[Geopackage] Geopackage Digest, Vol 18, Issue 4

Brad Hards bradh at frogmouth.net
Mon Dec 28 18:47:28 EST 2015


On Sun, 27 Dec 2015 11:02:28 PM Jeff Yutzler wrote:
> Brad,
> I don't think I understand what you are getting at regarding #132. Perhaps
> the issue is that I never got around to updating Requirement 2 to bump the
> version number. (I just added that as issue 188 / pull request 189.) A
> 1.0/1.0.1 implementation would have extension names as annexes and a 1.0.2
> implementation would have them as sub-annexes. We haven't been rigorous
> about avoiding SHALL statements outside of enumerated requirements.
> Addressing this could get complicated and could easily do more harm than
> good.
I think of patch level releases (x.y.z -> x.y.z+1) as "if the implementation 
was compliant before, it won't be non-compliant now". Alternatively, the 
consumer can rely on a conforming producer to make stuff that doesn't break a 
conforming consumer.

That won't be true for the change in #132 if the  provider implemented 
extensions, because the description is now incorrect. Its a SHALL, so 
consumers could be relying on it, even if you don't know who they are.

I think your change notes should reflect that, and perhaps be suggesting 1.1.0, 
not 1.0.2.
 
> I haven't heard specifically, but I assume there will be a 30 day review
> period or something like that before 1.0.2 can go up for adoption vote. I
> am just trying to get that clock started. We are still able to fix issues
> after that.
I think it would be more reasonable to  finish the spec, then give a reasonable 
(30 day) review period on that spec. I'm not part of the OGC though, so 
whatever works for you.

Brad


More information about the Geopackage mailing list