What's stopping someone from offering to buy lunch for someone at the office? All it would take is one meal. Say the person gives a receiving address, and now they can see the other person's transactions. Then from their own personal paychecks, they can cross reference the transactions rather easily and discover their income if people aren't using multiple addresses .. and just look for two transactions that are the same. Civility goes a long way, and in this case it can be used to discover private financial information about someone that just wanted a big-mac and fries. Again, steps can be taken to avoid this - but not everyone does this. How many people use their facebook login info on tens/hundreds of websites? They're not gonna do the work to stay private - because it involves potentially lots of work, and it needs to be done on a protocol level else it opens them to abuse.
When the time comes where cryptos are being used to pay salaries, I'm sure there would be apps for the currently accepted crypto to help with the privacy issue. The question is, will the government accept these anon coins.
Incorrect, you're detracting from the argument. I will make the point that giving your ISP access to infinitely valuable, and potentially unencrypted, information that they don't have access to right now is a scenario for disaster. Right now, they don't stand to benefit nearly as much if they were to reveal this information, nor do they deal with the costs of keeping it private. If this information were to become incredibly valuable (which it would, because it would be financial transaction data) .. in that you're now not acting through a bank an instead directly with your transactor through an ISP .. multiple layers of obfuscation are now removed. I'm making the claim that if your ISP were to have access to your transaction records, rather than a bank, you (and them) will be in a terrible situation and as such the cost of having an ISP in the first place would scale infinitely in trying to secure that information, or worse possibly sell that information. An ISP is not a bank, and I could not imagine a scenario where I would treat them as such.
You said "They would have a very good clue of my logical location on a network" so I assumed you were worried about other people knowing your location.
Hmm, I don't get it. Let's say, from an ISP's POV, they'll see from the blockchain that someone did a transaction and they're using them as service provider. Isn't it that most ISP use dynamic IPs, which means it's possible for multiple user to have the same IP? How would they specifically identify that you?
And also, is it impossible for bitcoin to improve it's anonymity by removing the IP infos from the blockchain for future transactions?
LOL! That means you'll end up with a mountain of dildos.
You're missing the point. It isn't about dildos specifically, that was just your example.
It is about tools being developed with that mountain of data that allow much of to be tracked, linked, and identified. It won't be targeted at you specifically (usually not at least), but it will be done in the aggregate and a great deal of identification will drop out. Have you ever posted a bitcoin address online? If so, and if you weren't extremely careful, that address can now be linked to your online identify, perhaps linked to other addresses of yours, and certainly linked to addresses of people who have transacted with you (who may also have posted addresses online, though not necessarily the same one you used) and indirectly (computers are good at this) linked with (many, many) other known addresses.
Here is one small example (though the graphs and tables in the paper are interesting, scary, and worth a look), and this is just the very, very beginning (think "cookies" as a privacy issue in web browsers a decade or more ago):
In this paper we explore this unique characteristic further, using heuristic clustering to group Bitcoin wallets based on evidence of shared authority, and then us- ing re-identification attacks (i.e., empirical purchasing of goods and services) to classify the operators of those clusters.
http://cseweb.ucsd.edu/~smeiklejohn/files/imc13.pdfLOL! It's too long.. I feel drowsy when I'm reading too much text
Ok let's just take my bitcoin address at my sig as an example. It's an address I used long time ago. How would other people know if I own the sender's address that sent btc to that address? How would they know what that transaction is for? How would they know where is my btc now and how much I currently have now?
If someone could post that info, I'd be pro anon coins right away
