5 ms·

JPMorgan's Blockchain Chief Is Setting Out on Her Own

51 points by wglb 9y ago

scrumper 9y ago
One has to wonder if it's related to: https://www.reuters.com/article/us-blockchain-jpmorgan/jpmorgan-mulls-spin-off-of-blockchain-project-quorum-sources-idUSKBN1GY36O https://www.reuters.com/article/us-blockchain-jpmorgan/jpmor... (In that maybe the bank decided not to, or will but wants someone else to lead it, or the spinoff will prove worthless, or...)
- mathattack 9y ago
  Banks are very political places too. Not the ideal home for a hacker.
  - cm2187 9y ago
    And if you print all the policies and procedures, you can easily build a decent wall with all that paper.
    - mathattack 9y ago
      Not just policies, but a lot of backstabbing too.
abstrct 9y ago
I had the pleasure of meeting Amber at a conference recently and am not surprised she is going out on her own. With her knowledge and contact list, I'd expect her next venture to be a success. This has been a common occurrence across many of the large classic players of the financial space. Keeping talent is difficult when a startup can be substantially more agile and effective. I know this isn't unique to the blockchain space, but it is certainly very prevalent. Potentially due to the fact that many leaders in the space were early investors in the technology too and a stable salary isn't top of mind compared to freedom and a challenge.
- atomical 9y ago
  This comment is so generic.
sorenn111 9y ago
I wonder at what point blockchain will hit a make or break moment in finance companies. At some point, it will have to provide direct, tangible value-add or it might get funding pulled from non-tech companies
mark_l_watson 9y ago
Probably like most people on HN, I have no idea how effective permissioned blockchain will end up being. I took the edX Blockchain for Business class last year, and was sipping the coolaid, but now I am not as optimistic. For small groups of people and organizations, a permissioned blockchain might be over engineering when a distributed and secure shared database would be fine. For very large scale applications like health care records or smart contracts involving many thousands of individuals it makes more sense, but good luck getting off the ground. I am still experimenting with two frameworks, one written in Rust and the other Haskell, but I am starting to think I am investing too much time.
- guiomie 9y ago
  I've come to this same realization, except I'm also still quite on the fence for big corp use cases. I feel like RPC APIs or REST could do as good as a enterprise blockchain with better performance. May I ask what are the Rust and Haskell based frameworks you are looking at? I've been experimenting with Hyperledger Fabric and might try out BigchainDb.
  - mark_l_watson 9y ago
    Exonum is written in Rust and Pact (and Juno) are written in Haskell. I was very enthusiastic last fall about setting up, on a very small scale, what my friend Ben Goertzel did with SingularityNet.io. I set up a placeholder web site http://hyperledgerai.com/ http://hyperledgerai.com/ but have been too busy with my day job to put enough effort into it.
- buckie 9y ago
  Is one of them Juno? https://github.com/kadena-io/juno https://github.com/kadena-io/juno. If so, I was lead eng on it so let me know if you have any questions. FYI we're hiring haskell devs to work on our private consensus layer (based off of Juno), our public consensus protocol (http://kadena.io/docs/chainweb-v15.pdf http://kadena.io/docs/chainweb-v15.pdf) and our smart contract language https://github.com/kadena-io/pact https://github.com/kadena-io/pact + formal verification system. hiring+hn@kadena.io.
fwdpropaganda 9y ago
One datapoint: I've started seeing small dev shops who offer to their clients that they can do things "on the blockchain". "Bitcoin is hype, but blockchain is the underlying technology." Seems like MySQL could solve the problem just as well.
- snissn 9y ago
  One can't "Roll back" in mysql. Mysql also can't easily create N master-master replication nodes.
  - muditsingh5000 9y ago
    I am not sure it would be practically feasible for a "rollback" in a blockchain based system. Miners would lose a lot of their reward if you suddenly say that this chain is invalid and all data has to be entered again. It would use a lot of computing resources as well.
    - ballenf 9y ago
      Rollback is typically only discussed with regard to private or closed blockchain implementations. Which just highlights the root "why blockchain?" question even further when discussing private implementations.
  - ashelmire 9y ago
    It is, in fact, fairly simple to track every change in a traditional db, who did it, and what the prior and subsequent state were. You can rollback such transactions without rewriting the whole db (as you'd have to do with a blockchain).
  - fwdpropaganda 9y ago
    Devs push updates to the nodes. They can do whatever they want with it.
- guiomie 9y ago
  I've been getting more serious about Hyperledger Fabric, and with time, I can't seem to see any use case that can't be solved by a typical distributed app. I see the utility of permissionless blockchains, but private blockchains, I don't get it.
icc97 9y ago
She's a very interesting speaker. Gave a talk at Def Con [0] on ways of talking to friends who are feeling very depressed and suicidal. [0]: https://m.youtube.com/watch?v=bfYJE-qCVaM https://m.youtube.com/watch?v=bfYJE-qCVaM
theocean154 9y ago
Most of the so called "emerging tech" projects at JPM are being phased out due to not delivering the promised returns on investment (lots of money for expensive developers to build out moonshot projects with little concrete deliverables).
- kolinko 9y ago
  The same is true for startups :)

