 It's time to get started on our closing plenary. And before getting to the main event, I would be terrifically remiss not to thank several folks. So I'm going to ask you to join me in doing that. First off, I wanna thank all of the speakers. I went to some incredible breakout sessions at this. And I hope you found some too. I really thought it was just an extraordinarily good and diverse set of breakouts. And I am deeply grateful to all of the people who came and put time and effort into making us all collectively good deal wiser now than we were when we arrived. Thank you. Second, I would like to thank some volunteers from Georgetown as well as Diane Goldenberg Hart's daughter who's not from Georgetown who helped out with some of the audio visual capture. In recent years, as many of you know, we've tried to really expand the amount of the number of sessions that we capture audio over PowerPoint. And there are too many of them for us to do by ourselves. So in recent meetings, we've been greatly helped and our whole community has been greatly helped by having a group of volunteers to assist us with that. So I thank them very much. And finally, I just wanna say a big thank you to the staff. They really did a tremendous job making this run smoothly. This is our first time in this hotel and so there was lots of things to learn and sort out and debug and we got a few things not quite right and we'll get those right next year and mess up some new ones. But overall, I have to say this just went amazingly smoothly and I really thank them for their hard work in making it look easy. Thank you. Now on to the main reason you're here because you're not here to listen to me thank people. So in 1996, Paul Evan Peters, the founding director of CNI died suddenly and unexpectedly and as we came to grips with that, a number of people wanted to do something in his memory and there were two things established. One is a scholarship which funds a master's student and a doctoral student. In a school of information or library science program. The other was the Paul Evan Peters Award and that was funded by ARL, EDUCAUSE, Microsoft and Xerox and we have given about, I think you're gonna be the sixth if my memory serves. Of these one every couple of years and they really recognize deep, significant and sustained excellence in some area that's really changed the, been a major game changer for all of us. The people who have gotten this award are all very distinguished. They've had amazing careers and Herbert is no exception, except in one particular I think, well, two particulars. I think he's still got a lot of amazing career ahead of him and also he really I think is genuinely the first of a new generation who did their work largely after Paul had gone. I think you can see at least the beginnings of the effects of and contributions of all the prior awardees in Paul's time. But Herbert really is the first of a new generation of important leaders to be recognized and that gives me a lot of pleasure too. You have in this nice little brochure here a good picture and a biography of some of Herbert's contributions. I would just say two things about those contributions. They're well enumerated in the text here. The first is that many of the people we've recognized have done sort of one good thing that was a very good thing and that had a lasting effect. Herbert has had one of the most fertile careers that I can imagine. He has just continued from one fabulous achievement to another and indeed built one on top of the other in a very creative way. People who can do that are very special. Most of us would be I think quite happy to have one really good idea. So that's one of the reasons I say somehow I think you've still got a lot of wonderful things ahead as well as behind you. I also take great pleasure in this because in a certain way I've really watched Herbert's career takeoff. I had a really wonderful unusual privilege of being an external member of his thesis defense for his doctorate at the University of Ghett and that was probably the first time I really got to know him pretty well and I've just been following his work ever since and so it's given me great joy to watch that whole trajectory. With that I ask you to join me in congratulating Herbert and as is traditional with the winners of the Paul Evan Peters Award. He will close the conference with a lecture in Paul's memory. So congratulations. Thanks a lot Cliff for your very kind words and thanks also to CNI for all the support that it has given all my projects over the years. So this presentation Cliff has been on my mind for numerous months well ever since you've actually sent me that email announcing that I had received the award and would I be willing to accept it? Like really? So thanks a lot for that of course. I wanna express a few thanks in general also for the people that have helped me achieve where I am now. First of all of course thank you to those that have nominated me and those that have endorsed the nomination. I wanna say a special word of thanks to the two people that initially brought me to the US in the context of the PhD research that Cliff actually refers to and those were Diana Markham then at Clear and Rick Lewis then at Los Alamos National Laboratory. They started this enormous 20 year adventure for me. I also and this may sound unusual I also would like to thank my employer the Los Alamos National Laboratory. As many of you I'm sure know that working for the federal government is not all that easy big understatement but then again I've been given enormous opportunities at the lab despite all the constraints and I'm not sure that I would have found another institution in which I would have been able to develop my research agenda in the same kind of way. This work means a lot to me because I consider it to be the highest recognition that I can receive from this very community that I've been active in for all these years. I need to with that regard thank all those people that I've collaborated with over the years and of course I'm not going to thank them all individually so what I did here is I created this virtual trophy that I offered them and what you see here is a word cloud and it's actually based on all the co-authors and all the titles of my publications and specifications that I've created over the years. Clearly this doesn't list all the people that I've collaborated with but definitely those that have had the highest the biggest impact on the direction of my work. This award means especially a lot to me because three people that have inspired my work most over these past two decades have previously received this award. That's Tim Berners-Lee, the inventor of the web, a fundamentally new communication medium that's Paul Ginspark of the Physics Preprint Archive actually initially known as xxxx.leno.gov. He early on actually in the days before the web understood that digital network technology