Re: Call for Community Feedback: Retiring IETF FTP Service

[Keith Moore <[e-mail]>]

I'm very opposed to this proposal.

FTP is a much better interface than HTTP for scripting, mirroring, and 
remote file access (i.e. mounting an FTP server like a "share" so that 
it can be accessed from one's computer just like any other file system)

FTP is also a very stable interface, whereas as HTTP + HTML are both 
hideously complex.   Use of FTP as a remote file access protocol is very 
widely supported.

Basically what makes an effective human interface, and what makes a good 
interface for computer programs, are very different. HTTP+HTML is 
arguably an example of the former (though I think it still has a long 
way to go and needs to evolve toward less complexity rather than more 
complexity); FTP is better at the latter.

But IETF should fundamentally be about promoting stable interfaces.   
That is, after all, the purpose of standards.

More deeply, I think this proposal represents a fundamental 
misunderstanding about what IETF is, or should be, about.   We need to 
be facilitating tool building, not trying to stamp it out by forcing 
tool builders to use deeply flawed APIs.   The idea that RFCs are only 
supposed to be accessed through web browsers is incredibly backward.

Also, *don't interpret traffic volume as an indicator of importance*.   
Sure, if nobody ever used FTP at all, it would probably be safe to say 
it's not important.   But a small amount of usage can still be of 
tremendous benefit to the community, and HTTP simply doesn't work as 
well (in practice) for that.

Keith

p.s. now having said the above, I will admit that in the past my scripts 
have always used the RFC editor's FTP server rather than the IETF's - 
why not go to the authoritative source of the information?   But I also 
have doubts about the RFC Editor continuing to provide what the 
community needs, as the separation of interests between the RFC Editor 
and the IETF seems to be growing over time, and the RFC Editor has 
undermined the IETF on at least one occasion.   On balance, I hope the 
IETF continues to serve as an alternate trustworthy source of RFCs, and 
also that the IETF continues to make RFCs available via both effective 
human interfaces and effective, stable machine-accessible APIs.