Call for Linked Research
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 16 | |
| Author | ||
| License | CC Attribution 3.0 Unported: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor. | |
| Identifiers | 10.5446/47542 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
|
3
8
10
00:00
Lecture/ConferenceComputer animation
00:23
Computer animation
00:57
Computer animation
01:24
Computer animation
01:50
Meeting/InterviewComputer animation
02:12
Computer animation
03:38
Meeting/InterviewComputer animation
04:40
Computer animation
05:22
Computer animation
06:43
Computer animation
07:10
Computer animation
10:36
Computer animation
11:03
Computer animation
11:37
Computer animation
14:03
Computer animation
15:09
Computer animation
22:16
Computer animation
23:23
Computer animation
24:23
Computer animation
25:12
Computer animation
27:47
Computer animation
28:50
Lecture/Conference
Transcript: English(auto-generated)
00:05
Thank you. Thank you. So the URL of the presentation is up there and there's a
00:25
short article about that I wrote a while ago. The presentation as well as article and everything that I'm about to talk about in the papers and so on are in HTML and RDFH so you can dereference this slide in question and if you have
00:45
something to say about that slide you can make it triple. So thank you for inviting me and let's get to it. I will let you consume this. All media are extensions of human, of some human faculty, psychic or physical, said by
01:06
Marshall McLuhan. So I'm going to start the talk from up here and then dive down and I think the philosophy of media and how it impacts the society is important. So to explain that I'm going to let you just consume these images.
02:04
Go back, read that again. So what he was talking about is basically the way we used media and our senses basically shaped our societies. We were in villages and telling
02:22
stories to each other. That's how we transfer knowledge. We use our senses to see and to hear and the body language to consume this information so it was an oral culture, right? And we went into print and became a
02:40
visual, primarily visual culture, right? Everything is written down, read in a sequential manner, represented in a two-dimensional space. It's internalized, it's somewhat antisocial and the explosion of the printing press. We have
03:03
the telephone, right? We're using our different senses again to hear, right? So while I'm talking on the phone, my other senses are not occupied, right? So I can actually look at something else as I'm listening to someone on the phone, whereas if I'm looking at a printed paper, my vision and my focus is
03:24
primarily on that paper. I can't really, you know, carry out a conversation or listen to someone give a speech while reading a paper. Maybe some people can but I can't. I think most people can't either. So senses pretty much tells us
03:40
how we consume this and what it really means in the end to actually have access to this information. So there's a radio. It's a bit social, right? People gather around. TV looks pretty social to me. They watch, you know, they can have their feedback, annotate what they're seeing in the living room. Hypertext,
04:01
the 60s, right? So it's interactive. Now you're using your hands and then, you know, who knows where the future is going but I'll just stop there. I mean, I don't, we don't need to go cover all types of media here but the point is that the media that we bet on or how we want to consume and transfer
04:22
knowledge, it entirely shapes the societies, right? So we went from oral culture to a written culture and then we have a interactive culture like the web, right? So people can talk on the web and, you know, tweet each other and so on.
04:41
So this talk is going to be about how you should, why you should control the link or academic research yourself and you should use a native web stack to do it. I don't need to convince you that, or anyone here or listening, that the web works. It's been working for fairly 26 years. It may fail but there's
05:07
no sign of it. So this electronic way of this medium that we're using to communicate and to interact and to share knowledge, it's been working fairly well. So then the question is how cool is scholarly communication, right? So
05:27
around 15th century, printing press came out around 1989 and 90s web, right? So what changed in how we share knowledge or research knowledge in this particular case?
05:45
What we're doing is, again, we're strictly using our visual system. We're reading each character, each word, one after another. It's non-interactive other than the fact that you may print a paper and, you know, annotate it and scribble on the side of it. It's somewhat antisocial because you're by
06:03
yourself, you're focused on what you're reading. Of course, you know, you may interrupt yourself and, you know, talk to someone else. And in the end what we're doing is we're just imitating paper, right? So we have this two dimensional space that we came up with to represent stuff which, you know,
06:24
anywhere from having scientific information to religious documents with typographical rules and so on and to today, right? And we're still doing the same thing. We're pretty much betting on that two dimensional system of way of communicating. But the web is about its interconnected ideas, right?