was able to fundamentally transform scholarly communication and then the third one is Brewster Cale of course the creator of the internet archive who early on in the very early days of the web understood that the communication medium was going to be so important that it just had to be archived. Interestingly my work is actually situated on the connection between all these people. On the axis that connects Tim and Paul you have all the work I've done regarding interoperability for scholarly communication. On the axis between Tim and Brewster you have the memento related work basically added the time dimension to the web and integrated web archives into the fabric of the web turn them into infrastructure rather than into destinations. And then on the connection between Brewster and Paul I have my quest to try and use web archival paradigms to archive the scholarly record that itself this notion itself pioneered by Locke's and David Rosenthal. So I'm ready to start the talk now and the talk will be about scholarly communication obviously and the decentralized web. And I'll first talk a little bit about what the decentralized web is all about. I then will motivate why I considered it relevant to apply or try to apply decentralized web concepts to scholarly communication. I will then share some inspirations that I picked up out there related to this quest and then I will end with some very bold speculations. As a matter of fact, a lot of that, a lot actually of what I will touch upon is big sky thinking. And I would like to invite you to sit back and relax and you just join me in this speculation. Forget for a moment about that report that you need to submit to the provost next week or that library system that you need to get off the ground in the first quarter of next year. Just come along, follow the line of thought and dream with me about a potential future that will most likely never exist. So the decentralized web. So first of all, I mean, you all know the web is inherently decentralized, a distributed system but still a couple of things happened so that there's some kind of a movement that has emerged that is trying to re-decentralize the web. To be honest, this was the topic that I chose to pursue in my sabbatical and I didn't find it all that easy to come by concise information, descriptions of what is all was about and I had to really dig myself into technical specifications, white papers and all to try and understand from the technicalities what this was all about. In case you're interested in this beyond what I will share with you today, I highly recommend a set of videos that were recorded as part of the decentralized web summit that Bruce R. Cale organized about a year ago at the Internet Archive. There's a lot of good stuff there. The thing is that there are really two strands going on in this decentralized web work. They share a concern and it's a concern about the enormous power and control that certain nodes on the web have established when it comes to access to information, privacy, censorship. It's concerned with a single point of failures also. And so the two first on the slide there, distributed file systems and blockchain, they're all about tackling these challenges by introducing new protocols, okay? They're also concerned with building, archiving and trust into the very fabric of the network. This is pretty interesting. I have, and there's actually really interesting work going on also specifically related to research and scholarly communication. I decided not to pursue this path because of the fact that it builds on totally from scratch protocols. And I thought understanding that scholarly communication hardly has even adopted the ways of the web. I thought maybe starting really from scratch is a bit a stretch too far. So that's why I decided to dig deeper into this other thread which more follows along with the lines of the web and is based on the HTTP protocol. This is work mostly coming out of the W3C actually. It's exemplified by the MIT Solid Project which I will talk a bit more about later. So the, this track that I studied finds its origin in one of these legendary notes that Tim Berners-Lee puts out every now and then, you know, like there was the one about linked data. And so this one is called socially aware cloud storage. And the consideration there is that over the past, let's say decade, the web has become dominated by a couple of massive portals. I don't really need to name them but I will Facebook, Twitter, Google and so on. And so these portals, and there's of course way more. These portals offer a very smooth service. It's free at the point of use, you know, but it comes of course with the consequence that the user, the user's behavior, the user's information is obviously the product that is being sold. It comes with a couple of other problems that come from this monopolistic kind of nature of these portals. The user has no mobility. It is not possible for the user to pick up its data from one of these portals and just drop it in another. There's hardly no interoperability between these systems. They have their own APIs if they even have them but they don't really interoperate. And then another aspect is that the functionality that is provided by these portals is typically constrained to the resources within these portals. So it's not like you can apply the functionality that comes with one of these portals and take it over here and apply it to resources in another portal. So there's this kind of connection between these things. So this note about Tim Berners-Lee is really a call to arms to put the user back in the driver's seat. And doing this by basically giving every user their own web environment in which they self-publish their information and where they have total control of their own information. It also comes with the notion of disconnecting between the application and the content. And so it's the notion where an application can exist typically browser site and it can work with information that sits there on different kind of pods. And it is up to the user to grant access to their own information. And this is done by means of web-based access control lists. So as I mentioned, all of this work is exemplified by the MIT Solid Project. And without going into technical details, I wanna illustrate by means of a scenario what the essence of the ideas really is. So I need you to listen very carefully to this because if you're not along with this, you won't understand anything of the rest of the presentation. So the idea here is literally that the user has their own web domain. So here's Alice and she owns Alice.info. And in that web domain, she publishes a profile. It's a machine-readable profile that lists some of her personal information. For example, what is her preferred web identifier and what are her contacts on the web? Also expressed in terms of web identifiers. Now in that domain, Alice runs her own