[rfc-i] URL checking

Joe Touch touch at ISI.EDU
Wed Oct 19 14:43:37 PDT 2005

Paul Hoffman wrote:
> At 6:42 PM -0700 10/18/05, Joe Touch wrote:
>>I suspect the cost of manual checking by others is prohibitive
> How much would it cost? How much is prohibitive? Determining both of 
> these answers is fairly trivial.
> It should take about an hour to scan the last 50 issued RFCs and 
> extract all the URLs to get a count to determine the average number 
> of URLs per RFC. It is safe to estimate that resolving a URL and 
> looking at the first screen takes less than 30 seconds.

Add some time to check to see whether the page content is the intended
one; it's easy to make sure it's a valid page, but harder to confirm
it's the _right_ page. And this is another check that needs to be logged
 (that it has been checked, what the status is, etc.). It's more than a
few minutes per RFC, when it all adds up.

> Dividing the 
> average number of URLs per RFC by 120, then multiplying the result by 
> the dollars per hour the staff who is doing this gets paid, the 
> multiplying the result by the average number of RFCs per year, tells 
> the cost per year of the RFC Editor adding this service. This can 
> then be compared to what the IAOC considers prohibitive.
>> - authors ought to do those checks.
> We do. We also check our spelling and grammar. And the RFC Editor 
> does all those tasks better than many of us do them.

Checking validity makes sense is closer to spelling and gramar. Content
is probably prohibitive, as I noted, because it depends on the context
of the citation (add 30 seconds to find all of those for each
reference), and the semantics of the citation (more time...)

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 254 bytes
Desc: OpenPGP digital signature
Url : http://www.rfc-editor.org/pipermail/rfc-interest/attachments/20051019/6fcb6869/signature.bin

More information about the rfc-interest mailing list