 Mae'n ddweud petodo'r cwestiyna iawn i'w throughoutach, if you would like to move forward because we're going to having a discussion about it, welcome everybody and thank you for coming to this assessment, Can you ask other than me and Lydia how many of you were at the session in Montreal in 2017? Nobody okay so that's going to be interesting oh you were there good, three of us are them. So what I wanted to talk about today was the relationship between a'r rest gwaith y cyffredin cyllid yn gweld yn gwychwydd. A'r cyntaf yn dda, mae gallwn gweld yn dod mewn cyffredin cwrdd yn ychydig lle i'r cyfrifedraethau cyfroedd yn ddoedd. Felly rwy'n dechrau sefydlu arnyn ni'n gweld amser o'r biologist ac yn yn i'r byw. Mae'r busud rydyn ni'n gweld yng nghymchiol ar y bach rydych. Felly rydyn ni'n ei wneud os oedd i erbyn i'r bywll ddoedd y dyfodol yn y cyżfyr a'r byw. I'm an interested amateur natural historian so I participate there and what I've observed over the years is that there is a divide between the core of the wiki species community who are very knowledgeable taxonomists, subject experts, many of them professionals, some of them leading amateurs of high renown and the rest of the wiki media movement and I've said to them in the past that in a nutshell if we were setting up wiki species today we wouldn't do it the way it's done we'd use wiki data to host the data and then we'd produce a front end something like skolia for those of you who know it to render the data in the way that wiki species currently renders what is very unstructured data in the main part and we had a discussion about this in montreal in 2017 and people agreed that things could be improved and in some small ways on wiki species they have been so some data in wiki species now comes from wiki data so for instance images for most individual species rather than the higher ranks we simply put a template on the wiki species page that says put the image from wiki data here and we call in whichever is the lead image for that species on wiki data which saves the taxonomist the bother of trying to choose which is the best image or argue over it or whatever and we do the same for authority control for taxonomists who are written about on wiki species so there are pages on wiki species for each taxon for each species and genus there are pages for each person who named a species or genus and there are pages for the publications that described that species or genus so the authority control for the people is also pulled in from wiki data we had an rfc a couple excuse me we had an rfc a couple of years ago and there was agreement in principle to using structured templates for the bibliographic data for the publications and once that's done there's no reason that that data couldn't be pulled in from wiki data but that hasn't yet been implemented um and that's partly through lack of volunteer time of people who are expert enough to write the lure code i'm a template editor but i'm not a lure coder so i can't do that so i wonder and this is going to be a discussion i'm not here to present slides or anything like that but i wonder if any of you have any views on what i've described so far any ideas for how we could take things forward any reasons why we shouldn't take things forward or if you can identify any other issues that might stand in the way and we'll have to use the microphone because there are people watching remotely so if you want to speak well thank you for organizing this meeting um i'm Fernando Chubi from Argentina i am a biologist i work also in paleontology i'm not a taxonomist i used to use wiki species some years ago at the university teaching with wiki species and wiki data and wiki Wikipedia and then i stopped using wiki species um i didn't find it easy to to integrate with other other tasks of students um but um i was wondering was which was the future of wiki species uh i am agir to listen to what you are going to show us and and i am ashing an integration of information of Wikiouteata w Wwikipedia and wiki species so that i can refer to the same queue etc from wiki data for species of genius or whatever i don't know ni yn ni wnaidd seithを cael ei wneud when the taxonomy changes, which is very usual when taxonomist revises, groove, but never mind. I just cared to learn. Thank you. One question which comes to my mind quite often, when I look at the well established long lasting wiki projects is if the media wiki software is a good tool for them. I know that Wikisource contributors struggle with the software. I know that it's not the perfect way of showing and navigating images on commons. And there are internal discussions in these projects about, is it really usable? Do we do something about it? Do we build a whole bunch of templates and technical solutions around it? Do we leave it as it is? Do we suffer? Do we move forward? So my concern here is, is a catalogue-like project like Wikispecies a good content to be built around media-wiki software? I think it's a very fair question. And I think it's, without being rude, it's an academic question because it's what's being used. We do have a catalogue that is being built in media-wiki. You could argue that it would be better built in Wikibase and then rendered in media-wiki or wherever else. But I think the option of saying, well, no, it's not the right software. We need some different software, isn't it really one that we have the option to take at the moment? I think it's a valid question, and it's certainly something that should be born in my perhaps in the longer term. Not directly to this, but I would be interested who here actually edits Wikispecies and who edits Wikidata and who edits both. So raise your hand for Wikispecies. OK. Wikidata? OK. That means you three both. I think that is part of the problem. The people we probably should be talking to are not necessarily here. This is very true. I've suggested in the past to people on Wikispecies that they should be here at Wikimania. I've reminded them that remote participation was available. I've seen no sign that anybody's actually turned up. As I say, there is a divorce between a very tight-knit and very skillful community and the wider Wikimedia movement. And I don't know