storage environment and it's called the pod, POD. I will mention that term a lot in the coming minutes here. So she operates that pod. And that pod is actually, you have to think about it as it's a web server on steroids. So it obviously it offers storage of Alice's personal information, but it also comes with authentication and authorization. And there's also a social component to this by means of this inbox. So there's an inbox where Alice can receive notifications from others on the network. Now with regard to authentication, it's not like we need to give every user in the world a username in Alice's pod. The idea for authentication is it's a global mechanism and once the mechanism is implemented, it can be used across different of these pod systems. So it's not one of these portals where everyone needs to go and get their own login account. So as I mentioned, there's also these applications. They typically run in the browser and Alice runs one of these. The application interacts with the pod so it supports all the standards that are supported by the pod. And using that application, she creates a document and she publishes it in her pod. She decides to make this public by use of these access control lists and how it's visible on the web at a certain URL. Now in comes Bob. And Bob actually reads, he discovers and he reads the document that was published by Alice. Now Bob has also drank the Kool-Aid of pods and so he also owns his own domain and he publishes his own profile and he runs a pod. Now this isn't necessarily the same pod. It could be another implementation but it's the pod that has the same functionalities and it supports the same standards. And Bob also has his preferred application to write textual information and uses it to write a comment to the posting that Alice did. And he publishes it also on the web in his own pod. And this is where it becomes interesting because we now have Alice's document and Bob's command to Alice's document. And these things are not connected. One thing that we've gained here is that both Alice and Bob are completely in control of their own information but the information is not connected so we need to solve that. Well one part of the equation is very simple of course. When Bob writes his comment about Alice's posting obviously he's going to link to it so that piece is solved. Now for the other part, this is where the inbox notion comes in. Remember Alice has a machine readable profile and in that profile you find information about the location of her storage environment but also of the location of her inbox. And so Bob's application figures that information out and can now send a notification to Alice's inbox basically saying hey, I wrote a comment about your stuff. And if all is well, that then allows Alice to link back to Bob's command. And as in the end state, basically we have these things that are bidirectionally linked but both of these parties have retained control of their own information. Now when I say there is a link from Alice's document to Bob's comment that can just be a link but that could also be that every time that Alice's document is rendered, Bob's comment is dynamically inserted as a comment into that document. Bottom line remains the same, they both remain in charge of their own information. That's a bit of the core idea behind this thinking, Tim Berners-Lee's thinking about his decentralized web thread. As I said, these pods are web servers on steroids. They support rewrite, delete operations, authentication, authorization and all of that is based on standards. So for example, for rewrite, delete, it's link data platform when it comes to identity, it's web ID and friend of a friend, authentication, web ID, TLS, open ID connect and access control is for authorization purposes. So Bob's analysis pod can be different implementations but they all support this standard. Same thing when it comes to notifications, that is also standard based and there's a spec called link data notifications that is being used there and as a payload of the notifications which are structured by the way and machine processable activity streams too could be used. Again, another W3C standard. So when you put all of that together, an ecology emerges or can emerge that looks a bit like this. So you have all these figureheads here, they all have their own pod and because they self publish in their own pod environments, you end up with a document network that it obviously has interconnected documents just like on the web in general. But in addition to that, because of these inboxes, there's a social network component that is being added on top of these storage environments and this is something that's going to be making it very interesting for what I'll talk about when applying these ideas to scholarly communication because this is actually the component where one could introduce discipline specific connections. So I started reading about this decentralized web movement and thought, well, it would be interesting to kind of see where we end up if we apply these ideas to scholarly communication and actually had two rational motivations that decided me to really go there. First of all, the web as Tim Berners-Lee describes it in his note on socially aware cloud storage and the scholarly web have at least one thing in common and that's monopolies of big players. And then the other thing is that when looking back at the Republic of Letters, when scholars were literally sending letters to each other expressing their ideas or expressing feedback to their peers' ideas, one is actually reminded that scholarly communication is an inherently social activity. And I think we've forgotten a little bit about that through the industrialization of scholarly communication and through the introduction of a whole bunch of intermediaries between author A and author B. So these were the two rational motivations that made me decide, yeah, why don't you go look into that? And then there was a third one, which is not all that rational, it's way more emotional and it's my deep frustration with the status quo of scholarly communication and with the fact that in the past two decades that we've all worked very hard to try and introduce change, we haven't actually, in my opinion, achieved an awful lot. Just two illustrations about that. Throughout my entire career, starting as a systems librarian at Gantt University until now working as an information scientist at the Los Alamos National Laboratory, this, the serial crisis, has been with me, has followed me along. This is not a crisis, this is a very unfortunate reality. And then there's this, it doesn't even really need a lot of explanation. I have said things along these same lines for about 10 or more years, so I thought why don't I let someone else say it for once? And so this is John Tennant at the recent conference who's basically sending the message that we all know, yeah, scholarly communication