Will block creation stall if too many miners suspend operation?

[coinableS]

For a cup of coffee a merchant probably doesn't need to wait for a confirmation but a retailer selling a TV or PC or car, or a person or ATM exchanging fiat will probably want to wait for a confirmation or two.

Why? Do you know many people that are able to execute raw transactions and double spends on their phone?  If they are at a retailer they will likely only have their phone or a paper wallet on hand as a method of paying.

[stdset]

For a cup of coffee a merchant probably doesn't need to wait for a confirmation but a retailer selling a TV or PC or car, or a person or ATM exchanging fiat will probably want to wait for a confirmation or two.

Why? Do you know many people that are able to execute raw transactions and double spends on their phone?  If they are at a retailer they will likely only have their phone or a paper wallet on hand as a method of paying.

If a Bitcoin knowing criminal finds such a retailer, who sells something expensive and doesn't wait for a single confirmation, the criminal doesn't need to send a doublespend from his phone. For example his accomplice could do that, from e.g. a PC. It's better to make the first transaction as low-priority as possible, and include considerable fee into the second. The criminal could also arrange a deal with a thievish pool operator: establish a direct connection from the PC to the pool, so that the second transaction is guaranteed to be received. The pool ignores the first transaction, confirms the second. In the case of success the criminal shares profit with the pool owner. The first transaction is unlikely to be included in several next blocks, so the criminal has enough time to finish paperwork with the retailer and drive away, he could even delay the second transaction release until he leaves.

[dwdoc]

Per above it would probably be a good idea to educate merchants that they should require high end transaction fees for zero confirmation sales.

[DannyHamilton]

For a cup of coffee a merchant probably doesn't need to wait for a confirmation but a retailer selling a TV or PC or car, or a person or ATM exchanging fiat will probably want to wait for a confirmation or two.

If a Bitcoin knowing criminal finds such a retailer, who sells something expensive and doesn't wait for a single confirmation, the criminal doesn't need to send a doublespend from his phone. For example his accomplice could do that, from e.g. a PC. It's better to make the first transaction as low-priority as possible, and include considerable fee into the second. The criminal could also arrange a deal with a thievish pool operator: establish a direct connection from the PC to the pool, so that the second transaction is guaranteed to be received. The pool ignores the first transaction, confirms the second. In the case of success the criminal shares profit with the pool owner. The first transaction is unlikely to be included in several next blocks, so the criminal has enough time to finish paperwork with the retailer and drive away, he could even delay the second transaction release until he leaves.

Why would a thief go to all that trouble?  Wouldn't it be easier to just pick up the box and walk out of the store? Or in the case of a car, just take it out for a test drive and don't come back.

There is no benefit to going through all these 0-confirmation false transaction efforts.

[stdset]

Why would a thief go to all that trouble?  Wouldn't it be easier to just pick up the box and walk out of the store? Or in the case of a car, just take it out for a test drive and don't come back.

There is no benefit to going through all these 0-confirmation false transaction efforts.

Looks like you prefer good and old physical violence? Sure, if you have enough firepower, you don't need all that complicated modern IT stuff.

[hhanh00]

Do miners replace a transaction with one that has a greater fee? I think they don't. In which case, a merchant just has to wait until his transaction reaches enough pools to ensure inclusion in the next block.

[dwdoc]

Miners are more likely to include a transaction with a higher fee. Therefore if a customer transacts with a merchant using a small transaction fee and broadcasts this to one set of nodes and then immediately pays himself the same bitcoins with a higher transaction fee which he broadcasts to a different set of nodes, the merchant may end up with nothing. 

[DeathAndTaxes]

Miners are more likely to include a transaction with a higher fee. Therefore if a customer transacts with a merchant using a small transaction fee and broadcasts this to one set of nodes and then immediately pays himself the same bitcoins with a higher transaction fee which he broadcasts to a different set of nodes, the merchant may end up with nothing. 

Any merchant accepting zero confirm txns would be smart to have a large number of listening nodes which detect the second txn and then halt the purchase.   

[hhanh00]

Miners are more likely to include a transaction with a higher fee. Therefore if a customer transacts with a merchant using a small transaction fee and broadcasts this to one set of nodes and then immediately pays himself the same bitcoins with a higher transaction fee which he broadcasts to a different set of nodes, the merchant may end up with nothing. 

Of course, but you are not answering my question.

[stdset]

Do miners replace a transaction with one that has a greater fee? I think they don't. In which case, a merchant just has to wait until his transaction reaches enough pools to ensure inclusion in the next block.

Normally they don't, they shouldn't even relay a doublespend. But modified, 'greedy' miners could do that. Also, nothing prevents a corrupted miner from colluding with a criminal to accept and confirm a particular doublespend while rejecting all other doublespends like honest miners do.