Grounding identity in a phone number is very reasonable for almost all normal usage. It makes recovery simple. It does block the ultra paranoid use cases though. Oh well.
Grounding identity in a phone number is very reasonable for almost all normal usage. It makes recovery simple. It does block the ultra paranoid use cases though. Oh well.
Session aims to provide anonymity, Signal aims to provide privacy.
Yeah if you compare that with Facebook messenger and other such services but if you want secure communication it's not reasonable.
In many jurisdictions, telecoms form an abusive oligopoly, and you need to provide a state-issued identity document to get a phone number.
That is not at all reasonable for normal usage - unlike well-known non-abusive authentication methods, such as a keypair; or its even simpler cousin, the username/password.
Oh, so there are "wrong questions" now? Yeah, I guess I'll have to get better at asking those real quick.
On top of that so many other things just inherently expect one to have a phone number. It would be somewhat odd to not have a phone number for most of the people I know and talk to through platforms like Signal.
So to your question of which is easier, having the state ID and a phone number is easier because I'll already have that for a multitude of reasons.
If you live in a place where its rare to have a phone number, then yes I agree Signal probably isn't a good choice.
What's easier, using a tool that's already in your hand or going to the tool store, searching for a new one, and swapping to that one? Just using the tool already in your hands, that you're already using, that you've been using for a long time.
It's exactly what you asked, just not the perspective you cared to look at.
If you insist that 2+2 equals "3 or 5 depending on perspective", ok sure let's work with your scenario of comparing registering a pre-existing state/telco ID with creating a brand new user/pass account (again: instead of what I asked you to compare). Well, suppose you were to lose your ID and/or your SIM - do you think you'd be able to renew them and regain access to your stuff in the same amount of time it's taking me to write this now?
Plus, I guess you've never seen bad actors buy fake ID/SIM accounts faster than you can block 'em - and much more quickly than any individual good-faith actor (with a single ID/SIM honestly linked to meatspace) would able to get out of harm's way in such a scenario.
>What's easier, using a tool that's already in your hand or going to the tool store, searching for a new one, and swapping to that one? Just using the tool already in your hands, that you're already using, that you've been using for a long time.
Hammers, nails... You're trying to make it make sense by dumbing it down.
In the end (again conforming to your framing), the "easier thing" is the one that fewer people will expend energy to prevent you from doing. And this last consideration kinda-sorta makes your reply be an answer to my question, except that I had to do the work of connecting the two, so now you owe me 100EUR.
I've answered it twice. Its easier to use the thing you already have than it is to make something new. Is that really difficult to understand?
> you could've registered a brand new account with username, password, email confirmation, and OTP 2FA, on any service that supports those
And then spend the many, many, many hours to share that new unique identifier with all the people I want to talk to. Or I can just continue using the same identifier they already know me by and have known me by for decades.
> I guess you've never seen bad actors buy fake ID/SIM accounts as quickly as you can block 'em
Whaaaa I thought you just told me a phone number is such an incredibly hard and challenging thing to get, now you're telling me anyone can easily get them anytime they want?
So phone numbers are incredibly challenging to get and yet people get them all the time easily. Otherwise, if phone numbers aren't hard to get and anyone can just freely get them what are we even really debating about?
For billions of users, having a phone number as their identifier isn't a challenge and is for sure the easier process and having to make a new unique identifier is a bigger deal and introduces far more roadblocks to effectual adoption. Its why WhatsApp uses it, its why iMessage uses it, its why Telegram uses it, its why WeChat uses it, and many others.
I asked which is easier to make.
This is, evidently, an extremely difficult question to understand.
>Whaaaa I thought you just told me a phone number is such an incredibly hard and challenging thing to get, now you're telling me anyone can easily get them anytime they want? >So phone numbers are incredibly challenging to get and yet people get them all the time easily.
Not phone numbers; accounts gatekept by them.
Not anyone; only bad actors.
>I thought you just told me
What I just told you is written in the parent post.
You have the right to misread it any way you like, and think whatever comes to your mind easiest. But since that's not how a conversation works, I will ask you to exercise that right somewhere I can't see you - and by the way you still owe me 100EUR, but since I'm not an actual cyberstalker, I won't be chasing you down to collect. So relax.
Which is easier to make? The one I don't even have to make, I already have.
Which pizza is easier to make, the premade pizza that's already ready to eat and is right in front of you, or the pile of ingredients in the pantry and fridge? Uhh...the pizza in front of you ready to eat? Put it on a plate and eat, it's there, it's ready to go. And practically everyone already bought their pizza and it's already in front of them.
And no, you didn't ask which is easier to make.
> What's easier: to obtain state ID, or to sign up to a website with your preferred username and password?
Where's the word "make" in that?
> Well, suppose you were to lose your ID and/or your SIM - do you think you'd be able to renew them and regain access to your stuff in the same amount of time it's taking me to write this now?
Well, I'll end up wanting to replace that ID and SIM for once again a multitude of other reasons, so having that identifier tied to that really doesn't increase any complexity. Its really not that big of a challenge for billions of users, its something they're going to already do. When the school needs to urgently get a hold of me, they'll call my phone number not refer to me by some handle on a jabber server with a population of one. This same thing applies over and over and over and over for tons of people.
Meanwhile if I've got other accounts tied to other processes that get lost in their own unique ways, I'll have separate issues to get those unique identities recovered. Sounds like more work, not less work.
You now owe me 20,000,000,000,000,000,000EUR, but since I'm not an actual cyberstalker, I won't be chasing you down to collect. So relax. Such a useless thing to add to the conversation.
My condolences to any children who have to interact with you.
>You now owe me 20,000,000,000,000,000,000EUR, but since I'm not an actual cyberstalker, I won't be chasing you down to collect. So relax. Such a useless thing to add to the conversation.
Nah chill, I'll be there shortly with the money.
It's worth mentioning that Session had started out as a fork of signal.
Obviously, I'm not really claiming that it's not possible people are experiencing this issue, but it can't possibly be widespread.
I feel like most likely people are using android skins that aggressively kill apps in the background.
There's something deeply wrong with the way signal delivers messages...
For this reason, it's hard to trust them. The encryption quality is irrelevant if the slop coded client is blasting random photos to random contacts.
Send a GIF to Contact A, Contact B receives random private images? Absolutely inexcusable slop code project. This class of state management bugs should not be possible with a well-architected client, period.
Signal's E2E encryption is more like End 2 Random End.
From GitHub comments: https://github.com/signalapp/Signal-Android/issues/10247
Greyson:
> Hi there, sorry, this issue was fixed in 5.17 (which hit 100% production on 7/21)
They had a difficult to reproduce problem reported in late December 2020, and got the fix rolled out seven months later.
Not sure your criticism "absolutely inexcusable slop code" is well considered.
Knowing that bug COULD exist, means that you cannot be sure that messages you send in Signal will make it to the recipient you intend given the poor quality. This means the E2E encryption is fundamentally broken, by the way. Because the client is lying to you about the true state of who it's about to send to.
The recipient text has fundamentally zero relationship to the true recipient of the message given that bug.
Having the UI and message sending code reference two different versions of state is incredible incompetence.