Linked Research

CC BY 4.0
Appeared In
CEUR (Central Europe workshop proceedings): Proceedings of the 3rd Workshop on Semantic Publishing, Volume 994, urn:nbn:de:0074-994-6

Information control is just an idea and should not to be mistaken as something new. It is an idea that's like a splinter in the minds of some, driving them mad.

The forces that wish to make a change and those that are resisting are tangled with one another. Thus, as the position of a polemic goes, I must explain to you how all this idea of denouncing change and praising obedience was born and I will give you a complete account of the system, and expound the bullshit, the master-builder of human laziness.

This is not an attempt to convince you with the same arguments that you have heard countless times about the wonderful world of free and open access to knowledge. It is not a roadmap on how to get there by ticking all the latest check-boxes. Such endeavours, while fruitful, are still doomed to endless struggle of changing control and power.

At the heart of our challenge are not technical or social problems. We tend to get sucked into the uncertainties and contravening evidence in practically any given argument. The details, so to speak, essentially slow down the rate of progress. Nonetheless, we do it in hopes of something different.

To take a trivial example, which of us ever undertakes laborious exercise to acquire and share research knowledge, except to obtain some pluses in one form or another, all in the name of making our contributions to humanity? But who has the right to criticize someone who chooses to enjoy the pleasures of mental masturbation if there are no negative consequences, or someone who remains silent, and changes nothing?

Do we then have the right to criticize those that stand in the way of others? How about the ones that wish to silence others because it may bring them potential future discomfort?

I feel I owe you an apology. My primary agenda is to ask for your consideration to make a change in whichever way you feel about the matter of publishing and sharing knowledge instead of sitting on the sidelines and watching things unfold and accepting whatever is handed out to you. Your voice matters.

This brings me to echo an idea that's similar, if not the same as one with which you are already familiar. The idea is for Web researchers to eat their own dog-food when it comes to sharing knowledge. Use the technologies and tools that are native to the Web. Continue to push the boundaries on how we acquire and disseminate knowledge using the Web stack. We have to intentionally tap into one another's knowledge by accessing, discovering, re-mixing, and sharing to the greatest extent possible. Some might say this would be employing public-funding to its fullest potential, while others would say that they are already doing this.

A proposal to shift towards Linked Research:

Start with HTML and enrich it semantically; encapsulate everything from meta-data, references, claims, conclusions, to steps needed to reproduce the research itself. Present it with a CSS that's helpful on screen and when printed. Use JavaScript to offer interactivity to better communicate your findings and allow further experimentation. When appropriate, use other technologies from that bag labeled "Web friendly". Publish the work at a URL so that both humans and machines can access and discover. Offer a comment system so that your colleagues can publicly provide their review or feedback to improve your work and so that it is immediately accessible. Announce your research so that you get the karma points that you seek all under your authority without having to ask anyone for permission to spread your contributions. On this day, you are both the author and the publisher.

Alternatively, we can simply continue on with the existing practices. We can continue to be at the mercy of "people in power" or needs of businesses because we are overly comfortable with an archaic pipeline that's constructed by those in charge. If rated using TimBL's Linked Open Data stars, it will get us 3 stars tops on a sunny day. Who cares about the stars right? That's the same star-system with which we try to slap government stakeholders or all other data-huggers. There is too much hypocrisy and arrogance in how we use our own technology, and it might be related to laziness and obedience.


13 interactions

Sarven Capadisli replied on

Perhaps there is some sarcasm in your statement, but I'll bite and respond any way.The point is to take charge and use the technologies that the SW/LD community is pushing. To eat one's own dogfood.To stop the hypocrisy on expecting others to aim for 5 stars but being satisfied with a much lower bar when it comes to sharing publicly funded knowledge.I think we can do better and should aim high instead of bending over and accepting what some groups with personal / private interest have in this matter.

