Corpus 2

[TokyoBouncyBall]

A summary would be good and I might do one later but Aston, data scraping, astonishing lack of contrition…

A thesis attempting to reach a desired conclusion using a corpus that includes material written by contrarians setting up straw men, material written by experimental AI, material written by bored posters arguing with their own socks, material written by students manufacturing quotations they can use in essays, material that can't be used because the creator has explicitly withdrawn permission.

I'm not an academic either, but I do have PhD and a few publications and I know just how much hard work went in to keeping what I wrote honest, and how rigorous the criticism was along the way. I cannot see how there is any validity in this exercise, and feel deep dismay that a discipline exists that permits - even encourages it.

And by the way that is what limits my sympathy for the student involved in this, because she must know just how flawed and fake her dataset is.

Unless she's really stupid and intellectually lazy, in which case she probably shouldn't have a primary degree, never mind be setting out to get a doctorate.

I have just spent an hour on the phone trying to get my pet insurance premium down because 5 years ago my dog had an expensive operation on his anus. It appears I can’t get another company to insure him because of his op. His anus is literally costing me an extra £3 a week. I may have spoken about it on here at the time. I can’t remember. I may have used the term arsehole. Or bottom. Or arse. So you’ll have to search them all. And I may have changed his sex on this post or the post I originally posted on. So you have to search for both sexes. But I promise the increase in pet insurance bit is true.

Aston - You can have that link free.

I think you absolutely could write a PhD thesis on the rise of gender critical feminism since 2008, and locate this within broader social and political change, and also consider in the context of previous ‘waves’ of feminism (although I have an issue with the concept of waves, because the standard waves do not include much of the early twentieth century activism on maternity and antenatal care and earlier campaigns for contraception and abortion (although some of this was tied with eugenics which makes things a little murky), but that is a disgression). You do not even have to agree with GC feminism to do this, you just need an open mind and a valid approach.

I also think you could consider the role of MN in this and the way in which language is used to both create community and to organise and engage with political activism and debate and indeed, to address bad faith actors. But to have any valid comparative value, you would need to look at other media and other discussion fora, and it would also be necessary to start from an open research question and not an assumption of ‘transphobia’ (how is this defined?). You would also need to understand how the site works, its history and most importantly, have consent to use it and any material quoted from it.

At least, these are the things I would be thinking about for a PhD on MN, and I would also be thinking that Sarah Pederson (hopefully I remember her name correctly) had already staked out the territory, so to speak, so what would I be adding?

What I would not be doing is comparing women posting to extremists based on a corpus of threads and research questions which were biased from the outset whilst riding roughshod over the issues of informed consent.

I, too, wondered about how this could form the basis of a PhD. However, as a linguistics graduate (a gazillion years ago now), I did assume that perhaps there was going to be some complex linguistic analysis as part of this, rather than just more straightforward analysis of the words being used. (I have no clue what that might be!)

Enough of behinds. I spent 5 years at uni and still can’t get over the ‘quality’ and small number of references to support the thesis.

edit: Pink News and Vice from what I recall

But surely the first job of the complex linguistic analysis would be to see whether it was possible to identify the 'real' opinions of the 'real' posters.

And only then to try to explore where (and indeed whether) they fit some template of 'phobic' views. There was no suggestion in anything I've read so far to indicate that that was going to happen.

It may be that it was going to be a PhD by publication rather than by thesis.

I think it is likely that it would be trying to define ‘linguistic transphobia’ in some way / doing some theoretical work related to this - makes the choice of data even more dubious.

It may be that it was going to do something interesting methodologically in terms of corpus linguistics(I doubt it with those supervisors) or more likely try and do something methodologically different for corpus assisted CDA.

It makes little sense for this team to be the supervisory team unless there was going to be a focus on linguistic ‘hate crimes’.

I would not expect that the research as outlined in the ethics would in and of itself be enough to write a whole thesis without some further focus on a methodological or theoretical development.

I think a study of the usage of the myriad words for arse from a socio-economic perspective would have more merit than Eden's proposal.,

But you love me and my expensive anus don't you mum?

Please tell me you still love me!

Please!

I'm sorry my anus is so expensive.

I'll forego the doggy treats!

Just as long as you still love me!

Tell me you do!

You do don't you?

Woof

.

