5 ms·

I wrote my first Fortran program in 1969. I have been active in the field of computing since then, both as an academic and as a programmer in industry. During t

I wrote my first Fortran program in 1969. I have been active in the field of computing since then, both as an academic and as a programmer in industry. During this time, there have been several times when I first heard about something new, something potentially game changing. I remember some of them very clearly: 1. Around 1971: a Wang desktop computer. Imagine: a computer that fit on a desk! 2. Visicalc (spreadsheet). Wow! I remember thinking finally, a breakthrough for non-programmers. 3. Relational databases. 4. UUCP - a way to talk to remote computers. 5. HTML/HTTP - we knew from the start that this was big And then there were many that were hyped but which I rejected as hype. Even when I was wrong, I still felt that there was something intrinsically wrong about this category of things. These things either vanished, limped along (and still do), or which I rejected out of hand, but actually succeeded. Like Virtual Reality (back in about 1984), Fuzzy Logic, Proving programs correct, and, of course Bitcoin. The latest in this category are NFTs which my intuition tells me is rubbish on so many levels. However, I think they will have some success, although I will never have anything to do with them (or Bitcoin for that matter).

pilingual 6y ago
What is the primary point for which you believe NFTs to be rubbish?
- tupshin 6y ago
  In what way do you think they have any intrinsic value?
  - capableweb 6y ago
    In what way do you think that you think about intrinsic value?
- XorNot 6y ago
  What legal force does an NFT bring to what it claims to authenticate? Even a certificate of authenticity for a painting is still notionally a legal contract between the authenticator and buyers: after all, there will be consequences if they cannot show they could reasonably support the identity they establish for something. What does an NFT do? Nothing. There are no consequences in the real world. You can make another blockchain with more NFTs. It doesn't even establish an interesting digital history for an item: they're tied to hashes, not data.
  - delaaxe 6y ago
    NFTs are not meant to be tied to something in the physical world. They’re certificates on digital creations
    - acdha 6y ago
      You misunderstood the point: it’s not physical versus digital but rather that they have no added benefits. Legal rights aren’t something only physical objects can have and ownership of a pure digital creation can be quite valuable but most NFTs don’t have that and you don’t need the expense and risk of using a blockchain to transfer them. If your goal is to show that you supported an artist, simple PKI or even a Tweet works better and is orders of magnitude less expensive because all of the value comes from a statement by the creator/rights-holder.
arthur_sav 6y ago
As with any technology we only see its impact with the pass of time. Bitcoin is holding pretty strong after 12ish years, you can't deny facts. You can question the practical applications but technology is often used in other ways than it was originally intended. In the case of Bitcoin it may prove to be useful only as a cross border digital store of value instead of an every day currency. Same with NFTs, you can dismiss the hype but that doesn't mean there are no practical applications that people may end up adopting.
- stickfigure 6y ago
  Bitcoin works because the supply is constrained. Scarcity is its primary attribute. There are no constraints on the supply of NFTs.
  - delaaxe 6y ago
    What I find mind boggling is that some creators sell 100 copies of the same jpeg
    - EVa5I7bHFq9mnYK 6y ago
      Yes, but it is known to everyone exactly how many they will sell.
- JeremyNT 6y ago
  > Same with NFTs, you can dismiss the hype but that doesn't mean there are no practical applications that people may end up adopting Bitcoin and NFTs have essentially two (closely related) practical applications: 1) pyramid scheme and 2) gambling / speculation on market volatility. So long as USD are cheap and regulation is lax, these will remain incredibly compelling use cases.
- graeme 6y ago
  Being a store of value is a tautology, not a use case. Name one other store of value with zero use in its class. Even art and magic cards have them. You can look at art, and people play with their non collector’s magic cards. Bitcoin is a pure faith enterprise. It’s only value is that people believe in it, despite the lack of use case and despite the ever rising waste of energy to power the network which rises inexorably with the market price. It is a tautological network where people enjoy seeing it run in a distributed way even though it does nothing else. How long can that last as something storing value? As mining costs increase you need an ever increasing supply of new entrants at high prices who are similarly enamoured with the idea of the network.
samvher 6y ago
At the risk of going slightly off-topic: I think it's interesting you mention "proving programs correct". My hunch is that formal proofs on code are currently too laborious to be useful in most business situations, but that they will become extremely useful as we start letting the machine/"AI" code for us. I might not have spent enough time in this field (so there might be some Dunning Kruger at play) but I think code proofs might be similar enough to Chess/Go that they can be automated, and I think they will be able to add guarantees to computer-generated code so that we can trust it more.
- gpanders 6y ago
  If the code proofs are automated, then how do you prove the proof? Pardon my ignorance, but I’m being sincere. I don’t know much about this field.
  - samvher 6y ago
    When it comes to formal proofs of code, once you have a proof it is generally straightforward to verify (in the same way that you could for example evaluate a boolean expression). The problem is in finding the proofs - that's what requires labor and intelligence. So the way I imagine it would work, is you tell a programming system which conditions the code it's going to generate should satisfy, and then it generates the code + proofs that it satisfies those conditions. It might be trickier than I imagine though (e.g. for Chess/Go the reinforcement signal might be more suitable for learning than what you'd find with proofs). If you want to know more about proofs you might be interested in the series "Software Foundations" (https://softwarefoundations.cis.upenn.edu/ https://softwarefoundations.cis.upenn.edu/).