06:48
We're intermingling on the web. We're sharing not just the documents but our opinions about these things. So the web is, you know, brings back that social connectivity that we had a long time ago. It's like it
07:06
is a global village as also Marshall McLuhan says. So the problem as I see with the scholarly communication is that, well, there's many problems. I think you're quite aware of that as well. One of them is a user experience problem. So
07:20
when we were looking at a PDF or Word document and so on, these are essentially paper user interfaces, right? Two dimensions. There may be some hyperlinks in there but it's, you know, when you submit a paper to a conference, what are the submission requirements? It says something like, well it has to be in 12 pages and this font size in English and, you know, make
07:47
sure you make sure you follow the Springer's layout rules, right? And, you know, given the age that we're, you know, the web, right? We're betting on that still. And that's producing, of course, data silos. These papers are
08:03
authoritative in a way. Documents are ending up on Springer or Elsevier or any, you know, name and publisher, right? We're handing out these documents to reside on their websites which we essentially don't have any control over. You don't own the domain, essentially. And they don't play well
08:22
with the web. They're just binary documents, PDFs. You can't get the source back. It's not that accessible. Have you tried searching for text in, let's say, Google Scholar for a particular fragment in a paper as
08:41
opposed to just a web document, right? When you look at the search results, you get a lot more out of it when you're looking at HTML document and search results as opposed to a PDF document, right? How many papers are out there which declare the hypotheses of the paper or how would you even extract
09:02
that information, right? Or what's the workflow step which resulted in where you get the data set from, right? None of that information is really easily accessible. It's just what you do is a big blob of text. You put it into a binary file, lock it away, give it to a third party, and then, you know, here's
09:24
your degree. So we have a discovery and a reuse problem. We can't query for these atomic parts because I want to know what your hypotheses is or what your methods were so that I can say something about it from my paper, right? I want to say I agree with that particular statement you made in that
09:42
sentence, in that paragraph, right? Not, you know, a document here and a citation at the bottom of the paper saying something about that, like which is arbitrary relation. It's not as specific as it could be. So if we can,
10:01
you know, discover, identify these parts which also means that we can all do other things around it, right? We can figure out the research gaps, right? So what are the trends of the research in that particular year? What are we discovering, right? Or find out new areas that we should be researching on. Or even further, what about the funding opportunities? How
10:25
should I assemble a team? Who should be on my team? If you link the profiles with authors and the papers, you can get to get to a lot. So I think that we have a paradigm shift that's needed and that's unfortunately been
10:43
lagging about 25, 50 years, depends how you look at it, 25 or so years. Web, 50 years maybe for the hypertext and the internet and so on. We're that behind. We have the technology but we haven't really adapted to that. So I propose
11:04
that we fix the user experience so that we can have better accessibility and also, you know, discover this information and move this information body of knowledge through different devices or through different senses in the end,
11:22
right? So it's not just printed material that you're looking at in your hand but or on the screen but you can move it to your device or hear it, watch it, whichever, whichever, however is suitable. So to do that I think that we need to start with an acid test and that's basically just a fancy way of
11:44
saying we need to figure out the use cases and, you know, how do we represent these scholarly documents, academic papers of course, that is if we if you buy into the idea that okay I want to be able to discover somebody's
12:01
hypotheses or arguments and conclusions and so on or the figures even, how do we do that, right? So we need if we agree as that one of the problems then we can sort of go after it. The technology, so of course I'm going to you probably have a good sense of what I'm going to propose but the technology
12:23
is not the important part, it's just that we agree on the problem and then if you have a different way of solving that, that's great. If your PDF can do that, if your bitmap image can do that, great, go for it. But there are other perhaps suitable technologies in place which seems to be prone to work and maybe
12:45
they're better bets to go with. So I won't go too much in this, I'll just fly through them. So there's a call for linked research and the idea is that you publish your work now. You don't need to wait for someone to give you a go
13:02
ahead. You can publish it on your own website or on your institution's website, at your library, your company, whatever gives you that sense of trust and control over your own research document so you don't have to give it to a third-party company, for instance. I don't mean that you know you
13:23
should publish on your friend's website but that the sense of trust is up to you. So whether that document will be around, how it will be accessed, these are the questions that you read, collectively probably have to figure out, but instead of just going along with, you know, status quo. So if you have
13:41
control the URI of your own article, you publish it using the Webby technologies, then you can do the rest of the stuff which, you know, helps with knowledge acquisition, better experience, you interlink it with the other documents, other arguments, and so on. And of course you want to motivate