Yes I agree about the hate crimes bit. After it has been published, and the mumsnet quotes are searchable easily on the search function to link to the username, what happens then?

As Aston think an actual crime has happened, do the police get involved and ask mumsnet for the real identity of the poster? If Aston think they’ll uncover crimes surely there is a moral duty to go to police before it’s even published? Otherwise it’s withholding evidence is it not? Or is the hate crime only a crime if the named victim says it is? In which case, the most likely named people to get notified are celebrities who can complain to police. I don’t know enough about hate crime but it all sounds the wrong way round - if you know crimes are being committed.

Obviously if they can’t find any hate crime then the research title is going to look silly. And there are lots of posters who are trying to stir things up on the feminist board so if they ‘catch’ one of those or even a poster who then says their posts were just a wind-up, surely that negates the research too after the fact? The only way of ensuring the crime has been committed is to put the criminal before court. So convictions need to come first, surely?

For those posters (e.g. Glinner) who posted as an identifiable human, isn't the term 'linguistic transphobia' actual libel / defamation by Aston?

So identifiable individual posters who did not give consent ( Glinner, Sarah P, DNK, Jo P and others) could all sue Aston for copyright breach, defamation and data breaches under DPA / GDPR?

Maybe we need to crowdfund.

😂
He is a rather gorgeous dog. And very well loved. But the butt of all jokes is the bloody insurance premium!

An interesting article on data scraping and its effect on the arts:

https://thecritic.co.uk/ai-and-the-great-data-robbery/

Just wondering if there's a timeline for any updates on this, @JustineMumsnet?

@JustineMumsnet
Please can you confirm that the dataset scraped for the PhD is not within your terms and conditions and that you have not given permission for it to be used? This is what I understand the position to be, but I would like to be sure.

https://www.linkedin.com/feed/update/urn:li:activity:7201199204860391425/

I know that if we post on Mumsnet, you 'own' our words. But this kind of thing, in the context of the whole Aston University palaver, really worries me.

We have, as users, no idea what threads you're deciding to use for this.

Interesting. I’m glad I’m not paying to be a premium member with stuff like this.

It’s mildly amusing though. Like the AI used to summarise reviews like on Amazon:

’Posters agree that the drive-blocking neighbour is a cheeky fucker, OP should block him in, and a diagram is required. That said, opinions are mixed on helpful rugby team moving it, accidental-on-purpose scratching and penguin bollards.’

MN don't own our words, we retain copyright but - I assume - we grant a license for Mn to use them when we use the site. I'd need to reread the t&Cs.

MN users are the content providers; our words are the product.

Coughs...

Actually...

Mumsnet DO OWN our words! We still retain copyright over our own words but Mumsnet also owns the copyright on our words and can use them however they wish in perpetuity 👇

This web site and its contents are copyright Mumsnet, all rights reserved.

By submitting User Content to us, simultaneously with such posting you automatically grant to us a worldwide, fully-paid, royalty-free, perpetual, irrevocable, non-exclusive, fully sublicensable, and transferable right and license to use, record, sell, lease, reproduce, distribute, create derivative works based upon (including, without limitation, translations), publicly display, publicly perform, transmit, publish and otherwise exploit the User Content (in whole or in part) as Mumsnet, in its sole discretion, deems appropriate. We may exercise this grant in any format, media or technology now known or later developed for the full term of any copyright that may exist in such User Content.

Yeah, that's a license. We retain copyright. MN's copyright is the website and content, separate from User Content. Copyright is quite a complicated beast, right enough.

Yes, this is why it matters what the terms and conditions are and whether data scraping, as has been done here, are part of these (as I understand it, not), and whether MN has given permission for part of the site to be used for this research.

It also matters on what grounds, without such permission, Aston thinks they have legitimate use, despite the content they are lifting being sensitive in nature and concerning children and vulnerable people, when posters cannot possibly have known their words would be used in this way (as it is against the terms and conditions to data scrape and MN, as I understand it at present, has not given permission).

I reported my own post yesterday morning asking about the first paragraph and await a response.

Both Newcastle and Manchester Universities are running late with responding to P. Murray's FOI requests. I was hoping for a rather more interesting reply than Aston's effort, sat down with my cup of tea, clicked on my bookmarks....and there's nothing new to read. :(