is extremely slow at adopting the techniques of the document web, the social web, and the semantic web. That's not like, we've not been told that it could be different. I already referred to Paul Ginsburg, but there were others in the mid-90s already that told us that the then new digital network technologies were able to fundamentally transform scholarly communication. These people understood that with these new technologies, it was not essential to fulfill the core functions of scholarly communication in a vertically integrated manner. They understood that registration, certification, discovery, and archiving could be handled separately and could be under control of different parties. And you see this illustrated in this depiction. This is done by Paul Ginsburg somewhere, I think, yeah, 2001. And there are two things that are really important here. First of all, the decoupling of registration from certification by means of the publication of preprints, obviously. And you see that in this picture at the bottom layer. Actually, Paul refers to it as data, stuff preprints that are posted in both discipline and institutional repositories. And that being decoupled from what he calls here the knowledge layer is where the certification, the peer review is applied, which here is instantiated by means of journals. So that's one thing that we can do with the new technologies. And then the other thing that's important here is that this bottom layer, this data layer, can be made freely available for all without any charge. This vision, however logical and beautiful, it may seem, didn't really generally happen in the last two decades. I still think this is the real way to go because it is the model that provides us with the guarantee that the raw information is available for free of all to use. And there's a guarantee that we can in an unhindered way build added value on top of this raw layer, both commercial and free added value. So while decoupling of registration from certification didn't happen, a couple of other things happened in the past two decades. Open access happened. So I don't consider open access a paradigm shift, but it's really an effort to find new business models to keep doing the same, with a very laudable goal, by the way, of providing free access to information to more people. And as a result of this movement, obviously we have more unhindered access to scholarly papers, a very good thing. There's also other consequences, though, of this movement. I came across this really interesting paper coming out of the UK. And what the authors are trying to do here is to figure out the money stream of research funding, and you see the research funders in the UK at the left-hand side of this picture, flowing through institutions, which you see in the middle of this picture, going towards the publishers which you see at the right-hand side of this picture. And so one of the findings of the authors is that it's actually extremely hard to combine this kind of information to quantify these flows financially. And hence they say, well, if we can't really understand how much money is flowing from research funding to the publishers, how can we even speculate about novel ways to do scholarly communication? The reason I'm showing this, of course, is that there's a new stream of money flowing to publishers. At the bottom of the circle there, you see your typical subscriptions that are being paid to publishers. And on top, as you all know, we have these new article processing charges, a new money stream that flows from research funding to the publishers. Now, if all of these APCs were about truly open access publishing, then all would be good. We would indeed have more open access. But it turns out, and this comes from another paper, that a lot of these APCs actually go to hybrid journals where there's both subscription fees and APCs that are being charged. And this trend is actually on the rise. So we're putting a lot more new money actually into the system. Another thing that happened and that has partly been discussed during the CNI meeting is that the established players in the scholarly communication system have actually increased their grip on the entire system. And you see that in the acquisition of established players, by established players of innovative venues for publishing analytics, the research process, and so on. You also see it in the increased competition, that institutional repositories face from research information systems. This is going on big time in Europe currently. And then the other thing that's really important is that the action radius of these established players has significantly increased from their original function as a publisher of content and way into the whole research enterprise. So I just give you this illustration. This pertains to Elsevier, but this is not at all only about Elsevier. And this shows how the initial business of Elsevier, which you see in the middle there, the publishing business, has now extended and Elsevier has tools to offer throughout the research enterprise. You see that at the left of this picture and also in research evaluation. So that is really as a result of buying a lot of tools, buying analytic capabilities, and buying a lot of content. And then one more thing happened that is really relevant for this presentation. And I have myself talked about this at previous C&I meetings. Increasingly, the research process, not only the outcome of research is visible on the web. There's, as a result of that, a dramatic extension of the scholarly record with a wide variety of artifacts. And as we move more towards open science, this is actually going to happen even more. Now these artifacts are deposited by our researchers in a wide variety of portals, all over the web. And the researchers flock to these kind of portals because they get good exposure there, collaboration facilities, and so on. But the bottom line is that these materials are actually leaving our institutions, the institutions where they were initially created, and they are not being archived because these kind of companies, these kind of portals, the last thing they focus in is on archiving the scholarly record. They offer other kind of functionality. So I hope that with this, I have adequately motivated why I thought and think it's still interesting to explore new possibilities for scholarly communication, new paradigms, new technologies, and so on. And I frame this quest really in terms of trying to achieve the scholarly comments. And with that regard, this little bit in red here that comes from the fourth 11 principles on scholarly communication is really core. How can we achieve a scholarly comments that is sustainable for the long run, fulfills the core functions of scholarly communication, and while doing that really doesn't forget about the archival bit. I will spend the rest of this talk, dwelling on these questions, and I'll consider