Mumsnet Corpus

[TokyoBouncyBall]

Not a TAAT, but a bit of googling as a result of a now deleted thread has led me to this:

https://fold.aston.ac.uk/handle/123456789/18

I note it says that the License is uncertain. Can you confirm that you have given permission for posts to be used in this way, or is there something that Aston might like to look into?

I note it says Users who wish to access this dataset must make a detailed application to FoLD and the researcher, as well as potentially gain additional agreement from an external organisation before they can be approved for access.

Given one of the uses it is being put to, I think it is a bit dubious to say the least.

Apologies if this has been asked, but there's a lot of pages to sift through since I last checked back - Has there been further updates from MN, or not yet?

Last I saw was about the chancellor saying they believe they have every right to the information here, did they ever answer the questions put to them by MN as Justine mentioned?

Not since the message to which you refer, no.

No updates, no answers to the questions.

I don't think we know the questions that MNHQ put to Aston.

No further updates since then.

I’m not expecting a quick turn around, VC probably underestimated how sticky our questions would be, and MN have hopefully hired a legal team to ensure that the questions asked cover all the relevant topics.

And if Aston haven't involved their lawyers, then they are even bigger fools than we think them already.

What I do wish we could get from @MNHQ is information about whether the corpus has been withdrawn from use in the meantime, because that ought to be the minimum response from the university.

I suspect lawyers on both sides are counselling silence.

I think so too. I think it's unlikely that we'll get updates from Justine quite yet. A lot for her/the legal team to consider (although I'm sure they're all reading the threads)

for you @JustineMumsnet!

I had a lovely lecture about how transphobic Mumsnet is recently. The evidence was weak and unconvincing. It's been mentioned before that complete obedience to the gendered soul orthodoxy is what is required. Reject everything else or you will burn in Rainbow Hell.

At least rainbow hell is pretty and has an absence of fire. Sunny with intermittent showers doesn't sound too terrible really.

I had a lovely lecture about how transphobic Mumsnet is recently. The evidence was weak and unconvincing.

I have had that lecture too 🙄

What I can't get my head around is the sheer audacity of secretly data scraping over a decade of data from MN without blinking an eye that they were breaking the terms and conditions of copyright. If they'd have paid for it, it would be worth millions.

The NY Times are currently suing OpenAI for scraping their articles to train ChatGPT. I've read that a settlement of $10 an article (in line with usual licensing for an article) would be $160 million. How many threads have been taken from Mumsnet by way of comparison?

The NY Times are currently suing OpenAI for scraping their articles to train ChatGPT. I've read that a settlement of $10 an article (in line with usual licensing for an article) would be $160 million. How many threads have been taken from Mumsnet by way of comparison?

I wonder if that's why Justine is still in the talking stage. Perhaps she's negotiating.

I'd be delighted to see MN get a shed load of cash out of this, on the understanding the data is deleted tho.

The lawyers may also be seeing it as an important test case that is going to influence the ways data way beyond Mumsnet can be obtained and used in the future. My gut feeling is that Aston have been really stupid and that their arrogance is going to blow up in their face.

Me as well. Quite the tirade.

By a new female colleague who doesn’t know me at all. (Discussing our social media guidelines amongst other guidelines)

Shame as I was looking forward to working with another woman.

I’m afraid I’ve judged.

I’ve just been reading about the NYT v ChatGPT argument, very interesting. Apparently if the ruling were to favour the NYT a worse case scenario for ChatGPT could result in a payout big enough to completely bankrupt Microsoft.

Obvs a U.K. case would be argued on a different set of laws, but presumably most countries are going to have test cases to settle these sorts of issues in the coming decade?

https://hls.harvard.edu/today/does-chatgpt-violate-new-york-times-copyrights/

That's possibly why the VC felt so entitled - what they are doing is so BIG he feels unchallengeable. But it has to be challenged at some stage, and it seems as though MN might be the one to do it.

This in The Times is interesting. Sorry I don’t have a share token.
https://www.thetimes.co.uk/article/reddit-considers-legal-action-against-ai-firms-for-unauthorised-use-pmp6t0t3vReddit considers legal action against AI companies for unauthorised use

Reddit considers legal action against AI firms for unauthorised use

https://www.thetimes.co.uk/article/7e87b54e-e709-452f-bb10-c0c8f9a7d4f4?shareToken=a374dd466e39e76e106ff1c6bfe3b62f

The CEO says that the company believes in the data being used for educational or research purposes but not for commercial gain. They do however see the potential for themselves in licensing this, and indeed, the value of the data they hold. See extract below.

Aston may say it's for research and education, but they've been awarded enormous amounts of money for this research, off the back of using women's data.

I'm not getting any less pissed off about this as time goes on.

Training in the extract above refers to training language models.
I wonder how many people who posted over the years ever realised that their anonymous online conversations about often quite sensitive topics (and I mean on here as well) would become the training ground for large language models and potentially commercially licensed data. I just don’t think that was conceivable some years ago.

Reddit are only taking legal action because they want to monetise the data. I assume Mumsnet will be having a similar train of thought.

Fine, but they couldn't do that retrospectively, they'd have to update their t&Cs and users would then have the choice. And I know what I'd choose.

What galls me about this whole situation is the fact that MN users are called 'transphobic' for having the temerity to want kids to grow up with as little fertility issues as possible, and that's why many MN users, especially FWR users, do not want kids medicalised with something that wipes out their fertility.

And the Aston blokes are there, laughing about a MN user's fertility issues.

It is just utterly vile.

Thinking back to the moment when you signed on to use MN, which for me I can't remember, you would have been presented with the T&Cs and would have indicated you'd agree to them to be able to use the site. Although almost all don't read these thoroughly, the gist would be that you stay within certain boundaries (or are removed), and that they follow the law (cyber security, data protection, GDPR, computer misuse for starters) and have guidelines that underpin the reasonable use of the posters, including intervention and consideration of things people flag up as offensive or dodgy, like 150 casino posts appearing somewhere. It wasn't conceivable, perhaps, but there were parameters on the forum that were trusted enough. The parallel entity of a scraped corpus shares, well, genuinely almost none of those qualities that I can think of, with the forum.

We don't know what the allegations of transphobia are in this students thesis.

What evidence is there that the people at Aston are "laughing" about fertility issues? If I've missed something then I apologise.

Your post is making an awful lot of assumptions and leaps.

Yes you have missed something. No it’s not this student’s thesis. Yes there is evidence. I could describe it in detail or go back and find the link but honestly since you are being all scolding and self righteous without even bothering to read the thread properly I am not going to go to the trouble.