Re: [apps-discuss] IETF technical plenary: the end of application protocols

[Scott Brim <[e-mail]>]

On Mon, Mar 21, 2011 at 12:55, Dave CROCKER <[e-mail]> wrote:
> On 3/21/2011 9:40 AM, Scott Brim wrote:
>>
>> On Mon, Mar 21, 2011 at 12:37, Dave CROCKER<[e-mail]>  wrote:
>>>
>>> I think Claudio raises another fundamental point, namely the loss of
>>> interoperability.  The proposed model is a pure silo, with no apparent
>>> ability to have variations or incremental capabilities that build on the
>>> base, except by the single vendor providing the proprietary protocol.
>>
>> Could you give a specific example?
>
>
> My email client accesses 6 different accounts.  The point of commonality is
> my client.  Typical silo solutions are for that silo only.

See response to Cyrus.

> As being touted, that would mean 6 different downloaded, proprietary access
> clients, with my having to switch from one to another.  Besides being a
> pain, this means that they cannot share data, such as getting a message in
> on one account and forwarding it through another.

I can give you an existence proof of that working today but I'm
getting hesitant to name specific services.

Let's get away from using email as an example.  Even within the mail
world, none of the web-based mail services are just mail, although
they offer a just mail API (IMAP, POP).

> Although some independent clients
> have gotten pretty good at providing an integrative experience, it is not
> possible to do group IMs except within a proprietary context requiring
> everyone to have an account with the same provider.

OK, a good example.  The difference between this and mail is that
there is no back end connectivity.  With mail you can use any
(including proprietary) interface to your service, and it will get it
to other services.  Since there are standards in the _back_ end, there
don't have to be standard user interfaces in the _front_ end.  I think
this is finally a point worth bringing up.