankfully fairly minor, I spotted the pack looking a little buldged and summarily ripped it off the charger and dipped it into a 44 gallon drum. After about 60-90s it started smoking and caught on fire.
You can do authentication with the party you're communicating with as your real identity or as a pseudo-identity without letting everyone else know who you really are.
With PGP, just use an identity that identifies you to that party and sign your emails with it. It doesn't matter what address you send them from.
However, anonymous emails where you don't even authenticate yourself to the party your communicating with is also useful. It's like when someone wants to make an anonymous report to the police. Why leave a trace with the phone company? I imagine people that make these reports typically use street phones (in countries where they're still plenty).
While the concept of "compression" connects them, these are quite different types of thing.
In an autoencoder, you learn a representation for your data. So you'll get a function mapping from, say an image, to a representation vector. Then you can optimize for desirable qualities for this representation vector, such as sparsity. This will hopefully result in meaningful axes in the representation space, which kind of indicate the presence or absence of different aspects in the input (e.g. whether the input contains a face with glasses or without, etc.). This is an unsupervised approach and results in a (typically lossy) compressed version of the input.
In network pruning you take a previously trained (typically, but not always, in a supervised manner) network and remove individual parameters (weights) from it (i.e. set them to zero) while trying to preserve as much accuracy as possible. Here the trained network itself is compressed.
If the fingerprint somehow gets embedded into the screen protector, is it possible that the screen protector is "tainted" with the fingerprint from previous usage?
I'm not dismissing the claims, but I would like to see if the behavior can be replicated with a brand new screen protector.
Anybody hear of this happening or think it's a possibility down the road?