Lack of randomness in pset template GlobalIds

[Moult]

The GlobalIds used in the distributed pset template definitions are not very unique. Here's a sample:

'3trzS0qRmHuO00025QrE$V'
'087P_0qXaHuO00025QrE$V'
'0h76K0qS4HuO00025QrE$V'
'0xpDi0qS4HuO00025QrE$V'
'1UNlG0qS4HuO00025QrE$V'
'3hy1C0qSGHuO00025QrE$V'
'1tQQG0qSKHuO00025QrE$V'
'2u_ay0qSKHuO00025QrE$V'
'073rA0qSOHuO00025QrE$V'
'1xexg0qU4HuO00025QrE$V'
'3_B7e0qU4HuO00025QrE$V'
'1cp2I0qUiHuO00025QrE$V'
'2YQUq0qV4HuO00025QrE$V'
'3AP4i0qWSHuO00025QrE$V'
'2Vni80qWWHuO00025QrE$V'

Is this a problem? Should it be fixed? Ping @berlotti @TLiebich @aothms

[aothms]

It's probably the uuid version (1,2,3,4,5) in play here, it's definitely not 4. It still allows for a lot (a bit less than) 18014398509481984 (64 ^ 9) defintions.

[Moult]

So related to https://github.com/buildingSMART/NextGen-IFC/issues/49 - should we specify a version? Sooner is better than later?

[berlotti]

policy for IFC 4.3.x is that property names and Pset names are unique. Names can be used as identifier.

[Moult]

@berlotti cheers - that clarifies that the GlobalIds are not "significant". If that is the case, this issue is a duplicate of https://github.com/buildingSMART/NextGen-IFC/issues/49 - should I start migrating unresolved issues from the next-gen repo to this issue?

[Moult]

Bump - if pset templates are included in a project and related to psets used within the project, does that raise the possibility of collisions to be something to be worried about?

[devonsparks]

@Moult - Where did we land on this topic? It seemed like there was good discussion in #49, but it was closed. The basic premise of being prescriptive on UUID versions to avoid collisions still seems sound.

[Moult]

No progress that I'm aware of.