On 29/12/2013 22:41, Philippe Beaudette wrote:
On Sun, Dec 29, 2013 at 2:25 PM, Emilio J. Rodríguez-Posada <
emi...@gmail.com> wrote:

Naw, it's a great idea.  Let's switch to building our own ARM-based
servers (by the way, which have already been a flop commercially),
using only unproven, low-volume available motherboards and having to
buy and assemble all of the rest of the components.  And then of
course, we need to design our own cases... and since these have such a
low performance, we'll need to have a lot more rack and datacenter
space, of course which comes with a cost... and we'll have to figure
out how to run our caching layers which require large amounts of
memory... and our storage layers which require large amounts of disk
space.....  At that point we'll probably need to redesign those boards
which are incapable of doing these things, so we'll need a team of
hardware engineers, plus a deal with a manufacturing plant.

So... I think with about a 100 million dollar per year research budget
we can do this.  Who's ponying up? ;)

Funny huh?

If we use free software, I don't see why we can't move to open-source
hardware ASAP.

Well, I think Leslie just listed a few, but I'll recap:
- low-availability
- Requires in house assembly
- Requires in house design capacity
- Substantially more rack and datacenter space required
- Insufficient for caching and storage layers
- Cost of manufacturing.

Add to the list people that know what they are doing.

Wikimedia-l mailing list
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, 

Reply via email to