I've already found one for you. No need to go anywhere.
http://www.bluenile.com/emerald-diamond-1-carat-or-less-good-cut-g-color-if-clarity_ld01435822?rf=pd

Not being able to make up one's mind on what a certain subject is or isn't kinda hints at how much of an expert one is in that subject.

So... this thread got my imagination running...

Not so far future... You get notified that your uncle you didn't even know about died and your presence is required for reading of his will. You soon find out that your uncle was one rich bastard but all he left you was a small diamond. So you decide to see how much the diamond is worth and take it to a specialist. While looking at the diamond the specialist notices there is an engraving on one of the sides of the diamond. Looking closely you realize it must be a QR code. The QR code is your uncles private key for his wallet containing hundreds of thousands of bitcoins. All in a small diamond

You could try to sue them, but finding the real address would be quite a pain, I imagine. This also would be one of the things to look at before signing up as a customer.

Nothing stops you from building a p2p network for messaging. There's really no need to build stuff into the client. And even if you think that being built into the client is a must, you can make your own client with the p2p exchange.

Hopefully we can get back on topic now.

Imagine how glad the customer will be when she has to skim through a hundred of these labels just to find out whether or not you accept megawallet. She will be even happier to see a simple slot for coins.

Are you looking at this from a merchant's perspective, from a bank's perspective or from a bank customer's perspective? The customer doesn't deal with the payment processor. For the merchants it's no different than putting your faith into mtgox. For the banks - this non-profit would need to earn their trust before they put a bit of funds in it. There's no need to transfer all of the funds to the payment processor, only an amount that is needed for the time that it takes the network to confirm all "instant" payments made from such a bank. In other words as the bank's client requests an insta-transfer, the bank connects to the payment processor and requests the same amount of funds to be transfered to the recipient specified by the bank's client. If the bank's account that is managed by the payment processor has enough money, the payment processor issues a transaction, if not - an error is returned. The bank can transfer some funds to the payment processor whenever it feels like and those will be used to issue transactions on behalf of the bank.

And if you bought an exclusive license to a digital product? A good example would be exclusive web-site templates - some of those cost thousands. Another is software. I suppose the wait doesn't matter that much in these cases, and people would be fine with waiting for some time, but only if bitcoin was the only way to pay for these things. More likely than not people would prefer paypal's instant service to waiting for a while. Besides, I don't think 6 confirmations takes just 10 minutes - it took an hour or so when I checked last (was some time ago, though).

I suppose it is, but IIRC even 0 confirmations takes time because of the decentralized nature of the network.

You would trust instawallet.org, but what would you do if a considerable share of customers was using fancywallet.com, bitwallet.org and a hundred more instawallet lookalikes?

Your idea fails here. First you can only trust a minor amount of coins which makes buying expensive things impossible, second if you're a merchant you have no way of finding out if you really were sent coins or not, it's the same as relying on the customer saying "I've sent you the coin, you'll receive it later, can I have my goods now?"

The best approach I could come up with to date is to create a central bitcoin payment processor for banks. This payment processor would have to be a non-profit organization and would ensure an even playground for all the banks wrt instant payments. The banks would store some money with this payment processor and as soon as a bank issues a transfer using the payment processor's api, the payment processor can confirm a transaction. So basically this would be a bank for banks.

He's trying to manipulate the market.

Suppose they become the main bitcoin bank, and if they offer instant transaction confirmation already this will attract merchants and users. If there aren't any competitors, they could one day decide to charge you 10% on all transactions. And so you have a choice: either have instant transactions that cost 10% more or not have instant transactions. Now, this is a very far-fetched situation, and more than likely there will be more banks in the future offering instant transaction confirmations. Everyone wants to jump on the band wagon once they know that it's profitable, don't they. So more than likely is that we'll have more bitcoin banks popping up that offer the same basic service to merchants - instant confirmations. Now here's where it becomes a problem: as there are many banks, they compete with each other for users, so the market is fragmented (especially considering an almost nonexistent barrier of entry). For users that might not be so much of a problem as it is a benefit, but think about merchants for a bit and you realize that it's beneficial for merchants to work with as many banks as possible that would allow them to receive instant payments. It would be a huge pain in the ass to review each bank, not to mention that it's also a risk to rely on this instant confirmation before the transaction is actually confirmed by the network. Now lets forget about these issues for a moment and focus on the experience of a user that tries to pay with bitcoin: the user clicks a "pay with btc" button on the website and is shown a page with a list of banks to choose from, now the user has to search through the hundreds of banks on that page just to fucking make a payment. Compare this with paypal or google-checkout and it's obvious that the user will prefer other payment options to bitcoin.

Got your attention now, didn't I?  

Anyway, I'd like to talk about one of the disadvantages of bitcoin over other digital currencies. Non-instant payments. In today's world we're used to instant payments. We go to the mall, pay with our credit card at the checkout and we're done, we use various digital currencies and payment systems online to again make payments in seconds. This is not possible with bitcoin even with 0 confirmations, and "0 confirmations" approach is quite insecure to be widely used. So in order for instant payments to be possible we need to have the payment confirmed in a faster way. The only one I can think of is by having trusted parties managing the money, thus when a trusted party confirms the money has been sent, you can be sure you'll receive a payment. In effect, these trusted parties are basically banks. However, because it's very easy to set up such a bank, because it takes few resources, anyone can create such banks. And this leads to the main question of this topic: how do we make sure fraudulent banks do not get included in the system? Or perhaps I'm going about this the wrong way entirely? What are your thoughts?

The bigger question is what do you do when a trusted exchange suddenly decides to take all your bitcoins and disappear? Which leads to another question: how do you know an exchange can be trusted?

Until you involve a few lawyers from a few countries that could answer those questions, you can't really consider any exchanges to be trustworthy no matter how secure they claim to be.

If gold is so good for savings, why have any fiat currency at all? Why not save your money in gold from the very start right until you need the money and only then convert them into fiat? It would be stupid to have always inflating fiat currency when you can have always deflating gold/bitcoin, wouldn't it? The problem with gold lies in the conversion to/from fiat currency. It's either too costly or too cumbersome. If it was easy you would prefer to store all your money in gold, and only convert into fiat currency when you need to. In a trade though, you would prefer to pay with gold because converting isn't free and involves a third party.

Bitcoin, however, is not gold, in the sense that it's very easy to convert to and from fiat currencies. The problem with bitcoin is in the risk of having it. You don't know what the price of bitcoin will be tomorrow, whether it's going to be 2x its price today or 0. And until you do you'll want to store some of your money in bitcoin, while still keeping most of it in fiat currency, since you don't want to lose everything in one day because you were storing it in bitcoins, but you also don't want to be the loser that didn't invest into bitcoin early and become rich with little effort. In other words if one wishes to solve the hoarding problem of bitcoin, one needs to up the chances of bitcoin becoming mainstream. Built-in deflation has nothing to do with hoarding.

Bad idea as a means to popularize bitcoin. Don't know if it's any good as a means of selling a digital product (likely not)

The best idea ever on these forums was paying bitcoins for getting merchants to provide bitcoin payment option. I think it should be expanded upon (one could make a website where some could donate bitcoins for the purpose and others could earn them by getting a website to accept bitcoins).

It might if you build it from source.

Your bitcoin client facilitates finding bitcoin clients even more

Ahh... I missed that somehow

mt.gox, you need to fix your charts. The dots for the same exchange rates are not on the same line.

Umm... as far as I understand, you really can't collect bitcoins. All that the system keeps are transactions between peers.

Hacking always seems to work.

I think you should first define your priorities. Why would anybody want to use your OS? Surely not just because it has a few programs installed by default?