[sldev] Opening the server source?

[Dirk Moerenhout]

That's a rather big "if". Any new startup will seemingly have this
because they don't need scalability or performance. If LL would start
anew with their current knowledge they could run a grid for 1/10th the
people without hassle too. The question is who would be able to run a
full alternative at the same size without failing? Obviously the
answer to that bubbles up while writing the question: any company that
has a different setup that faces the same challenges. Which brings us
to the like of Google. Though at that point the issue changes again.
What benefit would Google and alike have to try and compete with LL?
They could just buy and expand which would be tons cheaper than
starting anew.

Tuning services with a size as big as this is hard to do by open
sourcing some code. There are a lot of things that come into play
which you simply can't test just playing with code. To better the grid
LL is, for now, best off hiring more experts.

Dirk aka Blakar Ogre

On 6/26/07, Matthew Dowd <matthew.dowd at hotmail.co.uk> wrote:
>
>If someone did offer a alternative grid with a backend asset server
architecture which supported user backups without breaking the
permissions systems (using for instance some of the ideas in jira),
managed to address some of the scalabilitty/performance issues,
offered 24/7 support with minimal waiting queues, and offered an
import mechanism of full permission scripts/prims between LL grid and
their grid, I think they would give LL a run for their money at the
moment.