a technical component where I feel very well at ease. I will formulate ideas that are inspired by me looking into the decentralized web movement. And then I'll also touch upon an organizational component in which I will provide pointers to stuff of others that it is about thinking of organizational principles or foundations for a scholarly comments. All right, so here we go with the technical inspiration. So while I was exploring the decentralized web movement, I came across the work of this person here, Sarvan Kapadizli. He's a PhD student at Bonn University. He's a big fan of Marshall McLuhan, and he's probably equally enthusiastic about trying to change scholarly communication like I was about 20 years ago. So as part of his PhD work, he has implemented this tool, and a scholarly editor really, that is called Dokely. And so Dokely is a client-side application that fits in this whole framework that I explained. It can interact with these pods, okay? And so it gives you editing capability, but it also gives you this social connection, capability by means of the inboxes. It exists either as a browser extension or as JavaScript that is embedded in HTML pages, and it supports all these standards that I've mentioned before that a pod environment typically supports. But in addition to that, it supports a whole range of standards that are pertinent to scholarly communication. So for example, it supports web annotation, it supports Creative Commons licenses, a CITO, which is about semantic citation, PROVO, which is about province expressions, and so on. As I said, it's an editor. It allows one to edit HTML and RDF without actually knowing that you're editing HTML and RDF, so it's all menu-driven. And it's a platform for social interaction. So I'm going to show you, I'm going to walk you through a little scenario, and I'm not going to dwell on the editor capabilities because I don't think those are most crucial for this conversation. I'm going to dwell into the social interaction stuff. So what you see here at the left-hand side is a paper that Sarvan wrote and self-published in his own pod. And then at the right-hand side, you see a little menu system that pops up when you open the Docky Lee extension. You see here, or you don't, but there's a little picture there. This means that Sarvan is currently logged on to his pod environment. So he just put this paper out there and he said, well, you know, I would like a few people to review this. And so he says, oh, I'm going to click the Share button and he clicks the Share button and as a result of that, I see a new menu system and that gets automatically populated with Sarvan's contacts. Those are extracted from his friend-of-a-friend profile. So there's Tim Berners-Lee there, I am there, and Ruben Verborg from Genteser and a few other people. And Sarvan says, okay, Herbert, Ruben, why don't you guys review my paper? And he sends out this notification to our inbox, which of course are dynamically discovered via our profiles. I receive this notification in my inbox and if this looks geeky, well, it's because it is, all of this is, okay? But the message is clear. This message comes from actor Sarvan Kappa-Disney. You see his web ID up there. It's about an object called linked specification reports. So that's the paper he wrote and there's a message there. Would you please want to review this? Of course, I'm willing to review this. And the way I go about this, and this is really interesting, I think, is I'm not going to write a document all by myself, stand off from Sarvan's paper. I'm actually, in this case here, going to Sarvan's website, following the link that he just sent me, and going to create a derivation copy of his paper. So I am logged into my pod environment now, as you can see, because my icon is up there, and I click save as. This pops up a document and I can express where in my own personal environment I want to save Sarvan's paper because I want to express my review on top of that. As a result of that, a copy of the paper was created in my own environment and you see the URL down there. So you're clearly now in my pod environment. What is really interesting here is that I have not only made a copy, I've actually, in doing this, also recorded the provenance. You see there in red that this was derived from a paper in Sarvan's pod and at which day this was done. So now that I have this copy in my own environment, I go about my reviewing business, I edit, I comment, I annotate, okay? And when I'm ready, I'm going to use exactly the same mechanism to inform Sarvan that this thing, my review is now available. So I click share, Sarvan pops up from my friend of a friend list and I send a notification, good paper, but a few changes need to be made. Please publish a new version. And then Sarvan receives the notification. Again, the information is now about my review of his paper and it expresses my web identifier at the bottom there. Maybe you can't read it, but it's expressed as my orchid identifier and the message is indeed sent to Sarvan. So I find this remarkably interesting, a completely new way really of doing things, all in function of self-publishing your papers, self-publishing your reviews in your personal environments. And all of this, again, a document network overlaid with this social network that can be discipline-based. There's quite a few significant, actually, technical challenges that really need to be tackled to move this idea like Dokely to something that could be production ready. That's in the realm of authentication amongst other and we recently worked with Sarvan, my team worked with him, on integrating authentication with Orchid, actually, using the Orchid OpenID Connect. So basically, one can use this environment, a totally decentralized publication approach with a central piece of infrastructure that this community provided, like Orchid. It's also problems with versioning because LinkData Platform itself doesn't support that. But I don't want to dwell on that because there's really interesting conceptual questions, really, that when you think about the self-publishing model, how are you going to fulfill the core functions of scholarly communication? Registration, certification, awareness, and archiving. And I can't touch up on all of this, but again, I will show you what I mean, what the challenges are or illustrate by means of a scenario. So here's Alice again. And she's working on a new paper and she creates working draft one, working draft two, and then she decides, yep, this is ready now, I'm going to self-publish it. In terms of the scholarly record, this really means that she has registered a new paper, a new finding into the scholarly record. There's two really interesting aspects just in this one picture already. First of all, clearly we need the semantic distinction between those working drafts and the paper that Alice considers published so that we as readers or