14:01
others to do the same. So self-publish, self-archive, edit web space, you control and trust. That's one of the main messages. And while we're doing that, we're of course trying to identify the interesting parts of our documents, right? So if I think that my hypothesis is worthy of attention or someone to link
14:25
with, then I will go ahead and, you know, mark that part or declare that that is my hypothesis right there. So if you have something to say about my paper, don't just say I disagree with your paper, say that you disagree with that atomic part in that paper, right? Maybe you agree with, you know,
14:42
majority of the paragraphs, but there's a paragraph that you strongly disagree with. You can say I disagree with that. You know this, I will skip this, skip this. These are some of the slides for anyone that's not here. So here's an example,
15:04
which we can, so these slides are also interactive. I have it on a web page maybe. But if you look closely to that graph, so this slide, it says Sparkline by, usually proposed by Edward Tufte. It's actually loading from a live endpoint,
15:21
so some bunch of statistical data from a statistical endpoint, and it's an SVG which just loads it back into the page. So I have a paper that actually does the exact same thing. I just put that on the slide as well. So there's no difference between the slides and the papers. It's all HTML. Or here's another part of a paper or whatever. I can
15:41
interact with it. I can type some concept and type through. So I'm doing this directly in my slides, but it could be on a web page. So you announce it, you create awareness, of course, get some feedback. So one of the tools, of course, there are different approaches. So
16:01
I'm only proposing one, so one approach to do this, and this is the one I'm working on. I'm not trying to sell you this. I'm trying to sell you the acid test, you know, the call for research, right? That's the initiative. Those are the principles that we should go after. This is just one way of doing it. If that's useful to you, use it,
16:22
or use another tool that you prefer. So it's meant to be completely decentralized in a way that you as an author can decide where you put it using your own Web ID and your own personal storage space, and you can give access to whoever you want. So just on that
16:45
remark, if someone wants to come and annotate your paper, that annotation or the comment or reply or any sort of social interaction with that document, that information doesn't have to be embedded or necessarily live in that document. It can exist somewhere else, right? So your
17:05
annotation belongs to you. You are the author of that, and you decide where you want to store it, and it can be anywhere on the Web, right? Anywhere you trust or have control over. So Dokily, it's derived from Okidoki and Linked. We'll see whether that's a good name
17:24
or not. I don't know. It took me a while to come up with it, but I'm kind of stuck with it. So it's friends with the Linked Data Platform and Solid. Solid is a social Linked Data Platform which is friends with Linked Data Platform. It's just a thin layer which lets
17:41
you do, it's extends Linked Data Platform to do more social things. It's also friends with the Web annotations, work groups output, and the social Web work groups output primarily on the data model. Basically, the vocab that they have, it's used in Dokily. So when you make an annotation, you know, not only is it stored somewhere else, but, you know,
18:04
that RDF within it is using those vocabularies. So I'm happy to demo that better if you like. So I'll switch over to a little bit other things around this. The CUR workshop proceedings, which is the largest open access publishing computer science, will support Dokily. We
18:25
have a CUR make which is, it's on GitHub. So the green things are links. You can get to them on the presentations, on the presentation slides. You can basically export, for now, you can export the data that you get from EasyChair, run it through CUR make, and
18:44
that gives you, generates an HTML RDFA document. So now we have proceedings in RDF, right? So imagine all the CUR, from now on, all the CUR proceedings which were generated will have that. So now we have, you know, this all exists. It's not hypothetical. It works,
19:05
okay? So all my papers are published in that way, and CUR is taking this on as well. So we have now, we can now have a body of papers, right? All these papers with all their atomic parts linked, interlinked, discoverable, with the proceedings, right? So this paper
19:25
appeared in the proceedings. The proceedings says that that was, you know, that includes or has a part of that paper, right? The paper also says that this was a reply to a call for paper, right? So a call for paper is a particular theme of that year, right? So all the calls
19:45
linked with all the papers and proceedings, and this whole concept of pay as you go works. This has not been announced yet, but in the next weeks, the LinkedIn on the web workshop
20:02
will make an announcement that HTML and friends are welcome if you want to use your own tooling or Docly or whatever. That's fine. As long as, you know, you kind of keep things consistent with the rest of the submissions, you know, we may, as an interim solution, we'll go with
20:21
that ACM layout, which is fine, because what's important is that, you know, I can actually dereference your paper, right? So, like I said, I'm happy to get into more details on the tooling and link research in the breakout session. I hope you join. If you
20:41
have ideas or tooling that you want to suggest, that's perfect for it. So now a little bit about some fun stuff. So I published, one of my papers ended up, unfortunately, at Springer because I wanted to get my PhD degree at some point, so I had to comply with the rules of the conference and so on. So, you know, I signed the copyright
