 Mae'r ysgol fyddwyr o'r hunain yma, rydym yn ddau'r gwrth o'r wahanol, am fywr. Ac wedi'u pan fyddwn i'r awturdod yn fynd i'r eistedd, ac roeddent yn ddim yn fnwysgrifent i chi, mae'r yw ddechrau er mwyn â'r cystafeth a'r ynchyned â'r system, os y tuwch, yn fy wych yn amser, ac dydy'r cyllid eich bod yn ni'n gwybod, ac mae'n amser ond cyflodd yn ddau'r pierd. Roeddwn ni'n gweithio eich fawr am ychydig yn cael amser a'r argyrchu. Gweithio ar y fflaen iawn i cyfosio'r cyflwynt gyda'r gweithio'u heif~! Mae gwneud â'r gweithiooooo ac roeddwn ni'n gobeithio ar un sceisio ar gyfer y sefym Llywodol. Rwy'r pwysig yn felly'r gweithio ar gyfer yr archf Gyllidell. yr archfiu'r port oesol sy'n gweithio'r port oesol. Mae'r gweithio'r port oesol ymlaen nhw wedi'r ffawr hwn sydd oedd y bwysig ei wneud. Rhaid i chi'n mynd i ffordd sy'n bwylltio ar gyfer y Fyloedd, ond nid yw'n gweithio'r ysgolwyd Cymru, ond yn gwneud y ffordd yr ffordd, yw'r gweithio i'r Fyloedd yn eu gwahodd ysgolwyd. Yn oedd yn fawr bod chi'n fawr o ffordd o gwaith a wneud. Gweithio'r ffordd yn cyffredigol, yn cyfwilio'r cyffredigol, ac rwy'n i'n ffawr i'n gweithio i'r lleid o'r ddwyll. Mae'r ysgol yn yngyrch i'r ceisio, mae'n anghylo'i gael. Ond rydym ni'n meddwl am gael i ddechrau. Mae'n gweld yn odyn nhw, mae'n mynd i f向io yma. Mae'r yn yn iawn rhoi'r ffyrdd i Gweithdoedd yma o un draes yr un dros, os ysgol yw eu bod ei symud. Fy Beth fel ydych chi'n cael eu bod yn ddefnyddio'r gwahau sy'n cael ei wneud gyffredinol. It's not designed for the great unwash to communicate their ideas to the scholarly domain. It's designed for communities to talk to a first approximation within themselves. The other thing I was going to say is it's also embedded in a set of other communities with other tools. For example, we have very close collaborations with the astrophysical data service at Harvard and the Inspire, collaboration involving Slack and others. We really try and sit within this environment of many services to be part of that system and I think of annotation of how can we interact with other players to do parts that we don't have to do them ourselves. I see two key use cases that I would like to think about. One is how do we make archive work nicely with these great tools that have been developed for people to take their own private notes or share them in small groups or whatever. It seems, I think, from what I'm hearing, relatively straight forward. But then the really big one is how do you enable sort of more open commentary and discussion? As soon as you start talking about that you get into tonnes and tonnes of issues. About control, ownership, identity, authority, what's the community ready to do? That's I think where we'll spend most of the time. Felly, yn fwy o'r ffordd, y rhai i'r Llywodraeth a'r Llywodraeth Cymru yn ymweld yn y rhaid o'r prynnwys. Felly, rhai i'r Pdf, maen nhw'n ddweud o'r ffordd. Mae'r ffordd o'r rhaid o'r ffordd yn ymweld yn y rhaid o'r ffordd. Ymwynt allan o'r rhaid i'r pwylltau o'r pwylltau o'r prynnwys. Mae'r pwylltau o'r pwylltau o'r pwylltau. Mae'n ddweud o'r ffordd o'r ffordd o'r ffordd. Elin o'r gwybod yw'r hun, felly mae'n fawr o maen nhw'n ddweud o'r ffordd yw ymweld yn ymweld yn ymweld. A'on gymryd o'r naturículoch ar y tydd, ac maen nhw'n ddweud â'r rhai i'r fan na yw'r ffordd o'r pwylltau o'r ffordd. Felly, mae circumstances archive yw, mae'n ymweld yr ydych yn y gusagig i'r ysgolio. Felly, roedd adher yn gweithio'i rhai a'r grynnyd, mae'n gwneud 90% sefydl yn ymweld. Ychydig yn ymweld yn maen nhw'n ddweud. mae'n dweud i'r ddweud yn y mwynhau lleolau fflexible yn ddefnyddio'r ffordd. Fe wnaeth bwrdd iddo yn dweud y ddefnyddio'r ddweud yn ei ddweud i'r ffacbook app, ac mae'n ddweud bod yn dweud i'r ddweud i'r ddweud o'r cyfnod, a'r ddweud o'r cyfnodau nad oedd yng Nghymru, ac mae'n ddweud i'r ddweud i ddechrau'n ddechrau'n ddweud i'r ddweud i'r ddweud. Un ddechrau i ddechrau y ffordd, yn rhoi'r cyfnwyshwyl drwg Zeitbydd i Rice Book Api, sy'n rhai'r cyfnodd ar y dddolad gyda siŵr y cyroedd gwaith. Mae'n gallu gymryd yw'r ddiwedd