On Sat, 12 Nov 2016 05:59:36 -0500, mark lewis wrote:

 ml> while that may be true, it does so incorrectly... this is akin to
 ml> writing workaround code in your software to work around the bugs in
 ml> other's code... that's pure crap and bullshit... you don't work around
 ml> other's bugs... you point them out and get them to fix them... fidonet
 ml> found this out two decades ago... the internet hasn't reached that point
yet...

Good intentions...

I see no problem with a looser implementation provided doing so does
not break something else.  I would say that shoving fidonet messages
into NNTP is bound to break something anyway.

I haven't read the RFCs cover to cover and probaly won't ever do so.
I'll only read the parts that are relevant to what I am trying to get
working and then test a suite of readers against it.  Call it a 'best
effort'.

---