51 points by wglb 9y ago

One has to wonder if it's related to: https://www.reuters.com/article/us-blockchain-jpmorgan/jpmorgan-mulls-spin-off-of-blockchain-project-quorum-sources-idUSKBN1GY36O https://www.reuters.com/article/us-blockchain-jpmorgan/jpmor... (In that maybe the bank decided not to, or will but wants someone else to lead it, or the spinoff will prove worthless, or...)

mathattack 9y ago
Banks are very political places too. Not the ideal home for a hacker.
- cm2187 9y ago
  And if you print all the policies and procedures, you can easily build a decent wall with all that paper.
  - mathattack 9y ago
    Not just policies, but a lot of backstabbing too.

abstrct 9y ago

I had the pleasure of meeting Amber at a conference recently and am not surprised she is going out on her own. With her knowledge and contact list, I'd expect her next venture to be a success. This has been a common occurrence across many of the large classic players of the financial space. Keeping talent is difficult when a startup can be substantially more agile and effective. I know this isn't unique to the blockchain space, but it is certainly very prevalent. Potentially due to the fact that many leaders in the space were early investors in the technology too and a stable salary isn't top of mind compared to freedom and a challenge.

atomical 9y ago
This comment is so generic.

sorenn111 9y ago

I wonder at what point blockchain will hit a make or break moment in finance companies. At some point, it will have to provide direct, tangible value-add or it might get funding pulled from non-tech companies

mark_l_watson 9y ago

Probably like most people on HN, I have no idea how effective permissioned blockchain will end up being. I took the edX Blockchain for Business class last year, and was sipping the coolaid, but now I am not as optimistic. For small groups of people and organizations, a permissioned blockchain might be over engineering when a distributed and secure shared database would be fine. For very large scale applications like health care records or smart contracts involving many thousands of individuals it makes more sense, but good luck getting off the ground. I am still experimenting with two frameworks, one written in Rust and the other Haskell, but I am starting to think I am investing too much time.

guiomie 9y ago
I've come to this same realization, except I'm also still quite on the fence for big corp use cases. I feel like RPC APIs or REST could do as good as a enterprise blockchain with better performance. May I ask what are the Rust and Haskell based frameworks you are looking at? I've been experimenting with Hyperledger Fabric and might try out BigchainDb.
- mark_l_watson 9y ago
  Exonum is written in Rust and Pact (and Juno) are written in Haskell. I was very enthusiastic last fall about setting up, on a very small scale, what my friend Ben Goertzel did with SingularityNet.io. I set up a placeholder web site http://hyperledgerai.com/ http://hyperledgerai.com/ but have been too busy with my day job to put enough effort into it.
buckie 9y ago
Is one of them Juno? https://github.com/kadena-io/juno https://github.com/kadena-io/juno. If so, I was lead eng on it so let me know if you have any questions. FYI we're hiring haskell devs to work on our private consensus layer (based off of Juno), our public consensus protocol (http://kadena.io/docs/chainweb-v15.pdf http://kadena.io/docs/chainweb-v15.pdf) and our smart contract language https://github.com/kadena-io/pact https://github.com/kadena-io/pact + formal verification system. hiring+hn@kadena.io.

fwdpropaganda 9y ago

One datapoint: I've started seeing small dev shops who offer to their clients that they can do things "on the blockchain". "Bitcoin is hype, but blockchain is the underlying technology." Seems like MySQL could solve the problem just as well.

snissn 9y ago
One can't "Roll back" in mysql. Mysql also can't easily create N master-master replication nodes.
- muditsingh5000 9y ago
  I am not sure it would be practically feasible for a "rollback" in a blockchain based system. Miners would lose a lot of their reward if you suddenly say that this chain is invalid and all data has to be entered again. It would use a lot of computing resources as well.
  - ballenf 9y ago
    Rollback is typically only discussed with regard to private or closed blockchain implementations. Which just highlights the root "why blockchain?" question even further when discussing private implementations.
- ashelmire 9y ago
  It is, in fact, fairly simple to track every change in a traditional db, who did it, and what the prior and subsequent state were. You can rollback such transactions without rewriting the whole db (as you'd have to do with a blockchain).
- fwdpropaganda 9y ago
  Devs push updates to the nodes. They can do whatever they want with it.
guiomie 9y ago
I've been getting more serious about Hyperledger Fabric, and with time, I can't seem to see any use case that can't be solved by a typical distributed app. I see the utility of permissionless blockchains, but private blockchains, I don't get it.

icc97 9y ago

She's a very interesting speaker. Gave a talk at Def Con [0] on ways of talking to friends who are feeling very depressed and suicidal. [0]: https://m.youtube.com/watch?v=bfYJE-qCVaM https://m.youtube.com/watch?v=bfYJE-qCVaM

theocean154 9y ago

Most of the so called "emerging tech" projects at JPM are being phased out due to not delivering the promised returns on investment (lots of money for expensive developers to build out moonshot projects with little concrete deliverables).

kolinko 9y ago
The same is true for startups :)