7 ms·

Well, it's pretty clearly an aquihire. Zoom gets a team of highly skilled cryptographers and Internet protocol experts. Good for them. But that means the team t

Well, it's pretty clearly an aquihire. Zoom gets a team of highly skilled cryptographers and Internet protocol experts. Good for them. But that means the team that created Keybase as an innovate PKI store won't be working on that anymore. That's not Zoom's business, and probably won't be, as Keybase themselves never figured out how to turn it into a business.

asdkhadsj 7y ago
It bothers me that they even tried, honestly. Keybase seems like something that should be small, isolated, FOSS, supported by a foundation, etc. They could have built a business _around_ Keybase I'd imagine, but all they managed to do with this is invalidate Keybase and make people like myself, who feared their business motivations, feel vindicated for being paranoid. I'll never blame anyone for wanting to make money, to make a business, etc. But if you make a product that walks talks and acts like a FOSS project, but keep it to center your business around... I'll always be longing for a real, true FOSS replacement. In this case a good looking FOSS alternative came out a few months ago iirc. Though for the life of me I can't remember the name. edit: https://keys.pub/ https://keys.pub/ - though I will still miss KBFS
- dcow 7y ago
  How was their product ever marketed that way? They have open source clients because that security table stakes. They’re a solution to crypto anarchy because they help link your crypto identities to your social ones. None of that has changed. You talk like all valid software is free of corporate ownership/sponsorship. Why is zoom’s money somehow worse than e.g. softbank’s.
  - asdkhadsj 7y ago
    Oh I didn't mean to imply it was, maybe my "walks talks" bit was unfair. Rather I merely meant that Keybase, like Keys.pub, seems like a great isolated tool for the internet. Something exceptionally well suited for a foundation.
- coldpie 7y ago
  As someone who works at an open-source-focused business, I respectfully disagree. Unlike proprietary software, open source software doesn't depend on the broken window fallacy. As a result, it's really hard to make open source profitable. There's lots of different avenues to get there, and I don't like to fault someone for their efforts if the bulk of their work goes towards improving open source software, as I think Keybase did.
  - asdkhadsj 7y ago
    Possibly, but in this case I didn't expect them to make Keybase profitable, if anything I expect the opposite. I expect Keybase to be a FOSS, foundation for profitable extensions that the company builds and sells. Arguably I think they agree with me, about the extensions at least. As seen by their seemingly random directions of feature extensions that Keybase was prone to. My issue is not that they chose random features to try to make profitable, but rather that the core premise, a public keystore, was tied so closely to a for profit company. It would be like losing Git because Github went under. (Though, terrible example because Git works without a centralized repo, but it's just the first company <-> FOSS relationship that came to mind lol.)
    - semi-extrinsic 7y ago
      Maybe "losing Ubuntu if Canonical went under" is a better analogy then?
    - 411111111111111 7y ago
      It's actually a pretty good example to be honest. Keybase was a centralized key storage with value-add services such as file storage and chat. That was absolutely comparable to github, as you could've just gone back to manually syncing pubkeys and encrypting msgs. If github went away, you'd be without a lot of value-add services as well such as wiki, issues user management etc Realistically speaking, nobody is going to do that... And tbh, it was already dead in the water when they added crypto currencies... Just took a while for their money to run out. The actual difference is that there are enough competing products for github, not for keybase however, as that is just too niche
