Helen MacDonald

and including stakeholders of transparency, of responsiveness and the idea that you might need to go back on your recommendations in a pandemic as new evidence emerges and to provide sort of an ability for people to feed back and disagree with you and say, have you considered this or the other thing instead?

So that was kind of where my thinking was.

And I was interested also to talk to Per van Dijk at the end of last week about his thoughts as a guideline maker.

I think you're sort of saying the same thing as Per, in a way.

I think he's saying, you know, gather the evidence, make clear what the elements are that might feed into your guideline process.

And if you're posting on guideline platforms, what that allows you to do is to adapt that for your local context.

So you might say in Africa, well, none of this applies to me at all.

And then you've been saved a job of doing all the searching for the evidence.

Or you might say there are a few things that are different and maybe we'll

So if we shifted into the evidence and away from the guidance, one thing that's been interesting me about how evidence is accumulating in the pandemic is the role of preprint servers.

So this is the concept that you can put up your research before you submit it to a journal, before it's published.

And the scientific community can see it.

It's publicly available.

And they can also comment on it, which might lead you to alter it before you submit.

And we haven't had preprints in a pandemic before.

And I was interested to hear from you.

Joseph Ross, who works with the BMJ and also with MedArchive, which is the BMJ's preprint server, and to hear how he felt preprints are shaping the pandemic.

I think that's quite an interesting point, Carl, that it feels to me like when you look at these preprint servers, you've got to know a little bit more about what you're doing to be able to spot those flaws.

It's not like when you read a paper that's been peer reviewed and some editors have handled it and improved it and maybe ask the authors to make the limitations of the paper much, much clearer, perhaps ask them to be more cautious in their language.

So I think it is...