machines would understand Alice actually considers this a published paper. The second thing is, since the registration is concerned, we actually want to know that past registration, Alice doesn't tamper with this information. And this is of course where a self-publishing model differs significantly from submitting to pre-print archive or to publisher because in those cases, the paper is out of your hands, you can't really tamper with it anymore and the date stamp of priority has been given. In this case here, because Alice is completely in control of her own information, of her own pod, she can tamper as much as she wants with this post-publication. So a way to ameliorate this problem, and this is just speculation, is to actually combine registration with archiving. Archiving has to be done anyhow, remember? So here's the idea, let's just imagine there is such a thing like a scholarly archive. And so at the moment that Alice decides she's actually going to register her new paper in the scholarly record, a notification is sent to that scholarly archive, obviously with the URL of that publication, and now the archive is in business to pull that in and archive the paper. So we now have gained two things if we proceed this way. We have the ability to verify that Alice hasn't tampered with her paper post-publication by comparing the live and the archived copy. And what we also have is that if Alice decides to leave academia, she can actually, and we can actually find a copy in the archive. Now same thing actually applies when one publishes a review, like in this case here, Bob and Carol publish reviews to Alice's paper. This also really considers new registrations in the scholarly record, and as a result, the same mechanism could be used to archive these materials. There's another piece of work that has caught my eye as I was exploring decentralized web work. And this is the work by Amy Guy. She recently received her PhD from the University of Edinburgh Informatics Department. And if you don't know what you're looking at here, well, I didn't know either initially. But it turns out that this is actually her website, and that each of these little squares that you see are actually activities that Amy conducted as she went by her life, okay? And so for example, when I clicked that one little square there, which has a little money bucket on it, I actually get to see one of these personal activities. And we can see here that Amy bought some food, vegan food I need to add, at Luton Airport, and she actually spent five and a half British pound for that. This is pretty weird, but it's actually a thing. This is actually called a personal web observatory. And this is all about tracking your daily activities by using a wide variety of web portals out there. So Amy has accounts in a wide variety of these things that all track certain aspects of her life, for example, buying food, traveling, walking, biking, and so on and so on, and they're all different portals. And so her website, the system that she runs, which is called Slav, actually interconnects with these systems, uses API approaches, a variety of techniques to pull information from all these portals into her own environment. And as a result, she builds some kind of an event store that tracks the activities of her life. And so the website is a depiction of that event store, and the event store, of course, can also be queried. I was new to this, I hadn't heard about this, but it turns out this is actually really a thing, okay? And there's a whole movement called the Quantified Self Movement, and they engage in these kind of things. For example, for the purpose of starting to lead a healthier life. So they bring all that information into their central environment, their own environment, they run analytics on top of it, and then they say, oh, if I do this, this, and that, then the outcome will be a healthier me. There's also the Locker Project, and that was all about assets, artifacts, that people are dropping all over the web because they like the way these portals behave and all, but in the end, they also want it under their own control. And so they use all these APIs, connectors, they call it to bring the information in. If you have been listening carefully, you should probably by now know why I was so interested in this. Remember these artifacts that our scholars leave all over the web in all these productivity portals, these things that are leaving our institutions and that are not archived? Well, Amy's work actually provides inspiration for something that one could call a personal scholarly web observatory where one actually traces the activities, the artifacts that scholars drop, individuals drop all over the web. And where information about these new artifacts is automatically collected in a personal environment of that scholar. In addition to that, not only information about the artifacts themselves could be collected, also information about the interactions that peers have with these artifacts could be collected. And I'm sure that an awful lot of really neat analytics could be run on top of that. But that's not my major motivation to look into that. My major motivation really is to get hold of that artifact you arrive so that somehow we can collect and archive the artifact. So my team has started to rudimentary explore this notion of a personal scholarly web observatory and they used me as my subject. And so what you see here, well, what we did is we tracked my artifacts in the real scholarly record by means of my DBLP profile, also my artifacts in SlideShare and in GitHub. And what you see at the left-hand side is that I have way more papers than I have presentations and I have very little activity in GitHub actually. And then on the right-hand side, you see interactions by peers with the artifacts I put out there. And here we use, in order to collect that, we use the SlideShare API, the GitHub API. And then when it comes to my publications, we use MS Academic for citations and we use the Crossref event data for references in non-scholarly environments. So I kinda like what I see here because this gives an integrated view of my activities, not only the things that are traditionally considered scholarly, but the things that I really consider scholarly myself. So that I find interesting. It also is interesting because we are tracing stuff here not because they have a DOI like we do with altmetrics and with citations, but because I created them. It's on the basis of I have an identity in all of these portals, that's why these things are now being traced. And then the third thing that this shows, I mean you see that the amount of interactions with my slides are so much more humongous than with my papers, that basically means I shouldn't write papers anymore. This is another visualization, it's a heat map. It's done on the top of the same kind of event database that we collected about myself. It shows both