Mycelium Bitcoin Wallet

[roslinpl]

 Mycelium Bitcoin Wallet I like your product! I gave it 5 stars! Keep updating!

[Newar]

Regarding the spend from cold storage feature, if I spend some but not all of the funds on a paper wallet, where does the change go?

Back to the paper wallet.

[Jan]

Hi MyCelium people, good work on trading stuff but I heard you're using a tx radar for giving confidence to traders who receive BTC.

I think this is an extremely bad idea, as txs can be replaced after 100% propagation has been achieved by sending higher fee txs to replace the one you made. This attack only needs to work 1% of the time to be profitable but in reality can work much more (maybe even 100%) if pulled off correctly. You don't even need to tell the whole BTC network, just a few specific miners.

For instance if I submit my replacement tx to eligius using their web interface, then make the trade before eligius mines a block, if eligius mines a block it won't include the tx that got 100% propagation because that would've been rejected by eligius.

I suggest you remove this feature, and only using it for senders of BTC (not recipients).

The localbitcoins thing otherwise looks excellent. Thanks for this contribution. I don't mean to beat on this parade and be so negative, I like your work.

Thanks for your input Amir. The transaction confidence is not only based on network propagation, and never reaches 100%.
Here is what we look at:
- network propagation. We are listening to 500+ nodes
- length of the chain of unconfirmed transactions. If you tx depends on one unconfirmed output the confidence is halved. Halving for each link in the chain
- size of the fee of this tx and whatever unconfirmed tx it depends on
- Rela-time double-spend detection. If we detect one the confidence goes to zero.

We boil that down to a confidence number which never reaches 100%, and let the user take it into consideration when figuring out whether he wants to spend the time it takes to confirm with the seller. In the end it is an instrument that the buyer can use to help him judge the situation along with a number of other factors. Maybe he knows the seller, or he can see he has a five star reputation and 100 deals under his belt. He might also consider the seller a dodgy character and wait it out.

In the end there is no 100% guarantee, not even when the transaction has confirmed.

[Jan]

Wow, this really took off on reddit: http://www.reddit.com/r/Bitcoin/comments/1xq6wq/ann_myceliums_new_local_trader_feature_will_let/
Please up vote :-)

[Portnoy]

Hi MyCelium people, good work on trading stuff but I heard you're using a tx radar for giving confidence to traders who receive BTC.

I think this is an extremely bad idea, as txs can be replaced after 100% propagation has been achieved by sending higher fee txs to replace the one you made. This attack only needs to work 1% of the time to be profitable but in reality can work much more (maybe even 100%) if pulled off correctly. You don't even need to tell the whole BTC network, just a few specific miners.

For instance if I submit my replacement tx to eligius using their web interface, then make the trade before eligius mines a block, if eligius mines a block it won't include the tx that got 100% propagation because that would've been rejected by eligius.

I suggest you remove this feature, and only using it for senders of BTC (not recipients).

The localbitcoins thing otherwise looks excellent. Thanks for this contribution. I don't mean to beat on this parade and be so negative, I like your work.

Thanks for your input Amir. The transaction confidence is not only based on network propagation, and never reaches 100%.
Here is what we look at:
- network propagation. We are listening to 500+ nodes
- length of the chain of unconfirmed transactions. If you tx depends on one unconfirmed output the confidence is halved. Halving for each link in the chain
- size of the fee of this tx and whatever unconfirmed tx it depends on
- Rela-time double-spend detection. If we detect one the confidence goes to zero.

We boil that down to a confidence number which never reaches 100%, and let the user take it into consideration when figuring out whether he wants to spend the time it takes to confirm with the seller. In the end it is an instrument that the buyer can use to help him judge the situation along with a number of other factors. Maybe he knows the seller, or he can see he has a five star reputation and 100 deals under his belt. He might also consider the seller a dodgy character and wait it out.

In the end there is no 100% guarantee, not even when the transaction has confirmed.

Thanks for the quick response to this, and any and all other concerns and questions people have in regard to your wallet and other services.   

[westkybitcoins]

That's a really cool feature.  I love this wallet.  Now if only I could sign messages with it...

I implemented message signing for mycelium. It is very very likely that it is part of the next feature release.
We also use signing for other stuff internally

You'll get both, the cool feature, and message signing

Well doesn't that just beat all? 

Indeed.

I love this wallet.

[MarcoPon]

Please consider adding BitcoinAverage as a price source, and possibly as the default one - that would automatically resolve any temporary issue with any exchange site.

[mstevenson1972]

"Lets talk bitcoin" said this was already active but I can not find the “Transaction Confidence” graph and no update in google play app?

[apetersson]

"Lets talk bitcoin" said this was already active but I can not find the “Transaction Confidence” graph and no update in google play app?

no it is not active yet. we will publish testnet versions soon™

[Newar]

Please consider adding BitcoinAverage as a price source, and possibly as the default one - that would automatically resolve any temporary issue with any exchange site.

This was suggested before... and rejected as they do not want to be dependent on a third party.

[MarcoPon]

This was suggested before... and rejected as they do not want to be dependent on a third party.

Oh, didn't know or remember. Thanks!

[apetersson]

as requested, we now have message signing.

live on testnet and github. if YOU (our testers) give us the feedback, we can start to roll that out soon.

https://plus.google.com/107839384289577985803/posts/PTQ9kgWnqBK

https://github.com/mycelium-com/wallet/releases/tag/v1.1.7

if possible, test message signing with bitcoin-qt in testnet and with other clients that have testnet support.

[Jan]

Next release will feature Coinbase, BitPay and Kraken as exchange sources (in addition to Bitstamp, BTC China and BTC-E). MtGox however is going out.
Would you be upset if MtGox is no longer available?

[ZephramC]

Next release will feature Coinbase, BitPay and Kraken as exchange sources (in addition to Bitstamp, BTC China and BTC-E). MtGox however is going out.
Would you be upset if MtGox is no longer available?

I would not miss it.

[MarcoPon]

Would you be upset if MtGox is no longer available?

Absolutely not!

[RGBKey]

Do you plan on supporting BIP 70-73, and when?

[westkybitcoins]

Would you be upset if MtGox is no longer available?

Absolutely not!

+1

[bigbeninlondon]

Would you be upset if MtGox is no longer available?

Absolutely not!

+1

I personally wouldn't miss it, and I definitely think it should not be included when calculating the averages, but until it has truly tanked, some of your customers may want it in their option list.

[TheButterZone]

Next release will feature Coinbase, BitPay and Kraken as exchange sources (in addition to Bitstamp, BTC China and BTC-E). MtGox however is going out.
Would you be upset if MtGox is no longer available?

Nope. Permaban, as they've had so many strikes, they should be out of bitcoindom until the heat death of the universe.

[Rassah]

Do you plan on supporting BIP 70-73, and when?

Planning eventually, since everyone will be using it eventually, but not any time soon. First we have to finish LocalTrader, then we need to add BIP32 HD wallets, then we can focus of 70-73.