Oh yeah, the fact that certain wikipedia pages come under protection status 
definitely lowers their revert counts. And I accept that as well in my 
blog. 

Blog excerpt:* "This is more of a caveat. Some of these pages will have 
been placed under protection to stop getting abused. Protection could, for 
example, mean edits by anonymous users not appearing till they're approved, 
or if someone has been a registered user for less than 4 days old and has 
done less than 10 edits, they won't be able to edit certain pages. The page 
on Narendra Modi is under even stricter protection—only someone who's been 
a user for more than 30 days and has done over 500 edits can edit the page, 
and this has been the case since April. Such protections brings down the 
abuse levels for many pages, so that should be kept in mind."*

But then you could turn around and say, hey Shijith if you're aware of 
this, why didn't you factor it into your calculations? It's because I think 
there should be a limit to the level of complexity a journalist aims for in 
their story. I accept that I posted this in the datameet mailing list, but 
the story is aimed at a general audience.

Again from my blog: *"I've done work in the past that tries to be 
(conscientiously) rigorous, but I don't think journalism is the place for 
such work. Academic journals maybe, but not publications meant for the 
general public. In the pursuit of precision and conclusion validity, a lot 
of data journalism in India has become completely unreadable. I don't think 
there's anything wrong in admitting upfront that this post is a best-effort 
attempt, the conclusions may not be completely valid, but that hopefully 
this promotes the topic of wikipedia abuse in India as a legitimate avenue 
of research. And that some academic out there does a better job than I did 
in their paper. But I don't think a journalist should be the one doing that 
paper."*

As for the level to which abuse of certain wikipedia pages is organised, 
and how that coordination is done over Telegram and Whatsapp, because many 
of these groups are private groups, it's difficult to monitor that activity 
and find out which pages they are targeting.

But even if that's the case, I don't think I'll be missing out on any pages 
though. Right now the script goes through 150k pages (and once i've 
reworked the code, all the pages) which have been assessed by WikiProject 
India <https://en.wikipedia.org/wiki/Wikipedia:WikiProject_India>, the 
group of editors that maintains pages about India, and they cover *almost 
everything*. So no scope for missing out on disputed pages.

To your point about following OpIndia or other right-wing handles to find 
which wikipedia pages they are targeting, the only issue I would have with 
that is it won't capture abuse of pages that *aren't *a result of 
coordinated, organised campaigns. Abuse of pages that is a result of 
individual actors editing separately, but still devastating at an aggregate 
level, is also interesting to me. The right-wing agenda is pretty much all 
pervasive now, and people don't need to be prodded by politicians, media 
etc. to do their bidding, they pretty much act on their own volition now :(

On Wednesday, 22 December 2021 at 12:00:20 UTC+5:30 Shyamal wrote:

> Dear Shijith,
>
> This is very interesting work but I think it misses a lot of action due to 
> the rather simple approach to identifying dispute. An example is that calls 
> for action are often made from Whatsapp Group and on Twitter - such as 
> ("meat puppet") campaigns to make changes - for instance to alter the entry 
> on "Love Jihad" to read differently or for "Adam's Bridge" to be renamed 
> which have resulted in those pages being placed on protection - that in 
> turn drastically lowers edit revert counts. A script that follows 
> highly-followed twitter handles (OpIndia for instance) complaining about 
> Wikipedia pages might show up more disputed pages. 
>
> best wishes
> Shyamal
> https://muscicapa.blogspot.com
>
 

-- 
Datameet is a community of Data Science enthusiasts in India. Know more about 
us by visiting http://datameet.org
--- 
You received this message because you are subscribed to the Google Groups 
"datameet" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/datameet/ff7d7d96-427f-4bf8-91c4-4e9d5df8b36fn%40googlegroups.com.

Reply via email to