pilingual 6y ago

What is the primary point for which you believe NFTs to be rubbish?

tupshin 6y ago
In what way do you think they have any intrinsic value?
- capableweb 6y ago
  In what way do you think that you think about intrinsic value?
XorNot 6y ago
What legal force does an NFT bring to what it claims to authenticate? Even a certificate of authenticity for a painting is still notionally a legal contract between the authenticator and buyers: after all, there will be consequences if they cannot show they could reasonably support the identity they establish for something. What does an NFT do? Nothing. There are no consequences in the real world. You can make another blockchain with more NFTs. It doesn't even establish an interesting digital history for an item: they're tied to hashes, not data.
- delaaxe 6y ago
  NFTs are not meant to be tied to something in the physical world. They’re certificates on digital creations
  - acdha 6y ago
    You misunderstood the point: it’s not physical versus digital but rather that they have no added benefits. Legal rights aren’t something only physical objects can have and ownership of a pure digital creation can be quite valuable but most NFTs don’t have that and you don’t need the expense and risk of using a blockchain to transfer them. If your goal is to show that you supported an artist, simple PKI or even a Tweet works better and is orders of magnitude less expensive because all of the value comes from a statement by the creator/rights-holder.

arthur_sav 6y ago

As with any technology we only see its impact with the pass of time. Bitcoin is holding pretty strong after 12ish years, you can't deny facts. You can question the practical applications but technology is often used in other ways than it was originally intended. In the case of Bitcoin it may prove to be useful only as a cross border digital store of value instead of an every day currency. Same with NFTs, you can dismiss the hype but that doesn't mean there are no practical applications that people may end up adopting.

stickfigure 6y ago
Bitcoin works because the supply is constrained. Scarcity is its primary attribute. There are no constraints on the supply of NFTs.
- delaaxe 6y ago
  What I find mind boggling is that some creators sell 100 copies of the same jpeg
  - EVa5I7bHFq9mnYK 6y ago
    Yes, but it is known to everyone exactly how many they will sell.
JeremyNT 6y ago
> Same with NFTs, you can dismiss the hype but that doesn't mean there are no practical applications that people may end up adopting Bitcoin and NFTs have essentially two (closely related) practical applications: 1) pyramid scheme and 2) gambling / speculation on market volatility. So long as USD are cheap and regulation is lax, these will remain incredibly compelling use cases.
graeme 6y ago
Being a store of value is a tautology, not a use case. Name one other store of value with zero use in its class. Even art and magic cards have them. You can look at art, and people play with their non collector’s magic cards. Bitcoin is a pure faith enterprise. It’s only value is that people believe in it, despite the lack of use case and despite the ever rising waste of energy to power the network which rises inexorably with the market price. It is a tautological network where people enjoy seeing it run in a distributed way even though it does nothing else. How long can that last as something storing value? As mining costs increase you need an ever increasing supply of new entrants at high prices who are similarly enamoured with the idea of the network.

samvher 6y ago

At the risk of going slightly off-topic: I think it's interesting you mention "proving programs correct". My hunch is that formal proofs on code are currently too laborious to be useful in most business situations, but that they will become extremely useful as we start letting the machine/"AI" code for us. I might not have spent enough time in this field (so there might be some Dunning Kruger at play) but I think code proofs might be similar enough to Chess/Go that they can be automated, and I think they will be able to add guarantees to computer-generated code so that we can trust it more.

gpanders 6y ago
If the code proofs are automated, then how do you prove the proof? Pardon my ignorance, but I’m being sincere. I don’t know much about this field.
- samvher 6y ago
  When it comes to formal proofs of code, once you have a proof it is generally straightforward to verify (in the same way that you could for example evaluate a boolean expression). The problem is in finding the proofs - that's what requires labor and intelligence. So the way I imagine it would work, is you tell a programming system which conditions the code it's going to generate should satisfy, and then it generates the code + proofs that it satisfies those conditions. It might be trickier than I imagine though (e.g. for Chess/Go the reinforcement signal might be more suitable for learning than what you'd find with proofs). If you want to know more about proofs you might be interested in the series "Software Foundations" (https://softwarefoundations.cis.upenn.edu/ https://softwarefoundations.cis.upenn.edu/).