Sarven Capadisli replied on

I'm not held back personally as far as sharing my small contributions.However, the community at large is held back from making progress by restricting those that want to publish using the stack of technologies that's native to the Web. As I'm doing my PhD, I am technically in that group as I'm forced to provide my work in PDF by some conference organizers.One of the reasons for the restriction is because it brings discomfort for a select few (chairs, reviewers) as it would mean that they would have to deal with multiple formats i.e., PDF, HTML (and friends). Interesting priorities there.Another may be because there are business arrangements which is currently contrary to their profit models.I'm merely encouraging what many others before me and still do. The article came to  be for two reasons. One was as a summary to some of these discussions:[1][2][3][4][5] the other for ESWC SePublica 2013:[6][7] are well-aware of my position on this stuff any way ;)

Luca Matteis replied on

Don't blogs currently work this way? Most default Wordpress installations offer pretty good <meta> tags for things such as tags and other information.And this is pretty much the state of the entire blogosphere. It's not too bad.But I see your point regarding scientific papers. Semantically unfriendly PDFs with static data.I wonder if a switch can be made towards Web friendly technology rather than static PDF files.

Sarven Capadisli replied on

A complete switch would be great, but I initially didn't even suggest that. I've reached out to conference organizers on the public-lod/semantic-web mailing lists to simply allow alternative formats alongside PDF.The fact that they are opposed to allowing Semantic Web researchers to share their research "papers" to conferences using the native Web stack should tell you something. That's just pathetic.

Myles Byrne’s photoMyles Byrne replied on

It is poignant to see comments to the effect of, 'great work, but until there is mass adoption, science publishing and funding are keeping us stuck in MSWord and PDF land'. Such comments could be cited within the paper itself as proof of its rationale. Career scientists invested in playing the NatureCellScience game may not feel a strong incentive to make immediate use of this work, for indeed, it is not instrumental to climbing the status ladders of legacy institutions. But such scientists also sound out-of-date, wedded to forms of collegiality that have grown increasingly ineffectual over the last decades, leading to the current crises in reproducibility, access, and career viability.

A mass shift in the social organisation of science and web publishing is obviously underway. What is the point of waiting for the shift to be over, when you can be part of it happening? Remember how science began: by pursuing the spirit of exploration and discovery against the established order, in the name of knowledge for all. Now, we're beginning again. Whether you're inside or outside the ivied walls of NatureCellScience, the collegial spirit of science lies in this direction.

Anonymous Reviewer replied on

In this paper, the authors propose a web-based framework for scholarly communication, addressing a significant number of the processes involved. For that purpose, they present an exhaustive list of requirements, and highlight how some of such requirements could be implemented using web-based technologies and techniques.

The paper, without any doubt, is interesting and well written. Nonetheless, in its current, I think it is not mature yet to be accepted as a long paper. Among other things, and more specifically:

  • It lacks an analysis of limitations, difficulties and risks related to the implementation of some of the scholarly communication processes should be presented.
  • It should present evidences (e.g., user interface screenshots, statistics about collected data and linked resources, evaluations, etc.) of the presumably implemented prototype.

I thus suggest the paper may be shortened and accepted as a short paper.

Anonymous Reviewer replied on

The paper addresses many important issues that concern the scholarly publication transition process, but the scope of the paper is too broad to be covered in a single paper. Due to this very broad scope, the following issues arise:

I. Most of the paper is dedicated to generalities (requirements) that are covered very sparsely (Impact metrics, reward systems, quality assurance) and little space is left for presenting the work done itself.

II. The detailed part that follows (starting from section 4: Realization), only addresses a fraction of the scope (Authoring and Publishing). Remaining issues are then listed as ongoing or future work (Archiving and Preservation, Collections, Profiles, Notifications, Feedback, and Versioning). This gives an overall impression that the work described is currently mostly in the stage of development and thus targeting a full paper publication is probably premature.

