Yeah, this is just me tacking on extra features I'd like. My security-minded programmer brain can't help but think of all the edge-cases. It's something that is suddenly possible with distributed social media that never was before.
xthexder
joined 1 year ago
I assume you hold your own cryptographic keys, but I'm not actually sure how that works. Your client needs access to them to make posts, and it wouldn't make sense for the server to hold your private key, since that would mean the owner of your instance could make posts as you.
I haven't actually signed up to BlueSky to figure this out yet.
Edit: So it looks like users are authenticated using https://github.com/did-method-plc/did-method-plc But the keys are stored on the server, with an option to view your key for backup and migration. That does mean a certain level of trust with your instance, but you can self-host if that's a concern. A malicious host at least can't prevent you from rotating your keys and leaving (unless of course they steal your account entirely by rotating your keys themselves)
That's certainly a problem. It's one of the big reasons I think THC vapes should be both legal and regulated. In the states were it is legal, there's strict inventory tracking every step of the way.
Admittedly it's a lot harder to get people on board with regulating drug-free vapes, but I think it would be a good idea to have guarantees about what you're consuming just like food.
Well, I'm impressed they actually did test JUST the vape liquid, even though they're still calling them e-cigs.
Quoting from the journal itself:
There were no significant differences in changes of BAL inflammatory cell counts or cytokines between baseline and follow-up, comparing the control and e-cig groups. However, in the intervention but not the control group, change in urinary PG as a marker of e-cig use and inhalation was significantly correlated with change in cell counts (cell concentrations, macrophages, and lymphocytes) and cytokines (IL8, IL13, and TNFα), although the absolute magnitude of changes was small. There were no significant changes in mRNA or miRNA gene expression. Although limited by study size and duration, this is the first experimental demonstration of an impact of e-cig use on inflammation in the human lung among never-smokers.
The way I read this, it seems like there's a small correlation with inflammation, but there's no measurable risk of developing lung cancer from it (they were doing cancer research after all). Personally for an adult, I feel like "inflammation" is kind of a nothingburger, just stop vaping for a while and you'll be fine. But for kids developing habits, I can understand the concern.
If you'd like to point me at some studies go ahead. The only dangerous cases I've heard about were black market vapes that had other contaminants in them. It's been very hard to find reliable studies because most I've seen are self-reported using the entirely generic term "vaping" without any qualifiers on the kind.
The way you phrased that perfectly illustrates the current problem AI has: In a problem space as large as natural language, there are nearly an infinite number of ways it can be wrong. So no matter how much data we feed it, there will always be some "brand new sentence" someone asks that breaks it and causes a wrong answer.
I'm getting a lot of downvotes, and maybe I'm wrong about what kinds of vapes kids are using? Obviously if they're using nicotine vapes, that's bad and chemically addictive.
But I don't have a problem with kids vaping the drug-free, flavored juice. It can be habit forming, but so can fidget spinners. As long as it's not actually dangerous then I don't see the problem.
Banning fruity flavors sounds like it would inadvertently ban all of the drug-free vapes... Flavor-only vapes get you all the big clouds and cool-factor that's a big drive for kids, with none of the Nicotine or weed. Just inhaling the vapor on its own can be fairly safe.
Supposedly BlueSky has solved this by separating the data storage servers from the "relays" and "app view" servers, and since your account's posts are cryptographically signed, they can come from any instance as long as the signature matches.
That at least covers migrating followers and new posts, but I'm not really sure what would happen to old posts if a data server just went offline. I've still got more reading to do.
The main problem with that seems to be that the original server needs to be active to migrate. If the instance I'm on shuts down or is uncooperative, then my account history is gone. And for Mastodon, that's even worse if you have a bunch of followers. These are all reasons I decided to self-host before I built up too much of a presence.
Well, I guess that's a point against flavored vapes. I really wish there were more studies, because presumably not all flavorings would have the same effect. A comparison with unflavored e-juice would have been great.