On 6 Jan 2015, at 18:27, Dave Scott <[email protected]> wrote:
> 
> 
>> On 6 Jan 2015, at 15:49, Thomas Gazagnaire <[email protected]> wrote:
>> 
>> Thanks Louis for the great summary!
>> 
>>> - The forking and providing replacements would be really useful for Mirage,
>>> where we're having an active discussion about how to provide Xen-specific
>>> versions of certain packages such as Zarith.  Thomas (with any surname),
>>> opinions on this? 
>> 
>> My opinion: I was very keen to have this feature on the bug tracker, but 
>> with all the designs details in mind I'm much less keen now. It introduces a 
>> lot of complexity with pinning so would be nice to see what we can simplify. 
>> For instance, maybe we could forbid virtual packages - and pin only to real 
>> packages. Need to think a bit more about that.
>> 
>> a more meta comments: if we put design files in the repo (I'm not very fond 
>> of that, but why not), could we have a header specifying the status of the 
>> document (ie. draft, partially implemented, deprecated, etc...) and the 
>> target opam versions. And keep that up-to-date please.
> 
> FYI over in the xapi-project we’ve been experimenting with putting design 
> docs in a central github repo and having github render them nicely e.g.
> 
> http://xapi-project.github.io/design-docs/index.html
> 
> We stick a header on top like
> 
> ---
> title: thin LVHD storage
> layout: default
> design_doc: true
> revision: 1
> status: proposed
> —
> 
> and have a design doc index template
> 
> https://github.com/xapi-project/xapi-project.github.io/blob/master/_layouts/design_docs_list.html
> 
> So far it’s working quite well for us.

Oh wow, that's a lovely design!  I'm just doing a mini overhaul of the Real 
World OCaml commenting system that created GitHub issues from paragraphs -- I 
could imagine that fitting in quite well to let people leave comments and do 
spec review more easily.

-anil
_______________________________________________
opam-devel mailing list
[email protected]
http://lists.ocaml.org/listinfo/opam-devel

Reply via email to