Re: Fwd: Balsa default mail submission on TCP port 587, not port 25[major@satx.rr.com]

[<balsa microwave com>]

I'm not sure what anyone else was trying to say, but what *I* was saying
is that I beleive it is desirable to make any defaults (wether made
'default' directly by balsa, or by an underlying library) that apply to a
setting (specifically, port 587 for 'outgoing SMTP server hostname'), be
exposed somehow at the configuration dialog where that setting is entered,
so that someone making such a configuration without reading the
documentation (as occurs VERY often, especially for a user-oriented
application such as a mail reader), *especially* when the default is not
the 'currently well known and most frequently used' default (but even if
it was), and regardless of wether it is the 'official specification' for
the default.

It was discussed that balsa itself was not adding the default port to the
hostname, but that libESMTP was qualifying the bare hostname. I
acknowledged that the right thing to do would probably be to have a
'canonize' function become part of libESMTP that would accept the setting,
and return a fully qualified (incuding port number or service name)
setting that should be stored in the application configuration. I also
theorized that libESMTP probably did not hve such a function (yet).

To recap - I dont have a problem with the host[:port] syntax - I only have
a problem with there being no clear indication of what the default port is
*on the same screen the hostname is entered on*, especially when there is
both an 'most currently implemented standard way' and a 'new official
standard way' that are in conflict.

On Wed, 11 Jul 2001, Brian Stafford wrote:

> On Wed, 11 July 12:56 major wrote:
>
> Come to think of it, netscape uses exactly the same host[:port] syntax
> so balsa's behaviour is not without precedent.
>
> Or are you trying to imply that I'm so stupid as to have written what
> I claim to be a standards compliant SMTP library without having read or
> understood the relevant standards or made provision for the requirements
> of the standards?
>
> Brian Stafford
>