I have never implemented a crypto payment sollution for a physical shop... I did include crypto payments in online shops tough...

If you're not afraid of managing your own crypto (for example: if you're ok with exchanging the crypto you earned for FIAT on an exchange), it might be a good idear to start "small": you could buy a hardware wallet with multi coin support, paste some stickers at your door announcing you accept crypto, and if somebody wants to pay using crypto, just generate a new address for them on the pc behind your counter? Same can be done with several other SPV wallets (it doesn't have to be a hardware wallet).

Offcourse, if you want a more "automatic" or a professional looking approach, it might be a good idear to search for POS sollutions... I've seen a couple very professional looking devices popping up in the past... And some payment gateways make sure you're payed in your local currency by automatically converting your crypto to FIAT...

Personally, if i would have a physical store, i'd probably take the first route: just install an SPV wallet or use a dedicated hardware wallet to create addresses (QR codes) whenever a customer asks to pay with crypto... Let them send their funds, wait for a confirmation, finalize the sale and use an exchange to convert crypto to FIAT whenever i need FIAT money... This way your investment is very small, and you're able to gain some experience accepting crypto before going for an automated sollution.

BTW: if you're only talking about accepting bitcoin, you could look into btcpayserver, like the others have already mentioned...

I sometimes use getblocks... They have a free api that's more than sufficient for normal day-to-day implementations. There is a limit on the amount of daily calls tough.

getblock.io (i'm not affiliated with them)

Here's a how-to: https://electrum.readthedocs.io/en/latest/coldstorage.html

An airgapped setup basically means you never enter your seed on an online machine, nor transfer any walletfile that contains any private keys or seed information to an online machine.

• you create a new wallet on an offline machine and only backup your seed on a physical (offline), secure medium
• You export your master PUBLIC key to an online machine (trough QR codes, or trough a safe device) and import it into your wallet on the online machine
• You derive public keys from your master public key, hash them into the address and get receive funds using your online wallet
• When you want to spend the unspent outputs funding the addresses you created in the previous step, you use the online machine to create an unsigned transaction (since the online machine does not contain the master private key, the transaction cannot be signed on the online machine)
• you transfer the unsigned tx to the offline machine for signing, and the signed transaction is transferred back to the online machine for broadcasting

an airgapped setup is usually even more secure than a hardware wallet, especially if you use qr codes to transfer the master public key and the unsigned/signed transactions, you do a "proper" setup of the offline machine and you make sure you pick a trusted wallet. if you don't have a spare computer you can use as an offline machine, you can use a bootable usbstick to boot a linux distro with disabled network interfaces: just boot from the usb stick whenever you need to sign transactions and reboot into your "normal" os for the online parts of the process.

I'd suggest starting with testnet coins in order to get the hang of things... The whole process is doable even for not over technical users, but it does require some basic knowledge. There's a learning curve . I'd say there are gradations in security when doing an airgapped setup, ranging from relatively safe to completely paranoid, but in general, it'll be safer than using an online wallet even if you don't go into "paranoid-mode"

in short: you shouldn't...

You shouldn't blindly trust anything or anyone.

This being said: electrum is open source, anybody with a basic knowledge of python and blockchain technology should be able to verify each and every line of code to find out if the electrum dev's did (or did not) include any backdoors. Offcourse, nobody actually takes the effort to critically review each and every line, as a result, in the past, there were vulnerability's in older versions of electrum, and you'll never be 100% sure the current version is vulernability-free (but at least, you ARE able to verify the code if you really wanted to).
The best idear would be to use electrum in an airgapped setup, that way most vulnerability's won't hurt you, or at least use electrum in combination with a good hardware wallet.

one of the application of a merkle tree is the block header: you need to able to prove the transaction data inside a block didn't change just by looking at the block header... It's impossible to add all transaction hashes to the header, so we use a merkle tree.

This way, when a miner finds a block header whose sha256d hash is under the current target, the block can no longer be changed, since changing any transaction in the block will change the merkle tree and invalidate the block

true... I blurred out some data out of habit... But then i noticed i left the QR code intact, and i didn't blur out the order number in the url... But then again, i was exchanging some spare change that could easily be tracked down to me anyways, an exchange isn't a mixer... I didn't bother to re-upload the images completely unblurred nor did i bother to blur some extra data (and re-upload).

Sounds like a plan... The only tip i'd give you is to look into running electrum on an airgapped machine... If you are going to convert around 1/6th of your monthly income to bitcoin, it'll represent a big part of your funds, and it might be a good idear to keep it as safe as humanly possible.

you'll need this (official) electrum document to set this up: https://electrum.readthedocs.io/en/latest/coldstorage.html
I'd also suggest to test non-"default" setups with tBTC before actually using it on the main chain, that way you can make sure you know how everything works

An other option that hasn't been listed yet: https://chainquery.com/bitcoin-cli

This one lets you try out several commands on their online node...

It's a tad bit difficult to follow your question... But if i interpreted it correctly then: yes, you only need the seed phrase to restore your wallet or you can move the wallet file. When it comes to being able to spend your unspent outputs, both methods yield the same result.