21:03
form and I, you know, gave the LATAC to them. I included the URL to the paper, which is published on my site, and somehow ended up on Springer. So if you go to the Springer's Springer and get to this paper, you can decide what you want to do. If you're not, if you
21:24
don't have access to it from a library or whatever, right? And somewhat recent news, I'm sorry I'm picking up Springer, but it applies to all publishers. So a colleague
21:40
tweeted this. It seems, you know, it's good news that their meta data, I emphasize that, all the proceedings is now available as linked data. Great. But I still can't find out anything about your research, what the hypothesis is. Great that, you know, the authors or the keywords
22:03
or the abstract was available. I mean, I can get that from a website as well. You just have to, you know, do some cleaning, but I think we can do more. That's the whole point. So, I will leave it here. We can discuss. Questions are welcome. I don't know if you have more
22:25
time. I can even demo. We have time. So I'll just go a little bit more than planned, but yeah. So I just showed parts of Doakly. So it looks something like this, right? So
22:41
this is just a test page, right? It looks like a normal paper. I can turn that into ACM paper. It's just a CSS change. There's nothing fancy here. This problem was solved 10, 15 years ago. You know, I can link to the introduction. So each of these paragraphs or anything like I said, worthy of attention, you as author can decide, okay, I want that,
23:04
you know, fragment to be available on the web. So underneath it, there's RDFa and so on, right? I can, this menu needs an update, but I mean, you can do these things. You can embed some turtle or whatever. You can edit. So now this works like in browser
23:24
authoring, right? You can do the normal formatting stuff like you're used to. Now I can also, you know, okay, that screwed up. So let's do that. Anyway, that annotation
23:59
is not, it's stored in a different location, right? I'm the author because I signed in
24:05
with my web ID to this document and I have control, I had the right access to it. Everybody else can get a read access. So this document, the paper itself, you know, has a read and it acts as a view layer for that annotation, but you can access the paper by itself.
24:27
This is the paper, sorry, this is the annotation itself. Nothing fancy there to look at. There's another paper of mine. It's on our website, on my website. It's just like a blog post,
24:42
right? Nothing fancy. We have comments, right? It's track link or the argument link. So it's LNS, CS, right? So it's a Springer's layout. If I want to, you know, if I really
25:04
want the PDF for it, I can just print PDF. There's a print style sheet for it, right? Make it ACM. We can, I mean, generate hundreds of layouts if that's what we're after, but the goal is not, you know, that we have this fixed view of the academic work.
25:27
We may have a preferred view for it because maybe that's a better way of communicating it, but on the web, it doesn't need to be fixed, right? You want to move these documents around into different locations and different devices and so on and used by different people at
25:43
different times, not just today, but maybe in, you know, 50 years from now and so on. You can pretend that this paper is a W3C recommendation, right? You can pretend that
26:00
Tim Berners-Lee wrote this paper or, I mean, there's interactive parts of it. Like I said, this graphic, which I showed you earlier, it's also part of the paper. You can interact with
26:22
it. You can run queries, right, on a paper and it's just a sparkle query editor, nothing fancy, but it works, right? So, by enabling these things into these web documents, we let the reader sort of learn better, right? You want to, that's the goal. You want to get, you want to get a credit for your, what you wrote, but you also want to help the reader
26:44
understand it. So, if you have an algorithm, you don't just dump in the code there or the pseudocode there or whatever, the algorithm, but you can let the reader play with it, because then you can learn better. You can interact with it, you can change, you can execute the paper to see whether it's valid, right? Whether all these workflow steps are
27:05
right and whether you can reproduce the same results that paper was talking about, or if you change the values, what happens then, right? Can you break the paper? And there's all sorts of, like, this is all breakout stuff, but like, you know, we can talk about the
27:21
provenance, persistence of these UREs and so on. I'll just stop there, because I think, like, this could be 40 minutes or, you know, five-hour talk. And I think we were here.
27:43
No. No. We were here. Here. So, yeah, questions and comments. Tell me what you don't like and what your problems are, and why, what will it take for you to publish your own work under
28:01
web space that you control, instead of just handing out a LaTeX or a PDF or a Word or whatever to the conference and so on? Because I know that most of you have some sort of web space under the institutions that you're at, right? Your university probably gives you a web space that says here's, you know, you can, or you can ask IT if you don't know it,
28:20
but you can get a hold of it. So you're able to publish it. It takes 30 seconds to create a Wordpress account, 30 seconds to write a web document, so it's not that it's not feasible. The, like I said, it's, the HTML is proven to work, is yet to fail, so,
28:41
and there was a good reason as to why Tim Burnsley picked HTML or decided on something like HTML where you can view the source and have your document written in a declarative fashion as opposed to having something like LaTeX or TeX where you have to compile it and maybe break,
29:03
right? Instead of coding it, programming it. So, happy to hear your thoughts.
29:22
Thank you very much, Stavin.