[rfc-i] DOIs and RFCs

Joe Touch touch at isi.edu
Wed Feb 12 07:53:42 PST 2014


FWIW, this is a fine goal, but my experience is that universities each 
have their own culture for what they accept as a publication.

I just had a long conversation here at USC with someone who felt that 
RFCs weren't equivalent to journal publications, because journal pubs 
were scientific contributions and RFCs were 'just engineering 
specifications'.

(I pulled a 'respected' journal and showed how most standards-track RFCs 
were both as complex as much 'just engineering')

So while DOIs may help a very few, the more productive thing we could do 
if this were the goal would be to call (or republish) these as 
'Transactions on Internet Protocols' or papers in the "Journal of 
Internet Standards".

Joe

On 2/11/2014 11:40 PM, Eggert, Lars wrote:
> Hi,
>
> On 2014-2-11, at 23:56, Joe Touch <touch at isi.edu> wrote:
>> I don't see the benefit in creating another level of indirection for RFCs. We already have the rfc-editor URL as a way to identify them; citing them using an additional level of indirection through a separate service that may or may not be around in 20 years seems to be of no particular benefit and a lot of potential harm. E.g., the ACM Digital Library should link to the rfc-editor URL, which avoids dependence on DOI URL redirection.
>>
>> I similarly don't think its useful to include DOIs as part of citations.
>
> the main reason I have heard for DOIs is that some universities (e.g., some in the UK) do not count anything that doesn't have a DOI when evaluating the output of their professors and researchers. For those folks, having DOIs assigned to RFCs would be extremely valuable, and doing so may in fact attract more participation from academics, which IMO can only be good.
>
> Yes, the university departments are stupid to have this DOI requirement. But if we can make the problem go away by throwing a small amount of money at it, I'm all for it. IIRC this will be way cheaper that the bottled water for a single IETF meeting.
>
> (I agree with you that we don't need the referencing system provided by DOIs, or that they are useful to include in citations.)
>
> Lars
>


More information about the rfc-interest mailing list