Re: [bitcoin-dev] Let's deploy BIP65 CHECKLOCKTIMEVERIFY!

["Clément Elbaz" <clem.ds@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1623 bytes --]

I fail to see how the number of confirmations has anything to do with it.

With a non-upgraded Bitcoin software during a soft fork, you get the same
blocks as everyone else, and you get the same confirmed transactions as
everyone else. So you do have the exact same "writings" as everyone else to
calculate your balance.

The problem is that some transactions that are meaningless to you are
actually meaningful to people using an upgraded Bitcoin software.

Therefore during a softfork, while you can not miss the *existence* of a
transaction, you can miss its *meaning*.

If Bitcoin was just a decentralized whiteboard for people to write on it,
that would be no problem.

But as soon as you try to actually use Bitcoin (that is, calculate the
accurate balance of a wallet in a very broad sense), you can be led a wrong
result if you did not upgrade, which is a critical problem for financial
software.

And because nothing prevent people to send you transactions of a new type,
you have no way to "opt out" of this problem.



Le lun. 5 oct. 2015 à 14:16, Jorge Timón <jtimon@jtimon.cc> a écrit :

>
> On Oct 5, 2015 2:08 PM, "Clément Elbaz" <clem.ds@gmail.com> wrote:
> >
> > It will get correct results about :
> > - the existence every block
> > - the existence of every transaction
> >
> > It will get incorrect results :
> > - about the nature of some transactions
>
> Given the assumptions above, only of transactions without enough
> confirmations.
>
> > - and therefore, about the balances of some wallets.
>
> Not if the wallet waits for enough confirmations.
>

[-- Attachment #2: Type: text/html, Size: 2165 bytes --]