i'r maen i'r newid gyda'r ffordd neu wrth yr arfer yn croncau i newid i'r cefnodau ar gyfer y ffordd am y cyfnodd api iolaeth gŵl. Nid yn ei awrs i'r ddim yn ymddillog, gyda'r plans rwy'r gydig, a'r ll?] dyma'r gwylio, I would say that science depends upon argument and critical review to establish truth. People who discuss what science means might want to nuance that a bit. But while services like Archive have changed the openness and speed and the linkability, which is pretty important, they haven't really changed the granularity of discourse or the way it happens. Everyone who looks at this says that, yes, there should be commenting and discussion systems associated with articles. People have been requesting this feature for years, and it is absolutely terrifying. The reason it is absolutely terrifying to us is that this is the graph of the number of submissions per year. I apologize that my plane didn't have Wi-Fi, so it's not extended. I had only the data on my laptop. So about 80,000 submissions per year will come through Archive, and we have approximately two humans to help these users with all their problems, which means that we don't give them much help. The budget per article is about $7 for the whole of Archive, the tech side and the user help side. So it's a completely different scale to the sort of journal publishing world. Comments really get people riled up. We did this experiment with blog links when trackbacks came out. Our first stab was this little thing, I have a point of five blog links it says there, right? It's pretty hidden, one is very soft start, and it's still like that actually. The reason is, our first reason was, oh, we don't know how to fit that into the screen real estate. The second reason was, well, when the titles of the comments are crackpots, contrarians and free market ideas, and we already had complaints that these are sort of connected but hidden away, and we thought of having that, you wrote that, what next to my paper? And dealing with those problems is the one that really scares us. So I think that's a hard thing in the sort of completely open, someone can write something on my paper that I really care about. It's actually kind of interesting if you follow those links. There's a very interesting discussion by Jack Dyslar on the Archive trackback system and things and it gets into a slinging match against particular people and further down that thread it says, okay, we're even having to cut off this comment thread on this on the blog because we're starting to talk about Lubos Motel instead of the commenting system. So what do I want to get out of this? I would like to discuss with people a bunch of sort of techie issues about how I best reflect Archive structure out in the linked web to enable annotations, and happy to talk about that, but the sort of bigger scale is I want to get a sense, which I'm already getting, of what are the best tools to support the sort of personal and closed group annotation. I'd like to really explore the boundary between the web and PDF and find out how you can make those two work together. I'm really interested to see some of the notions of annotations tied to particular points in tech so they support multiple formats, but in the scholarly world, I think PDFs are going to be with us for longer than we might hope, so we'd better make them and the web work nicely together. And secondly, I really want to talk with people who have ideas about how we support more open annotation in the scholarly world without having this huge user support nightmare of dealing with, I don't want that next to my paper. And to me, that seems to rely upon some notion of sort of limited communities, Facebook in the scholarly and open world, and how does that work? There are a huge number of people trying to build social network systems of the scholarly world, but of course, everyone wants to build their own system, trap people within it, and it seems very hard to make one of these work in a sort of distributed system that is open and everyone can buy into. So, thanks very much.