4 ms·

This is patently false. Redhat is making bank off of this guy. Jeff has written 100s of Ansible roles for free and made them widely available. I've used dozens

This is patently false. Redhat is making bank off of this guy. Jeff has written 100s of Ansible roles for free and made them widely available. I've used dozens of his roles as a contractor for F500 companies to help get projects off the ground quickly. Now instead of giving him free binary compatible OSes to develop his roles against, he'll probably just drop support because there's too much hassle. The next time I go to install an OS, it'll be Debian, because most of the tools I'll use won't have packages for RHEL like OSes. This is so dumb it hurts, Redhat won't recover from this. It'll cripple their package ecosystem over the next decade.

bragr 3y ago
>This is patently false. Redhat is making bank off of this guy. Jeff has written 100s of Ansible roles for free and made them widely available. Firstly, creating value is not the same thing as creating revenue. Second, people aren't licensing Redhat for access to these. People license Redhat so when they have a bug that affects them, it gets fixed right away with guarantees. Unless that guy is getting paid to do those fixes, I don't see how this affects Redhat customers.
- geerlingguy 3y ago
  You'd be surprised how many paying RHEL customers use community-contributed Ansible roles because there's no way for a company even as large as IBM to employ experts in all the various software RHEL users install to maintain integrations. That's the long-term effect of starving the wider community: you lose having 'great' integrations, and you end up with poorly-written and barely-maintained integrations for anything that's not extremely mainstream, and so you start losing the customers who use anything outside the core set of products extensively.
  - bragr 3y ago
    I'll just copy paste what I said in another comment: >I'm not saying this is the morally right, "best" in the long term sense, or otherwise socially optimal choice, I'm just saying it's the standard IBM choice.
    - geerlingguy 3y ago
      Ah, missed that ;)
  - berkes 3y ago
    You make it sound as if it's without doubt that a) a significant number of RedHat users require Ansible and b) those users will stop using RedHat if this Ansible integration becomes unavailable. I highly doubt both. I don't have numbers to back it up, but neither do I see any indication that it will matter to RedHats baseline.
Quarrelsome 3y ago
> It'll cripple their package ecosystem over the next decade. but in a way that's really hard to measure and doesn't show up on a bean-counter's spreadsheet. So that means corporate will consider it "good business" and will just blame something else for the immeasurable failures a decade later.

bragr 3y ago

>This is patently false. Redhat is making bank off of this guy. Jeff has written 100s of Ansible roles for free and made them widely available. Firstly, creating value is not the same thing as creating revenue. Second, people aren't licensing Redhat for access to these. People license Redhat so when they have a bug that affects them, it gets fixed right away with guarantees. Unless that guy is getting paid to do those fixes, I don't see how this affects Redhat customers.

geerlingguy 3y ago
You'd be surprised how many paying RHEL customers use community-contributed Ansible roles because there's no way for a company even as large as IBM to employ experts in all the various software RHEL users install to maintain integrations. That's the long-term effect of starving the wider community: you lose having 'great' integrations, and you end up with poorly-written and barely-maintained integrations for anything that's not extremely mainstream, and so you start losing the customers who use anything outside the core set of products extensively.
- bragr 3y ago
  I'll just copy paste what I said in another comment: >I'm not saying this is the morally right, "best" in the long term sense, or otherwise socially optimal choice, I'm just saying it's the standard IBM choice.
  - geerlingguy 3y ago
    Ah, missed that ;)
- berkes 3y ago
  You make it sound as if it's without doubt that a) a significant number of RedHat users require Ansible and b) those users will stop using RedHat if this Ansible integration becomes unavailable. I highly doubt both. I don't have numbers to back it up, but neither do I see any indication that it will matter to RedHats baseline.

Quarrelsome 3y ago

> It'll cripple their package ecosystem over the next decade. but in a way that's really hard to measure and doesn't show up on a bean-counter's spreadsheet. So that means corporate will consider it "good business" and will just blame something else for the immeasurable failures a decade later.