The biggest issue, in my view, is that you cannot have both scalability and secure offline transactions in a system like Bitcoin -- specifically, in any system that lacks currency issuing authorities. Chaum showed this long before there was a Bitcoin (sorry if this is not an article you can easily find for free):
In layman's terms, the result is this: every time you make a secure digital cash transfer, the amount of information needed in future transfers of the money increases. The increase is, in the best case, linear in the number of offline transactions. Traditionally, that is solved by having a "bank" that issues the initial information, and which will re-issue currency to reset its size.
In the case of Bitcoin, there is no such authority, meaning that either you'll hit this scalability problem or you cannot have secure offline payments. In Bitcoin's case, solutions have been given for the scalability problem; it is safe to conclude that offline payments are impossible.
Lacking offline transactions is a big deal in practice. It means that we cannot meet anywhere "off the grid" and expect to be able to transfer money. It means that gas pumps need reliable Internet connections. It means that hot dog stands need Internet service. In other words, it means that Bitcoin cannot replace paper money -- whereas Chaumiam-style systems could, or at least have a stronger case for being able to.
There is also reason to doubt Bitcoin's security model. See, for example, this work on analyzing the publicly available information in Bitcoin to de-anonymize its users:
Really, if I were to be as conservative about security as possible, I would say that the only security Bitcoin provides is protection against double-spending in online transactions; I would not rely on anonymity, and I would not even think about offline payments. Considering how much work was done in the early 90s on digital cash systems that not only protect against double spending, but which also allow secure offline transactions and which ensure the anonymity of users, I don't think it is unfair to say that Bitcoin has technical weaknesses compared to other systems.
> Lacking offline transactions is a big deal in practice.
If lack of offline transactions is bitcoins biggest flaw, I am going to invest all my money to bitcoin. Come on, even if bitcoin gains 1% market share in e-commerce or in a niche such as online gambling, it would be a very big deal. Nobody cares if it can't be used for offline transactions if it can be used to circumvent the gambling regulations.
> I don't think it is unfair to say that Bitcoin has technical weaknesses compared to other systems.
Are these other systems used anywhere? Like a gambling startup that made over $400 000 in profit last month.
"Nobody cares if it can't be used for offline transactions if it can be used to circumvent the gambling regulations."
In other words, the future of Bitcoin is in the black market. Which means that the police are just going to look for people who are engaged in Bitcoin transactions, and particularly those engaging in large volume transactions, and they will just happen to be patrolling in that person's neighborhood.
"Are these other systems used anywhere?"
The timing was bad; David Chaum tried to create a company that facilitated digital cash in the 90s, but faced so much trouble and so much resistance by the US government (which at the time was actively thwarting cryptography deployment on the Internet) that his company ultimately went bankrupt. There is nothing stopping a digital cash revival if you want to start it up again -- smartphones could add something interesting to that mix, like phone-to-phone payments (and unlike Bitcoin, you could do this in a way that does not require either phone to have Internet service -- so people could transfer money in New York City's subways, for example).
"Offline transactions are possible in the near future with a semi trusted third party"
I may be mis-reading that, but it looks like that is not an offline transaction, but a transaction facilitated by a payment processor. They seem to be trying to solve the problem of quickly verifying a payment.
Offline payments should allow this:
1. Alice receives money from some online payment.
2. Alice sends money to Bob via an offline payment.
3. Bob sends money to Catherine via an offline payment.
4. Catherine spends her money online.
In that scenario, Bob should not be required to ever do anything online. In other words, it should be possible for parties to only ever deal in offline transactions, without ever going online, as it is possible for people to only ever deal in paper cash.