On 25 Oct 2012, at 01:25, Michael Richardson <mcr+ietf@xxxxxxxxxxxx> wrote: > Sabahattin Gucukoglu <listsebby@xxxxxx> wrote: > SG> * text/paragraphs (or whatever), a completely different identity that violates the length limits > > SG> Apple Mail and Microsoft use this text/paragraphs. > > Do you think it would be worth writing a specification for text/paragraphs? Only if we can get a whole bunch of defective code reading and writing it. There might be a case for that, given the laziness they've demonstrated in simply abusing text/plain, but it's also just possible that they might, just might implement format=flowed instead. But it's unlikely. :-( > Heuristically, it's not that hard to identify, and a small patch for > mailman would at least mark email as being in that format, so that at > least, IETF lists could have email that complies to some standard. It would be easier and simpler and probably more participant-friendly (though not, see below, quite as good for changing the running code :-) ) to perform imperfect conversions from text/paragraphs to text/plain; format=flowed. All we've got to do is identify any line longer than 80 characters and, assuming that it runs to the end of a paragraph at all times, add a forced soft line break encoded using F=F's stuffing rules. Any line terminated before it's too long is assumed to be a manually inserted hard line break. Now that I come to think of it, now is the time to see if the Tcl MIME parser will help me out writing a quick and dirty proxy server for my own machine … > (Whether or not we then drop email that doesn't have a text/plain part > is a second conversation) You want to punish Apple and/or Microsoft? That's how to do it. :-) Cheers, Sabahattin