III. The paper presents several strong points that are unfortunately left on a very general level, concerning authoring and publishing of research papers or exploiting in full extent available web technology. At least a few examples of topics that could become more central in the paper (or, each of them could become a separate paper per se):

  • How concretely front end Web technologies, as mentioned at the beginning of the section 3.1, contribute to the transition process of scholarly communication;
  • How to implement a quality assurance process of scholarly publishing, mentioned as a requirement (section 2.3. Impact metrics an reward systems), measuring impact and implementing reward system, which is traditionally a major issue in the ongoing scholarly communication transition process.
  • How to efficiently attach research data to papers is also a very interesting problem to tackle.

The authors should perhaps make a choice whether the goal is (1) to provide a general overview of the currently ongoing scholarly communication transition (something along the format as was for example published in [1]) and that would more comprehensively embrace the scope of this complex subject. Or, (2) to select one (or a few) issues that the authors consider as central to the transition process of scholarly communication and focus on these in much closer detail (e.g. exploiting fronted web technologies, measuring impact, reusing research data,..), or (3) focus on the description of a tool that helps advancing the scholarly communication transition which should be described in much closer detail pinpointing major contributions of this tool.

[1] Kriegeskorte, 2012 - An emerging consensus for open evaluation: 18 visions for the future of scientific publishing. In: Frontiers in Computational Neuroscience, 6:94.

Anonymous Reviewer replied on

The paper proposes a research agenda for linked research. The authors argue for an infrastructure in which all artifacts of research publications - CfPs, papers, proceedings, as well as micro-structure parts of publications (proofs, arguments, etc.) - are provided using the Linked Data paradigm. With the dokieli authoring application and an example for semantically enriched CEUR proceedings, the authors also present an example toolstack and implementation.

While the paper is quite extensive about the requirements for a Linked Research tool stack, in my opinion, the potential impact and advantages are a bit vague. For example, the authors state that each paper can be linked to the original call for papers, but it is not clear what benefit such a link brings. Adding a high-level vision about possible applications - e.g., improved scientionmetrics, semantic search for related works, etc. - would clearly improve the paper and justify the impact.

The authors claim that a strength of their approach is that tables and diagrams can be automatically updated, since they are linked to the underlying, possibly dynamic data (section 1), and that multiple copies of the same paper can be provided at different locations (sec. 2.1). On the other hand, versioning is only mentioned as an item for future work. Here, I would have appreciated a more thorough discussion of the possible impact of such design decisions. Being able to change a publication after its acceptance is a crucial paradigm shift in scientific publication and communication, and hence, a critical reflection of such a design decision should be added. A similar point holds for the idea of LOCKSS (3.3.), where versioning is crucial.

Along similar lines, 3.2.1 discusses the submission by URL instead of by submitting a paper. Such an approach would allow authors to change a paper during the reviewing process, which could totally flaw the entire reviewing process. Again, this is a paradigm shift w.r.t. current submission practices, which calls for a deeper discussion.

In section 3.3, it is stated that "the author is responsible for maintaining their service and domain name". This is used as a contrast to centralized services like arXiv. However, even personal pages usually not technically maintained by the authors, but by their organization (university, company, etc.), or hosted by an internet service provider. Thus, the comparison is a bit unfair.

Section 5 is embarassingly short. Furthermore, many aforementioned services (e.g., ResearchGate) are missing from the comparison.

In total, the paper misses a good motivation, the related works section is substandard, and the impact of ground-breaking design decisions should be discussed more thoroughly.

Minor remarks: - Fig. 1 should contain a cyclic arrow from research to research labeled "reference" - in 4.1, a figure (dokieli-edit-menu) is referenced which does not exist. Maybe the authors submitted a URL, and it has been added in the meantime? ;-)

Anonymous Reviewer replied on

This paper presents a framework for scholarly communication, and addresses a number of the processes involved. In general the opinion of the three reviewers is that the paper is well written and interesting. However there a number of serious comments:

  • the scope of the paper is too broad
  • not mature yet to be accepted as long paper
  • it should present evidences
  • discussion of related work/implementations is insufficient

For these reasons I advice to accept this paper as poster