Re: [mif] Adoption of API document as the MIF WG document

[Simon Perreault <[e-mail]>]

On 2011-12-19 10:44, Ted Lemon wrote:
> On Dec 19, 2011, at 10:04 AM, Simon Perreault wrote:
>> Why not just specify the high-level API right away? Why is this intermediate
>> step necessary? As you say, all the underlying functionality is already
>> available on the various platforms out there.
>
> Which high-level API should we define, Simon? There is no single high-level API
> that addresses all use cases.

I disagree, I think it is a solvable problem.

> And while there is piecemeal support for features
> of the low-level API, at the moment hardly anybody supports all the required
> functionality, as far as I can tell. Also, a lot of the functionality is hidden
> under mid-level APIs that obscure features that would be required to implement
> the high-level APIs we want. This is why I think this work is important.

Understood, thanks.

> This is an easy problem for you to solve: write a document describing the
> high-level API you think we need.

Yeah, I know. *sigh*

I just don't want the WG to see this document as something that allows us to 
place a "DONE" checkmark next to the API milestone. There is still API work to be 
done.

Simon
-- 
DTN made easy, lean, and smart --> http://postellation.viagenie.ca
NAT64/DNS64 open-source        --> http://ecdysis.viagenie.ca
STUN/TURN server               --> http://numb.viagenie.ca