the DDDWG has invited the commissions of the IOCR to do something that's impossible, by which I mean unmanageable. Wonderfully, Brian used the word possibility of convergence,

clearly is still there, and I still believe that. I think that what we've asked the commissions to do to define their metadata, deal with diversities, all of these things, I still believe is a reasonable request. Now, the Commission on Biological Micro-molecules chairman himself is here, and I'm logged to the fact that, or by warning, that I would

invite him to comment. But I think there is a weakness in my position on one aspect. And I think that when we look at the neutron facilities, back to 1973, for example, with the ILL, these predated are central facilities, the synchrotron facilities. Therefore, they

are the core celebra of whether it's realistic for the raw data back in 1973, which the ILL itself has said we can't guarantee that the software is there to analyse it,

whether it's realistic that we could assume it might be possible. So I agree that there is a weakness in my managerial position where I'm not certain in my mind that, let's say, the 1973 ILL data, for example, will be processable because

the expertise may well have retired, the knowledge of what is the metadata that surrounded the measurements of those data sets may not be available. So anyway, I'll throw that open as a point as chairman where I do feel that we are in a bit of a wobbly position

to think that we could hope that all raw data can be defined metadata-wise, software-wise. I mean, I've highlighted Matthew. I've put you on the spot there. I mean, would you like to comment on that or just say, how true, John?

I don't know what the files are like. I know there are files in those directories. I have to go and see if it's possible to do anything with those files and what

is actually contained. I can't really say any more than that. Just know that they're there. Okay, fine. So let's give Tom the chance. I mean, we do have agreement that within MX there is diversity, whatever I might quibble about the word explosion. So Herbert, James

and myself do agree that there's a strong diversity going on with the facilities generating the MX data, for example. Would you like to comment? I agree with everything you said, John. Well, maybe not everything, but most. Actually, just from the point of view of the Commission on Biological Macromolecules, I'd point out

that we strongly support the idea of identifying what the metadata ought to be and so forth. I would say I'm not sure that our commission is the right people to do that, however. And subcontract the task. Yeah, we might be able to subcontract the task. Yeah, exactly. But I think that we're totally in support. We're totally on board

with everything we're doing. Who's next? Mike Probert, Newcastle. I think you need to make a clear distinction, though. You just spoke about the possibility that data from 1973 might not be complete

and usable still. There is actually nothing we can do to change that position now. The only thing that we do have the capability to do is change the possibility to future prove our current data. And I think that's very much what we're about to try and ensure with respect to raw data and not on the process and derived data for the

very splendid efforts of the databases that are going on already. But the raw data is something different. Just a quick comment on the metadata definitions for the high-pressure commission. We are, I guess, pretty close for the metadata

definition for the diamond and we saw that the different techniques and the different pressure cells are still waiting. So probably it will be a stepwise process, right? So we'll have to wait. Maybe it will not come in just one paper or just one at a time.

And about the data from the history, not the raw data, but something like five years ago, paper, based on the structural factors which were actually deposited with paper

written in Akta in the early 1980s. There were structural factors deposited as copies from the microfilm, from line printer, which were almost illegible. And I asked for the

hard copies of the British Library and they provided me hard copies. I paid something like 40 pounds for the hard copies. But actually, there were some pages missing. So even the structural factors deposited with the papers in Akta are incomplete,

not telling about the raw data. So I had a summer student with me over the summer transferring the rest of that microfilmed, poor quality data onto the web. You can all enjoy that. I just wanted to respond to Camille in my capacity as chair of ComSips. I believe that

the work that has been done is sufficient that we should now make it a ComSips project to get that dictionary done. And if you like, that's an immediate outcome of this workshop. This work has now been done and we can go to the next stage of actually producing a dictionary.

And certainly it will be stepwise and it's intended to be an evolutionary thing anyway. So we should come up with a plan to work together. I must say, I'm in awe of these chaps like yourself who have got this command of these

ontologies and semantic whatevers. And it's great that you've come all the way from Australia and please lead us forward in the whole thing of the proper future use and future proofing of using the raw data. What are your plans for us?

I don't think there's much point in me making plans because so much depends on what everyone else does. So in this particular case, if the High Pressure Commission don't do anything then

nothing happens. I mean all we can do is provide an accessible way to develop metadata. And that's what I'm personally trying to work on, is to make it easy for people to produce these dictionaries. Because it seems to me that people are still, they find it quite, what Brian showed, some of the text from the dictionaries, it seems that people still find

that quite inaccessible. And one of the thrusts of my talk was that really you don't need to know all of that stuff, you just need to provide those four things. And the rest can be done by someone who's happy with the dictionary. Our keynote speaker who opened the whole thing

surely deserves 140 characters at least of her view of the matter. Okay, about metadata and if in future, especially regarding all the data sets made from ILL,

I don't think we really need to preserve the software that was used to process their data. All that we really need is a clear description of the formats that are being used. So if you want to go back to old data, that is the thing we should do, try to define what

the format is. In the future there's always possibility to analyze those data. So we should get things very clear. Also with respect to the metadata I was thinking, I mean if say 20 years ago someone thought okay let's put as a metadata the angle chi of the goniometer, then in future it may happen that no one really knows what that means.

So it has to be clear cut as you can. We need to describe in very well in depth what we mean by the metadata. So we shouldn't be having any confusion in the meaning of

the things we put in the metadata. Thank you very much. I think we can convene now for the 3.25 for final wrap-up session. Just before we sign off, the world at large has been watching

us over the last couple of days. I'm deeply indebted to our technical staff who have made the webcast possible. I don't know how many people have got out of bed early on a Sunday morning to tune in, but the presentations of the last couple of days will also be made available

on our website in the next couple of weeks when they've all been edited and spliced together. So when Rovin is but a hazy and happy memory to you, if you want to revisit any of the discussions and presentations over the last couple of days, that will be available to you. Thank you very much. And to thank Herbert Bernstein and John Westbrook.

Obviously they're about to be turned off from the web stream, so really appreciate their contribution remotely as well.