3 ms·

Oh man. I'm such a Power fanboy...and this sort of response is just so cringeworthy that I really hope someone else starts making Power ISA 3.0 chips. I mean,

by andreiw 10y ago

Oh man. I'm such a Power fanboy...and this sort of response is just so cringeworthy that I really hope someone else starts making Power ISA 3.0 chips. I mean, hello? Director of Ecosystem and Alliances really needs an explanation of why OpenBSD is useful and why it should run on Power8/9 as well? Need I remind that processors only work when there is software to run? IBM should be excited by anyone wanting to spend time getting stuff to work. And, hello, with the cheapest (Tyan) box currently around $3k, not many people are going to be buying these on their own...OTOH what is $3k x 10 for IBM? Nothing.

andreiw 10y ago
Uh oh, IBM suit down voting me. Must have touched a sore spot... Hey buddy, it's not my fault your executives in charge of curating the Power ecosystem are so freaking clueless. I suppose it's a miracle they spelled OpenBSD without making 3 mistakes.... Speaking of which, when are we going to have hardware approachable to system developers? I mean something that isn't 2U, doesn't need to go into a rack, doesn't sound like a 737 taking off on boot and doesn't cost $3k with only 4GB of RAM? A tower within $2k sure should be reasonable to make.
mikaelhn 10y ago
While OpenBSD is widely known to tech people, it's not on frontpages like Linux has been, and it has zero backing by marketing and sales departments like what corporations have e.g. RedHat or Oracle. IBM has traditionally been good at enterprise customers, and tech, and I guess the need of an explanation reflects that this is a big cultural space. Credit is due in the fact that they take time to ask the question. My personal take is that if IBM's committment is to broaden an ecosystem (as I think I see hints of with the OpenPOWER initiative), then donating is a fairly obvious choice as it's a fantastic value deal - one more major OS supported. However in a traditional enterprise framing for very large customers only, maybe OpenBSD support wouldn't make sales happen. This is still an extremely recent conversation and I am sure a range of decisionmakers are involved indirectly. Let's see how it turns out. I do agree that real hardware in developers' hands is required for real architecture support. Let's see what kind of physical footprint and price point the Talos OpenPOWER motherboard will have in 1-2 years from now.
- mikaelhn 10y ago
  Just for clarity, this query has been on the table of the competent person for something like 72 hours. Please wait with making conclusions until 3-6mo from now, so something like the second or third quarter 2017. It's easy as an individual to think that an enterprise should react in an instant, but getting things through a heavily interlocked environment with tons of people and departments takes effort and time.

by andreiw 10y ago

andreiw 10y ago

Uh oh, IBM suit down voting me. Must have touched a sore spot... Hey buddy, it's not my fault your executives in charge of curating the Power ecosystem are so freaking clueless. I suppose it's a miracle they spelled OpenBSD without making 3 mistakes.... Speaking of which, when are we going to have hardware approachable to system developers? I mean something that isn't 2U, doesn't need to go into a rack, doesn't sound like a 737 taking off on boot and doesn't cost $3k with only 4GB of RAM? A tower within $2k sure should be reasonable to make.

mikaelhn 10y ago

While OpenBSD is widely known to tech people, it's not on frontpages like Linux has been, and it has zero backing by marketing and sales departments like what corporations have e.g. RedHat or Oracle. IBM has traditionally been good at enterprise customers, and tech, and I guess the need of an explanation reflects that this is a big cultural space. Credit is due in the fact that they take time to ask the question. My personal take is that if IBM's committment is to broaden an ecosystem (as I think I see hints of with the OpenPOWER initiative), then donating is a fairly obvious choice as it's a fantastic value deal - one more major OS supported. However in a traditional enterprise framing for very large customers only, maybe OpenBSD support wouldn't make sales happen. This is still an extremely recent conversation and I am sure a range of decisionmakers are involved indirectly. Let's see how it turns out. I do agree that real hardware in developers' hands is required for real architecture support. Let's see what kind of physical footprint and price point the Talos OpenPOWER motherboard will have in 1-2 years from now.

mikaelhn 10y ago
Just for clarity, this query has been on the table of the competent person for something like 72 hours. Please wait with making conclusions until 3-6mo from now, so something like the second or third quarter 2017. It's easy as an individual to think that an enterprise should react in an instant, but getting things through a heavily interlocked environment with tons of people and departments takes effort and time.