how to resolve that. I've tried and it's not easy. I think another aspect that concerns me is the model of the taxonomy that's used because it's not a field where there is absolute truth. It's like you might argue there is in physics or chemistry. I'm sure there are physicists or chemists who might disagree. But there are different agreements as to which are valid taxons and which should be merged or split or whatever. And it's noticeable sometimes that Wikipedia, Wikimedia Commons, Wikidata and Wikispecies don't necessarily agree with each other. And you sometimes have two, three or even four different versions of the taxonomy. It is possible in Wikidata to represent multiple versions of the truth according to different sources. The other projects, including Wikispecies, all attempt to represent one version of the truth. Life would be a lot simpler for everybody if the one version of the truth that was being represented by each of those projects was the same one version of the truth rather than a version of the truth. Sorry, was it you who raised your hand, Michael, or you? Even in perhaps one of the most well-studied, well-understood groups of organisms, birds, there are at least three official different lists of all the bird species of the world maintained by different official organisations, none of which agree, none of which are subsets of the other. And that's quite without the actual regional checklist, so the official New Zealand checklist of birds drawn by the New Zealand Ornithological Society does not agree with the Cornell or the other two either. So we don't, even, and that's in birds, that's a very well understood group. Now, in less and more obscure groups than who knows which particular classification it is we were supposed to be representing. At the beginning you said that you would not be getting into taxonomy, at least for that part of decisions I was holding back, but I'm sorry, it's time for me to be blunt. If there are multiple taxonomies circulating in the world, and there are, and if the solution adopted the wiki species is not sufficient to address them, and if wiki data addresses this issue sufficiently, I think that sort of answers the question of to migrate or not to migrate. I know that the taxonomies on wiki species have a version of the truth, how they resolve the issue of the different bird lists, I don't know, but they only represent one model. Whether wiki data is correctly capable of representing all the different models of taxonomy, I don't know. I said it's capable of representing different models, whether it does represent them, and whether it's capable of representing all of the models, I'm not so sure, but again I'm not in a position to judge that. But it's a very, it's clearly a very complex system. For the sake of our online audience. For the sake of the online audience, it's technically possible for wiki data to say that according to some sources the species belongs in this family and according to a different classification it belongs in another family. The problem is that you want to normally come up with a hierarchy, a single hierarchy you could use to start answering questions or running search results or something. I mean it wants to say which is the deprecated, which is the preferred version of that classification. Wiki data is still hierarchical and still wants one single nested hierarchy. So that leaves a question of, and I promise I'll shut up, I'll leave, and I don't want to dominate this discussion anymore. So that leaves a question of doing the footwork of actually setting up the taxonomies in wiki data if, as you are saying, the system is capable of accepting what you refer to as different versions of reality. Let's say, and there is no reality, there are just sources, right? So if wiki data is capable of accommodating different versions of reality. Now is the question of the footwork to either the footwork to actually reflect the multiple taxonomies in wiki data or employ an automated system about a script. Whatever to populate those taxonomies properly with the individual branches. Thank you, interesting thoughts. I'd like to extend Lydia's question. Lydia asked how many of you edit wiki data? How many of you edit wiki data about taxonomic names of living things? Can I ask how you find doing that on wiki data when it comes to the issue of different models of taxonomy? Does anyone like to comment? I guess I'm Alex from Australia. I've got a science background, but absolutely nothing to do with biology or anything like that. But I have found it interesting. I guess for me being quite ignorant, I would say not very knowledgeable on taxonomy and biology in general. But I still really enjoy putting in tax on authors and years that things were named. I guess for me it's just a difficulty of understanding things like recombinations and synonyms and so on and how to model those. I think it has got better. I remember a few years ago, there was nothing about how to represent those. I think it's one of those things like published books and so on where the model has been developed or iterated on and built over the years. The documentation gets better over time. If I checked now it would probably be better than the last time I checked. But I guess most of what I do is just putting in authors. That's one of those things I find really useful about the wiki data way. I suppose to say wiki species and wiki pediar is that it normally is just the surname appears and there's no actual link to the biologist who did the name or the classification. As in Infobox it will usually have no author at all. The point we're skirting round that you're quite aware of is that wiki data can't agree what a taxon is. So there's two different models. If we look at the common house sparrow, Pasodomesticus, little brown bird. Now there's an item for Pasodomesticus. Does it refer to a little brown bird? Or does it refer to the name Pasodomesticus which was coined by probably Linnaeus in the 18th century and this particular publication and so is a subset of this genus Pasor and what not. And some folks in wiki data really, really want to be able to say that Freddie the sparrow in a Hollywood movie is an instance of Pasodomesticus. But