ietf
[Top] [All Lists]

Re: query about ID/RFC statistics

2015-05-27 09:56:39
On Tue, May 26, 2015 at 6:44 PM, Jari Arkko 
<jari(_dot_)arkko(_at_)piuha(_dot_)net> wrote:

aren't these things listed in the XML and this a 'quick' xml xoath
parse away from win?

Yes, they are in some cases… the difficulty with getauthors has been the
special cases. The non-XML… the non-numbered sections… the oddly
intended stuff… the misspelled stuff… people’s names in different
formats… the people who misspell their names :-) or at least their
co-authors names :-)

You could of course say that we should ignore all that broken stuff.
I wanted to have a smaller error rate, hence included many special
cases. Almost all of this is data-driven, so once you add a pattern
line the tool will recognise it in the future.

Anyway, I do have a set of tools that i really have no time to maintain.

Which is very unfortunate, because they are super useful.


One group of tools is getauthors/authorstats, the one that collects
document statistics.

I often use these tools to figure out what drafts I have active, and
which need a pokin'.

It is operational, but if taken over by anyone
else it needs a rewrite. Despite being somewhat data driven, the
rest of the code is a hack upon a hack.

Another group of tools is the IESG statistics tools, which would
be very interesting, but are no longer operational due to interface
changes to how the data tracker presents itself. It too would need
a lot of work.

If anyone is interested in putting time on these tools, let us know!
For instance, we could start with a Code Sprint project in Prague.

I think that this would be great -- eventually I think that they
should become IETF / tools maintained...

W

The IETF also runs official tools projects that are funded with
IETF funds. So far I have not considered these tools so business
critical that we’d need to have them done commercially. And some
of them have been up and running through community effort,
i.e., me, Lars, and a few people who have sent me edits.
Let me know if these are so critical that they’d need a more
official IETF attention.

Jari




-- 
I don't think the execution is relevant when it was obviously a bad
idea in the first place.
This is like putting rabid weasels in your pants, and later expressing
regret at having chosen those particular rabid weasels and that pair
of pants.
   ---maf


<Prev in Thread] Current Thread [Next in Thread>