The wallet file potentially contains meta-data (like labels), if you move the wallet file from computer 1 to computer 2, you also get to keep the meta-data on computer 2, if you use the seed phrase to restore your wallet on computer 2, you lose information like labels (unless you use a plugin to synchronise the meta-data)

It's usually the wallet that takes care of this for you... Usually, you can just import the 3 private keys into one wallet and the wallet should take care of the rest... Offcourse, this is the theory, in reality things sometimes go sideways.

Do be carefull when exporting and importing private keys! make sure you do spend some attention to opsec (best only to handle private keys on an offline machine)

EDIT: just on a sidenote, when doing this, i usually use electrum... I just create a new wallet with the wizard, chose "import bitcoin addresses or private keys", then you get a screen to paste your private key(s). Do click the "info" button, since there's a trick to import (for example) (nested) segwit and this is discussed in the "info" menu...

If you have a wallet that generated 3 addresses for you, and you used these 3 addresses to receive funds on 3 different occasions, your wallet will just see 3 unspent outputs it's allowed to send...
If you send an amount bigger than the largest unspent output, your wallet will just create 1 transaction with 2 (or 3) inputs and one (or more) outputs. Do note that the fee you pay is denominated in sat per vbyte of transaction data. A transaction's size depends on the amount of inputs you use and the amount of outputs you create, so the bottom line is that the more unspent outputs you combine in a single transaction, the bigger it will be, the bigger your transaction fee will have to be (but re-using the same address wouldn't have changed this fact, since receiving funds 3 times on the same address will also result in your wallet seeing 3 unspent outputs it's allow to spend).

TL;DR; you don't have to create 3 transactions to spend your 3 unspent outputs, as long as the addresses you received funds on were generated by the same wallet (it's usually even possible to combine unspent outputs funding addresses that were not generated by the same wallet, but this requires quite a bit of expertise)

To add to Charles-Tim's reply: it's also a good idear to look at the "Tx" column inside electrum and use an address with Tx count 0 (= addresses that were never funded).
This way, you increase your privacy level... If you re-use your address, it decreases your privacy...

Also, once you spend funds from an address, you expose your public key. This is not a big issue, but there have been arguments in the past about a very tiny decrease in security funding an address that was already spent from and thus has it's public key exposed... The bottom line is that you shouldn't worry about this to much, since those discussions were merely theoretical...

But still, don't re-use addresses for privacy reasons

I tested out eXch.cx... Here's my walktrough:

Calculating the exchange rate:



Creating an order:



Letter of Guarantee:



After  confirmation:



My Wallet:



So far, i can say that the whole experience was painless... The exchange rate was good for a no-KYC swap site... Everything went fast and smooth

Edit: I posted my review of this campaign in this post... But it should have gone in the discussion thread... Sorry!

I cannot delete this post for the next 24 hours due to board restrictions, but i'll try to remember deleting it tomorrow.

As for now, i posted my review here: https://bitcointalk.org/index.php?topic=577207.msg62918160#msg62918160

I'm having a bit of spare time lately, and i'm thinking about filling it up by offering some scripting services... I'm not a very big expert, but i do have some experience (i have many more finished projects, but most of my scripts are not on my public repo, since they were written for my employer and they're part of his intellectual property)... Working on a no cure no pay basis...

I can write in

• golang: these are older projects, my programming style did improve lately : demo project 1:my old website, demo project 2:a simple apache CLF logfile parser, demo project 3:a tool to run in your crontab or scheduled tasks to alert you when the optimal tx fee drops under a certain target fee
• python: demo project 1:a tool to check how far a certan tx is from a coinbase reward (written after a discussion with a mixer), demo project 2:my tool to check for unspent outputs on bitcoin forks (like bitcoin cash)
• php: demo project 1:a POC php function to deserialise segwit transactions, demo project 2:a VERY simple php testnet faucet, demo project 3:A site i coded in php/mysql

I'm also proficient in relational database design and my daytime job is being a unix system engineer....

I'm looking for smaller automation projects that i can do for cheap in a couple of hours... Need someone to write a script to parse a webpage and shoot you a mail when a certain event arises? A go program to provide a simple API to your relational database? A script to automate a small repetitive task? A very simple and basic homepage without a nice layout? A small database design? Reply in this thread or shoot me a PM
I prefer writing in golang, but if you really insist on python or php that would work aswell.

I'm currently not looking for bigger projects... If something would take more than a couple of hour of work, i'll probably decline. Sad to say, but most crypto-related programming would require more time than i'm willing to spend right now.

Payment only in "real" crypto, the one with a blockchain behind it and a decent market cap and exchange support (BTC, ETH, LTC,...), payment either upfront (with a guarantee of a refund if i cannot complete your project), using an escrow of your choice (you pay the fee) or after the project finishes (if you have a long and positive bitcointalk reputation).

I opted for trusting my daughter... I only have one anyways... I have 2 hardware wallets which i keep in a safe place... They both have the same seed, the same pin and i used the same extension word... My daughter knows where i've locked up my hardware wallets and she knows where she can find the extension word and she knows my pin by heart. She knows how to use my wallets as well. If i die, i don't want her to have to start a big puzzle restoring a seed and importing it into a wallet. I want her to plug in a ready made device, punch in a 8 number pin code, take the piece of paper where the extension word is written on and take control over my money.

