[Opensim-dev] AssetBase and metadata
Hurliman, John
john.hurliman at intel.com
Mon Feb 2 18:19:55 UTC 2009
>-----Original Message-----
>From: opensim-dev-bounces at lists.berlios.de [mailto:opensim-dev-
>bounces at lists.berlios.de] On Behalf Of Sean Dague
>Sent: Monday, February 02, 2009 5:47 AM
>To: opensim-dev at lists.berlios.de
>Subject: Re: [Opensim-dev] AssetBase and metadata
>
>Teravus Ovares wrote:
>> Is there any reason that we don't request items from the asset server
>> internally by the inventory UUID instead of the asset UUID?
>> Requesting assets by inventory UUID would make it a LOT simpler to
>> apply permissions at the trusted service level instead of at the
>> simulator level.
>
>To do that sanely, we'd have to merge the inventory and asset servers,
>which honestly, wouldn't be such a bad idea. They have a hard time
>living on their own anyway.
>
> -Sean
>
>--
>Sean Dague / Neas Bade
>sdague at gmail.com
>http://dague.net
>
That's exactly the approach in the Cable Beach asset/inventory server. As Teravus pointed out, assets and inventory have very different characteristics when it comes to storage and caching (inventory is almost identical to metadata). The server* has different backends/frontends for inventory, but shares authentication and authorization modules between inventory and assets.
* In the current prototype implementation of Cable Beach asset/inventory, a single process is used. There are a nearly infinite number of techniques to break services apart, put them together, run different pieces on different machines, etc. The implementation details of what code is executing on what machine is not as important at this stage as defining the interfaces, and asserting that there is or is not a tight coupling between assets and inventory. If inventory defines the many to one mapping that allows multiple people to own the same asset with different permissions, I think a distributed grid must tightly couple these two concepts.
John
More information about the Opensim-dev
mailing list