creation of new artifacts and others interacting with those artifacts. And then the big white spot that you see there is because a lot of these APIs don't allow you to go a long time back in time, but going forward you could obviously fill this all out. So this is where I leave the technical domain and I veer into a domain that I have no clue about really, which is organizational matters, that's your forte actually. What I have observed though is that there is a very strong sentiment living currently in many quarters that academic institutions should take back control of scholarly publishing and of the scholarly record. And one of the recent indications I ve seen of that is this posting by Shan Sutton, which really is a call to arms. It's actually co-authored with a group of very prominent library deans and directors. And it's a call to arms regarding the establishment of an academy-owned scholarly communication system where access to publications is for free and eventually the act of publishing itself is also for free. So the idea here is that academic institutions, libraries, will actually provide and sustain the infrastructure for a future scholarly communication system, something like the scholarly comments. These ideas are also expressed in this short paper by David Lewis and he gets actually concrete and he says all academic libraries should spend two and a half percent of their budget on deploying and sustaining such infrastructure. And then there's Bjorn Brams, by many considered an extremist voice when it comes to thinking about future scholarly communication systems. In my opinion, a man with quite a few good insights. And for years, he has been shouting off the rooftops that it is academic institutions that should provide the infrastructure for scholarly communication. He actually compares providing or investing in that infrastructure with the investments academic institutions made decades ago in network infrastructure. And he says when these institutions made these investments, they really didn't know what the outcome was going to be either, but they did it and it turned out to be really great that they did. Bjorn goes actually one step further because he says that we have a one to two year window now where most of the papers are available to anyone to grab thanks to Sci Hub. And he says we just should all cancel our subscriptions and use the money that becomes available and invest that in initial infrastructure for scholarly communication. That's probably where the extremism comes in. Shanssettens posting explicitly talks about two parts for the future, two kind of approaches to achieve the scholarly comments. One is the inherently discipline oriented preprint servers with a reference to decentralized OSF preprint platform, preprints on the rise, very good news. And the other is institution based repositories with a reference to the next generation repository report by core, I actually contributed myself to this report. This report has a very strong emphasis on interoperability because the vision that is being brought forward is that of institutional repositories as many nodes in the network but together forming the basis of a new scholarly communication system and one can only achieve that when there's deep interoperability between these systems. So what I've told you thus far really raises the question of could there be a third track? Do we need to keep open minds? Could there be such a thing as an other publishing model based on these ideas of self publishing? So I'm going to call it from out the researcher pod and I'm going to explore this idea and this is where I veer into total speculation, okay? So you see here the researcher pod and it consists of two main components. At the right hand side, the communication platform and that's the thing that's inspired by MIT solid and Dochili and then on the left hand side, this notion of an artifact tracker which is inspired by Amy Gee's work. This tracker though doesn't only track the artifacts that researchers leave out there in portals on the web but also the things that authors are self publishing in their pod environments. So all of that goes into this event store. Now obviously we cannot expect a researcher to operate their own pods. That's where you guys come in. So we're going to say that these pods are actually hosted by an institution. So every institution is going to host the scholarly pods of their respective researchers. I volunteered you for that. I'm going to volunteer you for one more thing actually. Sorry about that. I said I was going to speculate. The other thing is of course archiving and I'm going to put that in the institution's lab also. I think there's a nice alignment with the real task of libraries of academic institutions and archiving the scholarly records. So what we're going to do is as this event store records information about new artifacts, we're going to send that information about these artifacts onto an archival process that is operated by institutions and that archival process is going to collect one artifact after another and it's going to dump it in the archive. So let's pull all of this together. First from the perspective of the researcher. So a researcher here in this paradigm has a personal domain for scholarly communication. One could actually say the researcher starts his life or her life as a researcher by acquiring a personal domain. It's not institution bound. It's truly personal and I'll come to why later on. The researcher pod is for scholarly activities and it operates in the domain. So this is truly the researcher taking responsibility for the publication of papers and reviews in their own environment. David Rosenthal and I had a conversation about this on the 411 list and he says, in this kind of a paradigm, you achieve a level of persistence purely because of self-interest of the scholar. So our pod has three functions, self-publishing of papers, overlay reviews, comments and annotations, notifications, which is where the discipline component comes in, okay, and then the self-tracking, keeping track of everything that the researcher is doing out there on the web by use of APIs, notification mechanisms and so on. So this researcher pod, definitely this event store, eventually starts to provide a very comprehensive overview of all the contributions that this researcher makes and potentially of the interactions that the researchers peer has with these artifacts. The pods, because they support all these standards are globally interoperable and the standards actually to achieve this exist. The tools are currently really brittle and that's a massive understatement based on my experience. And then there's the notion of institution enabled. So the institution provides a hosting platform and of course operational support so that the researchers can run their pods in those platforms. And the researchers domain resolves to the institutional platform. And