- StavrosK 7y ago
  keys.pub doesn't have the single most useful feature Keybase has: The ability to verifiably establish a secure channel with anyone given their Twitter/Github/whatever username.
  - floatingatoll 7y ago
    Their homepage advertises 'keys pull username@github' as an example. Is the missing piece you describe here simply the command 'keys chat username@github'?
    - StavrosK 7y ago
      No, it's the cryptographic attestation so you know you are getting the right key.
      - Squithrilve 7y ago
        They don't support it? That's weird (maybe a missing feature) given that it's quite easy to add to anything that has signed metadata, see e.g. this for OpenPGP: https://github.com/wiktor-k/openpgp-proofs#openpgp-proofs https://github.com/wiktor-k/openpgp-proofs#openpgp-proofs
        StavrosK 7y ago
        I'm not very familiar with the service, but AFAIK they don't. It would be great if they added it. EDIT: It looks like it might, from the front page, I will try it out to make sure. If it does, that'll be great! EDIT 2: It sort of does, but it's on a per-key basis, not an entire identity. You can publish proof on Twitter/Github/whatever, but it's only for one specific key, and it's one key per service, which means you can't only have one identity and multiple services.
buttersbrian 7y ago
It doesn't seem certain that they won't dogfood in addition to using the expertise internally.
usrusr 7y ago
The biggest challenge with acquihiring is retention. Allowing the acquired team to continue what they were doing is the only somewhat foolproof strategy to deal with it. It's a question of pocket depth and expectations. How much will the new team contribute to the "home" product? If expectations are too high chances are that much of the team won't stay and the acquisition will turn out to be a waste of money. With lower expectations however, continuing to fund the project in question can be bargain for getting a pool of in-house consultants to occasionally tap into for the "home" product, if they are really as good. And even if retention wasn't a problem at all, skilled people are not inherently skilled, they need to keep challenging themselves in their area of expertise to stay sharp. If the "home" product was failing to foster in-house expertise before then chances are high that it's a problem based on culture and priorities and experts injected from outside would quickly lose their edge. Keep them on the project they became experts on and they stay experts.
- labster 7y ago
  Retention isn’t going to be a problem in this market. Hiring in general has almost disappeared. And if you get hospitalized with coronavirus without health insurance, it will almost certainly lead to bankruptcy. It’s too risky not to have a job right now.
  - jki275 7y ago
    I don’t think hiring has disappeared - I got contacted by four recruiters just yesterday alone asking me to apply to vacancies they’re trying to hire for. I wouldn’t quit my job (and I’m not looking anyway), but there’s plenty of hiring going on.
- benhurmarcel 7y ago
  Relevant: https://dilbert.com/strip/2014-07-31 https://dilbert.com/strip/2014-07-31
hinkley 7y ago
My experiences with acquihires have not been good. People went to work for this company based on the domain, the people, or the culture. With the acquihire they change the domain first, and the culture about a year in. Then the people start to leave, and it's just a job, and one you didn't even apply for. On my worst days it felt like I was sold like cattle, and I would have seen more upside by hiring on someplace else.