of course it's not because Pasodomesticus is actually a label, a two word label given to a concept of a taxonomic name. It may not even be a valid name. It may not even actually be used anymore. It might be defunct but it still exists in wiki data because you have to have all the synonyms. So what we have to do instead of say Pasodomesticus, Freddie the sparrow is an instance of an individual of the species Pasodomesticus. And the battle seems to be new no much more about this than me between people who want to treat the taxonomic names in wiki data as free floating labels with a particular hierarchy. But you can't say something like well this name has a range of Eurasia and has been introduced to New Zealand because you can't say that about a two word label. That's not a thing. It's the sparrows that have been introduced to New Zealand. The sparrows that currently have the label Pasodomesticus attached to them but they might change that label one day and go by a different name. And that's not really how wiki data, because wiki data doesn't seem to be able to agree on how it's going to model those facts. Does that seem fair to you Andy or what do you think it is? Sadly yes it does. Does anybody else want to add anything before I move on? Okay so we're over halfway through our half hour slot and we've identified a number of issues with wiki species. Around lack of integration with the rest of the movement or lack of integration with other wikimedia projects and possibly about the software not being appropriate. And we've identified some potential issues with wiki data but Lydia's here so they'll all be fixed very quickly. No some issues around modelling which may take quite some time to resolve. They certainly haven't been resolved easily so far. What I'd like to do in the next 10 minutes is talk about, I won't say solutions because these may be insoluble or very difficult to solve problems. But the next steps towards a solution. So what can we do to, as wikimedians, to be more welcoming to those in wiki species who perhaps don't see themselves as part of this movement. As people who do any wiki species what can we do to be more collaborative around that work. What can we do as wiki datans to improve the modelling of the taxonomy in such a way that it's more easily to integrate with wiki species. And what can we do on wiki species to use data from wiki data more effectively even if we can't do the taxonomic data yet. Because as I said at the beginning we started with things like the authority control for the people and the images of the species. So we did the low hanging fruit rather than trying to address the taxonomic modelling issue. And as an aside one of the things that concerns me is there is a lot of data in wiki species that is not being captured in wiki data or indeed any of the other wiki media projects. So if a species is in wiki species there will be a wiki data entry but as I alluded to when you spoke the paper that originated that species name and the author of that name may be in wiki species as unstructured prose. But it's not captured in wiki data, it's not captured in wiki media. How can we make wiki species more structured so the data can more easily be integrated into wiki data and eventually be served from wiki data. Anybody got any ideas? Well I'll put it another way. If you had to do something what would you do first? Pardon? Cyw? Cyw is on wiki species. It's very little used. Part of the problem is the rendering because the wiki species editors like to render their sources in a different way to the wikipedia from which the template has been copied. And we need a lure editor to help us modify the citation template so that it renders the correct way and then I think very quickly we'd get consensus to use that. But yes that's a good first step I think. Do we have any lure coders in the room who would be interested in getting involved with that? I'll talk to you about that a bit more later then. Anybody else got anything? Has anybody else got anything that they would suggest? How much are people on wiki species excited about visualisations and visualisations? As far as I'm aware that they're not at all, they just want a flat database or flat catalogue. I'm also an editor on wiki species and there has been work done but it had to be very labour intensive. I think they coded in Python so that it generates kind of like the family tree or like a bubble map of how things are related with one another. And I believe there was a lot of uptake but that was at least good ten years ago. So there was someone that was active in the community but given that the page constantly updating and there's images being added, removed and etc. So for now I would say it's presented in like a flat database format but there's definitely some interest towards kind of like a linkage or visualising kind of way. So in general to get people excited about wiki data and understand it, visualisations that show how this data goes beyond what is in one item but exploiting the actual graph is usually helpful to get people excited. And I was wondering if one opportunity would be to make visualisations more accessible. So this could be something as simple as a link in an info box to some tool that does a visualisation or even thinking about some visualisations on wiki but that maybe as a future step. I think so and I think again we're going for the low hanging fruit. We could perhaps make the skolia link for the authors more prominent and possibly even embed something within the page on wiki species and that would hopefully wet people's appetite to do more with it for other types of content. So I'll undertake to go and have a look at that. Anybody else? Well we have only five minutes left so I'm getting a signal from the back five minutes left but I'm quite happy to wrap up there if everybody's happy that we've covered everything that they wanted to discuss. We haven't had any feedback I take it from online. No questions? No? Okay. Well thank you all for coming. I hope that if we do this again in another six years we have a bit more progress to discuss than we have after the last one. But keeping good heart and be optimistic and thank you for your contributions.