then the institutions also take care of the archival aspect of all via this mechanism that I've explained. One could discuss what the exact nature or architecture of this archive should be but there's several options available. Now one more exploration of this idea and this pertains to the mobility of researchers and the persistence of the Scholar record. So here they are again. Alice and Carol work at institution A and Bob works at institution B and obviously in this paradigm they all operate their own pods. And again in this paradigm whenever they create a new artifact either out there on the web or in their pod it automatically gets archived into this scholarly archive. Now at one point Alice decides to move over to institution B. Obviously the stuff that was already archived while she was at institution A remains in the archive. And then Alice what does she do? Well she just grabs her domain and she grabs her pod and she moves it along. It's portable. She moves it along to the new institution and she drops it there on the hosting platform and life is good again. She can just start self-publishing again. Obviously everything that she now creates from within institution B gets archived again in the scholarly archive. It's just that now it's going to happen via processes at institution B. Now Carol decides to completely leave academia. She's had enough of it. And then well a couple of things can happen. Her employer or Carol herself can decide to keep this pod running. Maybe not make any new contributions because the research career is over but whichever way the materials that she created are in the scholarly archive. Or she could completely give up on this pod and start the fundamentally new life but again her materials are sitting in the archive. And basically the same story if Bob goes belly up because his institution could say, well you know Bob is so important that we're going to keep his pod alive. Or as again David Rosenthal suggested, well Bob could actually have a will where he endowed his pod and his domain to keep on going for a while. Or again of course the pod could just totally disappear but again Bob's materials, his artifacts are in the scholarly archive. Now persistence is not only about having archive the stuff it's also about seamlessly being able to get two materials that sit in the archive. Remember we've created the document network and Alice's papers may point to Bob's artifacts and we want these things to keep working when they leave their pods behind basically. So when Alice moves from institution A to B there's not a problem because she holds on to her domain so all the links just keep working. When Carol and Bob give up on their pod we have developed technologies that actually allow us to follow these links that existed on the live web or that exist on the live web that would typically lead to four of fours that now would directly and seamlessly lead into the web archive, the scholarly archive. And these technologies include PIDs and robust links that comes out of my team related to the memento work. So in summary one can think of these researcher pods as a different kind of institutional repository. It's contributor centric instead of document centric. So it offers something truly for the researcher. The disciplinary component that is missing really from institutional repositories is instantiated here by means of social network functionality overlaid on these self-publishing environments. There is, as I showed you, built in portability, built in mobility and the thing is archive ready if we have these event stores that communicate with an archival layer. And then in closing, it was really not my intention here to convince you that researcher pods are the way to go. I just wanted to illustrate that there are degrees of freedom beyond the notion of preprint repositories and institutional repositories that we might want to explore when deciding on creating the scholarly comments. Self-publishing as a concept may be too far removed from the current practice although one can really trace it back to the ideas of Stephen Harnett in his subversive proposal and even to the early days of the Republic of Letters. While it may not be really feasible, I think discussing it, contemplating it still remains extremely interesting because by understanding why it is not feasible, we may get new insights into what is actually feasible and that could be pretty enlightening. From a technical perspective, for the very first time in the two decades that I've worked on these kind of things, I feel confident that the standards, maybe not necessarily the tools but at least the standards to deploy the scholarly comments really exist. And that's why I also voted 10 out of 10 on that very first question here that was, you know, it's a poll by Bianca Kramer that was issued at the recent force 11 conference. Can the scholarly comments be realized with existing technology? My answer was 10 out of 10, yes. Technically we can do this. So that leaves us with the organizational questions. Can academic institutions take global collective action to work towards funding, building, sustaining the scholarly comments? I'm afraid that I do not have the expertise to answer that question but many in this room actually do. And so I hope that with this talk I've given you a little bit of inspiration and I've given you a little nudge to taking concrete action with this regard. Thank you much. Cliff, do we have time for a few questions if there's any? One question? If there are any. I'm tired so you don't have to feel obliged. No, I don't see any questions I think. Okay. I think you have stretched people's thinking in so many fascinating directions and I have to say there is so much work you alluded to there that I'm unfamiliar with and wanna go explore and I suspect I speak for many people here. I think you've given us an amazing amount of things to think about and I really love where you ended this which is this is a set of things to think about not a blueprint it points to possibilities that perhaps haven't been adequately considered. Exactly. And those kinds of opening ups of possibilities are so valuable to us. I think you'll agree with me that the nominating committee for the Paul Evan Peters Award have picked an incredibly worthy awardee. I'd like to thank that committee. Their members are enumerated on the award brochure you have and I won't take your time here but especially Herbert, I'd like to thank you not just for this wonderful talk which is such, Paul would have loved this, seriously. But for everything you've done for our community thank you so much. With that I wish you safe travels. I wish you good holidays and a good new year and I hope to see lots of you in San Diego in the spring. In between we will get all of these presentation slides and videos up as soon as we can and we'll keep you posted on our progress on CNI Nouns. Travel safe. Thank you.