asdkhadsj 7y ago

It bothers me that they even tried, honestly. Keybase seems like something that should be small, isolated, FOSS, supported by a foundation, etc. They could have built a business _around_ Keybase I'd imagine, but all they managed to do with this is invalidate Keybase and make people like myself, who feared their business motivations, feel vindicated for being paranoid. I'll never blame anyone for wanting to make money, to make a business, etc. But if you make a product that walks talks and acts like a FOSS project, but keep it to center your business around... I'll always be longing for a real, true FOSS replacement. In this case a good looking FOSS alternative came out a few months ago iirc. Though for the life of me I can't remember the name. edit: https://keys.pub/ https://keys.pub/ - though I will still miss KBFS

dcow 7y ago
How was their product ever marketed that way? They have open source clients because that security table stakes. They’re a solution to crypto anarchy because they help link your crypto identities to your social ones. None of that has changed. You talk like all valid software is free of corporate ownership/sponsorship. Why is zoom’s money somehow worse than e.g. softbank’s.
- asdkhadsj 7y ago
  Oh I didn't mean to imply it was, maybe my "walks talks" bit was unfair. Rather I merely meant that Keybase, like Keys.pub, seems like a great isolated tool for the internet. Something exceptionally well suited for a foundation.
coldpie 7y ago
As someone who works at an open-source-focused business, I respectfully disagree. Unlike proprietary software, open source software doesn't depend on the broken window fallacy. As a result, it's really hard to make open source profitable. There's lots of different avenues to get there, and I don't like to fault someone for their efforts if the bulk of their work goes towards improving open source software, as I think Keybase did.
- asdkhadsj 7y ago
  Possibly, but in this case I didn't expect them to make Keybase profitable, if anything I expect the opposite. I expect Keybase to be a FOSS, foundation for profitable extensions that the company builds and sells. Arguably I think they agree with me, about the extensions at least. As seen by their seemingly random directions of feature extensions that Keybase was prone to. My issue is not that they chose random features to try to make profitable, but rather that the core premise, a public keystore, was tied so closely to a for profit company. It would be like losing Git because Github went under. (Though, terrible example because Git works without a centralized repo, but it's just the first company <-> FOSS relationship that came to mind lol.)
  - semi-extrinsic 7y ago
    Maybe "losing Ubuntu if Canonical went under" is a better analogy then?
  - 411111111111111 7y ago
    It's actually a pretty good example to be honest. Keybase was a centralized key storage with value-add services such as file storage and chat. That was absolutely comparable to github, as you could've just gone back to manually syncing pubkeys and encrypting msgs. If github went away, you'd be without a lot of value-add services as well such as wiki, issues user management etc Realistically speaking, nobody is going to do that... And tbh, it was already dead in the water when they added crypto currencies... Just took a while for their money to run out. The actual difference is that there are enough competing products for github, not for keybase however, as that is just too niche
StavrosK 7y ago
keys.pub doesn't have the single most useful feature Keybase has: The ability to verifiably establish a secure channel with anyone given their Twitter/Github/whatever username.
- floatingatoll 7y ago
  Their homepage advertises 'keys pull username@github' as an example. Is the missing piece you describe here simply the command 'keys chat username@github'?
  - StavrosK 7y ago
    No, it's the cryptographic attestation so you know you are getting the right key.
    - Squithrilve 7y ago
      They don't support it? That's weird (maybe a missing feature) given that it's quite easy to add to anything that has signed metadata, see e.g. this for OpenPGP: https://github.com/wiktor-k/openpgp-proofs#openpgp-proofs https://github.com/wiktor-k/openpgp-proofs#openpgp-proofs
      - StavrosK 7y ago
        I'm not very familiar with the service, but AFAIK they don't. It would be great if they added it. EDIT: It looks like it might, from the front page, I will try it out to make sure. If it does, that'll be great! EDIT 2: It sort of does, but it's on a per-key basis, not an entire identity. You can publish proof on Twitter/Github/whatever, but it's only for one specific key, and it's one key per service, which means you can't only have one identity and multiple services.

buttersbrian 7y ago

It doesn't seem certain that they won't dogfood in addition to using the expertise internally.

usrusr 7y ago

The biggest challenge with acquihiring is retention. Allowing the acquired team to continue what they were doing is the only somewhat foolproof strategy to deal with it. It's a question of pocket depth and expectations. How much will the new team contribute to the "home" product? If expectations are too high chances are that much of the team won't stay and the acquisition will turn out to be a waste of money. With lower expectations however, continuing to fund the project in question can be bargain for getting a pool of in-house consultants to occasionally tap into for the "home" product, if they are really as good. And even if retention wasn't a problem at all, skilled people are not inherently skilled, they need to keep challenging themselves in their area of expertise to stay sharp. If the "home" product was failing to foster in-house expertise before then chances are high that it's a problem based on culture and priorities and experts injected from outside would quickly lose their edge. Keep them on the project they became experts on and they stay experts.

labster 7y ago
Retention isn’t going to be a problem in this market. Hiring in general has almost disappeared. And if you get hospitalized with coronavirus without health insurance, it will almost certainly lead to bankruptcy. It’s too risky not to have a job right now.
- jki275 7y ago
  I don’t think hiring has disappeared - I got contacted by four recruiters just yesterday alone asking me to apply to vacancies they’re trying to hire for. I wouldn’t quit my job (and I’m not looking anyway), but there’s plenty of hiring going on.
benhurmarcel 7y ago
Relevant: https://dilbert.com/strip/2014-07-31 https://dilbert.com/strip/2014-07-31

hinkley 7y ago

My experiences with acquihires have not been good. People went to work for this company based on the domain, the people, or the culture. With the acquihire they change the domain first, and the culture about a year in. Then the people start to leave, and it's just a job, and one you didn't even apply for. On my worst days it felt like I was sold like cattle, and I would have seen more upside by hiring on someplace else.