If she robs me, she robs me... I'd rather have my only child robbing her father than dying and not being 99% sure she'll be able to spend my hard earned money.

I'll try to simplify the things a bit... Let's take a step back into a simplified version that disregards many of the key aspects of bitcoin....

I'm making a payment system between me an my friends, in order to do this, we have an excel-sheet, we each put $100 in a jar and we add 3 records into our excel-sheet:

 1  friend1  $100  unspent  initial block reward
 2  friend2  $100  unspent  initial block reward
 3  friend3  $100  unspent  initial block reward


I'm friend1, i now send $50 to friend2... The exelsheet looks like this:

 1  friend1  $100  spent  initial block reward
 2  friend2  $100  unspent  initial block reward
 3  friend3  $100  unspent  initial block reward
 4  friend3  $50  unspent  used line 1
 5  friend1  $50  unspent  used line 1


Now friend3 uses line4 to send $25 to friend2

 1  friend1  $100  spent  initial block reward
 2  friend2  $100  unspent  initial block reward
 3  friend3  $100  unspent  initial block reward
 4  friend3  $50  spent  used line 1
 5  friend1  $50  unspent  used line 1
 6  friend3  $25  unspent  used line 4
 7  friend4  $25  unspent  used line 4


After a while, we decide to implement a hard fork... We come together and say: from now on, we'll no longer use dollars as the payment amount... Starting from line 8, we'll switch to euro's (at an imaginary exchange rate of $1 = €1)... When i now send my $50 to friend2, the ledger will look like this:

 1  friend1  $100  spent  initial block reward
 2  friend2  $100  unspent  initial block reward
 3  friend3  $100  unspent  initial block reward
 4  friend3  $50  spent  used line 1
 5  friend1  $50  spent  used line 1
 6  friend3  $25  unspent  used line 4
 7  friend4  $25  unspent  used line 4
 8  friend2  €50  unspent  used line 5

However, friend3 does not want to follow our new rule.. He insists every amount has to be entered in dollars... We now have a hard fork: friend1, friend2 and friend3 agree on the first 7 lines... But only friend1 and friend2 agree on the transaction in line 8, friend3 only agrees on the first 7 lines, he rejects line 8 and he'll be able to add his own transaction on line 8, making this a hard fork... We'll continue with 2 versions of our document: one that uses euro from line 8 and higher, and one that has always and will always use dollars...

If we kept working in dollars, but we decided to add a new extra worksheet in our excel document keeping track on metadata about our transactions, we could have created a soft fork: the people that wanted to use the extra sheet to add metadata and the people that did not want to use this extra sheet all agree on the rules about the first worksheet, so they can both keep adding their transactions in one document... It's only that some of the friends will use the second worksheet aswell...

Offcourse, in the situation with 3 friends putting fiat money into a jar, this seems ridiculous, but in a situation with thousands of anonymous users and coinbase rewards, it does "work"... All users agree on the blockchain untill the height of the fork, and at this height, the two blockchains grow apart.

Just FYI: this is not exactly how bitcoin works... It's just an (over)simplification about how hard forks and soft forks more or less work when mapped to an "easyer" problem.

I was coming here with about the same info tranthidung and ETFbitcoin already gave, with the possible extra info that wasn't explicitly posted by tranthidung, but was already discussed previously: so it's clear to everybody: yes, the OP is right, cloudflare IS a MITM... Packages get encrypted by a symetric key between your workstation and cloudflare, cloudflare decrypts and checks it's cache, if no cache hit, cloudflare sets up a new symetric key for communication between their servers and (in this case) the server running bitcointalk. There isn't even any 100% proof there is a symetric key used for the connection between CF and bitcointalk's host, but i'm willing to take Theymos on his word, since he has no incentive not to set this up on his site. This means no network operator will be able to capture and read your traffic (since it's encrypted), but cloudflare will be able to access (and store) every unencrypted package you send to bitcointalk.

I've actually spent a lot of time on this subject in this post: https://bitcointalk.org/index.php?topic=5247838.0 (EDIT: link was already posted by ETFbitcoin) and even tough the images no longer work, the theory can still be applied to bitcointalk aswell... But like i said: back when CF was implemented, this was discussed by a lot of members (including Theymos), and the usage of cloudflare was more or less accepted by most users at that time (eventough we all knew the obvious MITM attack vector that was introduced at that time).

In the past, there was a discussion of also setting up bitcointalk as a hidden service for people that wanted to mitigate cloudflare, but the proposal also had it's downsides and it didn't gain any traction.

If you want another example, i used to mine coin magi (in a very distant past)

https://github.com/m-pays/magi#intro

9 years ago, there already was a coin that combined POW en POS

Users can already set any fee they want (if they're using a more or less decent wallet), and miners already pick which transaction they include in the block they're trying to solve...

Offcourse, at the moment, some nodes started to reject transactions with a fee that's to low since their mempools are already full with unconfirmed transactions.