Unicode 9.0 will be released in June, so you probably won't see official fonts released before then. If you just want a font to play around with, luke-jr has made a font with the Bitcoin symbol in its official place here: https://www.reddit.com/r/Bitcoin/comments/3refb4/btc_symbol_added_to_droid_and_console_fonts/

In any case, thank you to the bitcointalk.org participants who helped me with the proposal - the original thread is here if you're curious: https://bitcointalk.org/index.php?topic=1170902

Ken

Thanks everyone in this thread for your support and helping get BTC into Unicode! Special thanks to runpaint for digging up a bunch of usage examples - that made a big difference.

Ken

I would like to get the opinion of native Thai users.  What are the key features of the Baht symbol: are two lines okay, does the line have to go all they way through, etc? Do you consider BTC and ฿ to be the same character or different? What do you think of using ฿ for Bitcoin?

These may seem like strange questions, but the answers are important for font design.

Thanks,
Ken

Thanks. I'll add Digitalcoin Foundation to the supporters. Any other organizations that are in favor this proposal: just let me know.

Ken

Unicode is the standard that most computers use to define what text characters exist and how to encode them. In the bad days before Unicode (e.g. the 1990s), different computers used different characters and stored them different ways. If I used Russian and you used Japanese, we couldn't exchange files. But with Unicode, everyone's computer handles characters the same way. I can send you Ӂ ڠ ᅗ Æ א ₳ ☢ 五 and your computer can handle it (as long as you have the fonts installed).

The point is that computer companies aren't going to patch random characters into their OS. But if you get a character added to Unicode, then everyone will support it. The Unicode Committee has the people who decide what goes into Unicode. The purpose of my proposal is to convince them that putting the Bitcoin symbol into Unicode makes sense and isn't just some crazy idea I came up with.

(I've oversimplified somewhat but this is probably already more than you want to know about Unicode.)

Ken

Lol. Yes, my previous Unicode proposal was for the group mark symbol used by IBM mainframes in the 1960s. It was approved, so that symbol will be in the next Unicode release. You may be interested in my related article on mining Bitcoin on a 1960s mainframe: http://www.righto.com/2015/05/bitcoin-mining-on-55-year-old-ibm-1401.html


That explains things. For those just joining, my proposal to add the Bitcoin symbol is at http://righto.com/bitcoin-unicode.pdf and is much more extensive than the unrelated rejected 2011 proposal.

Ken

Hi runpaint. I already have some multi-lingual running text examples in the proposal, but you've found some new languages, so thanks for finding them. The Greek was is especially interesting - I had looked for one in Greek, but didn't find it. I'll add some of your examples to the document.

At this point, there are probably enough examples, so you don't need to spend more time looking.

Thanks again for your help,
Ken

Thanks, runpaint for finding those examples! I'll add them to the proposal.

Ken

Your support here is helpful since it shows the "Bitcoin community" is mostly in favor. So thanks, tspacepilot. Petitions can help for new emojis, but for text characters the Unicode committee mostly looks for examples of the character being used in "running text".  That is, the character needs to be part of text, not an icon or logo. I think I have enough examples, but if you come across the Bitcoin symbol being used in books or magazines, let me know.

Ken

LiteCoinGuy, Za1n, Mickeyb, Rustynomad, runpaint, etc: Thanks for your support.

goosoodude, the character I successfully added to Unicode before is the "group mark symbol", used by IBM computers in the 1960s. http://www.unicode.org/L2/L2015/15083-group-mark.pdf

jt byte: My proposal covers BTC not ฿.

wearehatetherules: I haven't submitted the proposal yet. The Unicode Committee's next meeting is in November.

Ken

Yes, the wiki page is referenced in my proposal. The problem is that BTC isn't in the Unicode standard, so you can't easily use it and need to use a workaround like ฿ or Ƀ or a custom font. By adding BTC to Unicode, the character can be conveniently used everywhere. (This forum uses a special font that allows it to display the symbol.)

I've written a proposal for the Bitcoin symbol BTC to be added to the Unicode standard. This would make it much easier to use the BTC symbol in text.

The proposal is at http://righto.com/bitcoin-unicode.pdf. I've successfully proposed the addition of a symbol to Unicode before, so I'm familiar with the process.

Take a look and let me know of any errors. If you are known in the Bitcoin community or represent a company, please let me add your endorsement to this proposal - your support will help sway the Unicode committee.

Thanks,
Ken

Thank you for your reply, but I'm asking something much more obscure that I noticed after studying transactions in hex for hours. As you say, since a coinbase transaction isn't spending anything, it doesn't need a scriptSig and arbitrary data can be put in the coinbase field.

But I'm talking about the scriptPubKey in the output. A normal transaction puts the public key hash here, but a coinbase transaction puts the actual public key here.

That is, the scriptPubKey of a normal transaction is:
OP_DUP OP_HASH160 addr OP_EQUALVERIFY OP_CHECKSIG

But the scriptPubKey of a coinbase transaction is different:
public-key OP_CHECKSIG

A consequence is that to spend something from a normal transaction, the next transaction will have the scriptSig:

But to spend something from a coinbase transaction, the next transaction will have the scriptSig alone without the pubkey:


As far as I can tell, there is no reason for the two transactions to be handled differently. I expect you could do either transaction either way.

My question is if there is a reason for these two different scripts? Or is it just an arbitrary implementation detail?

This is the place for feature requests? I request that the transaction page should indicate non-canonical scripts. This would make it much easier to detect transactions that are suffering from transaction malleability.

Desired behavior: There should be an indication under "Input Scripts" that the transaction is using OP_PUSHDATA2 in a nonstandard way if I go to https://blockchain.info/tx/bba8c3d044828f099ae3bc5f3beaff2643e0202d6c121753b53536a49511c63f

As it is, even the rawtx page doesn't show anything wrong.

A technical question: Why does a coinbase transaction put the public key itself in the scriptPubKey, while a regular transaction puts the public key hash (address) in the scriptPubKey? (And as a consequence, the scriptSig to redeem a coinbase transaction doesn't include the public key, unlike the scriptSig to redeem a regular transaction.)

Is there a motivation behind this, or is it an arbitrary implementation detail?

Reference: https://en.bitcoin.it/wiki/Script#Standard_Generation_Transaction_.28pay-to-pubkey.29

A bit of a technical question: when computing/verifying the signature on a transaction, why is the scriptPubKey script from the previous transaction copied into the transaction being signed? Obviously the scriptSig is not available at the time of signing, but why not just leave the script section blank?

Since the hash of the previous transaction is already included in the data being signed, I don't see any security benefit from doing this. It seems like a pointless complication to me, but I assume there's a reason. I've studied the wiki page https://en.bitcoin.it/wiki/OP_CHECKSIG but I can't figure out the motivation behind this. Anyone know?

Thanks.