Exploring of the pre-publication communication

Two months ago, we announced a new free RePEc service that allows RePEc users making a fragmentation/annotation of reading papers and linking whole papers and/or its fragments by scientific relationships. These new tools are publicly available at sociorepec.org. It can help researchers with their everyday academic work, like discovery, analysis and writing new papers.

Using these tools researchers create private or public micro research outputs (annotations, relationships, etc.). If it is public, SocioRePEc can initiate direct scholarly communication between the researchers who used some papers to create micro outputs and the authors of the used papers. Such direct communication takes place while researchers are collecting findings, manipulating and organising the findings, e.g. as their manuscripts. Thus, researchers have an opportunity to come to scholarly communication before the manuscripts become traditional publications. We called this the pre-publication communication.

Recently we presented our vision of the possible impact of pre-publication communication in a position paper “End of Publication? Open access and a new scholarly communication technology“.

We are looking for partners (organizations or individuals) to explore the pre-publication communication.

We want to find out how useful pre-publication communication is. As the first step, we propose some experiments with SocioRePEc facilities:

1. Competitive selection. The basic pre-publication communication provided by SocioRePEc is public. That means the system allows experiments with creating some elements of competition. Members of the research community can trace the “author”<–>”user” pre-publication communication. Then they compete with the author by offering the user better research results or more efficient solution to her/his research problem.

2. Identification of the “neighbours”. We can think of researchers using research outputs of other researchers as “neighbours” in the global scientific labor division system. Pre-publication communication can help researchers to find out who their neighbours are. It can get the neighbours better collective intelligence. They can interactively adjust and adapt their “supply” and “demand” to get better mutual effect from their direct research cooperation.

3. Exploring threats. Do researchers appreciate that pre-publication communication is an instrument for identifying problems in and reducing potential threats to the credibility of their work? To shed some light on this question we need some additional qualitative study on how a research culture (formal and informal norms, rules and motivation) can be developed that can lead researchers to adopt pre-publication scholarly communication.

4. Publication as aggregation. It is also important to find out what could motivate scholars to adopt the idea that the future of research publication is aggregation. Neylon wrote about this: “If we think of publication as the act of bringing a set of things together and providing them with a coherent identity then that publication can be many things with many possible uses” [1]. Possible questions for the experiments are: What kind of forms in general can research outputs usage have in, say, economics? Will researchers agree to share micro research outputs in order to benefit from the pre-publication communication? Under what circumstances could researchers adopt the idea of “publication as aggregation”?

5. Transparency in research. What changes in research practice can initiate global pre-publication scholarly communication between authors and users of research outputs? How can this improve the transparency and credibility of their research findings? Answering these questions will imply some study of, for example, the community of RePEc users. We see them as a pro-active group of scholars open to innovations in the field of global scholarly communication technology.

We rely on grant support, sponsorship and community donations to get started.  Please consider making a donation or support us in another form (email for contacts).


[1] Neylon, C. The future of research communication is aggregation, Science in the Open Blog, Published: 10 APRIL 2010. Available online: http://cameronneylon.net/blog/the-future-of-research-communication-is-aggregation/

Annotating of papers in PDF

SocioRePEc provides an annotation service for papers in PDF files. RAS users can annotate RePEc papers in PDF files. At this time RePEc’s ArchEc and CitEc projects are building a collection of PDF files. SocioRePEc uses PDF files from this collection or, in some cases, from its original source.

If a PDF file specified in paper’s description at RePEc is available, the SocioRePEc system provides additional features. The features vary depending on whether a user is logged-in with RAS credentials or not. E.g. if you are logged-in, you will see your private annotations in paper’s PDF.

The first case is when you are not logged-in.

1. While a paper has no annotations yet, you will see a link to the PDF and another link to login screen –



2. When a paper already has annotation you will see two additional links. The first – opens the paper with the annotation. The second – opens the annotations only –


When you are logged-in you have additional features.

1. You will be able to open the PDF file with an interface that allows you to annotate the text –



2. If the paper already has some annotations  you can view them separately –



When you click the link “annotation and fragmentation” you will see a text of the paper. To create an annotation, just select a fragment of the text. Then you see two icons. The icon on the left opens a form to create an annotation. The icon on the right only highlights the fragment –




The form looks like this (on the right) –


At top of the form you see the text fragment that you selected.

The field “Title” is optional. If you leave it blank, the system will make one for you. It is used within the SocioRePEc tables of content.

The drop-down list “Relationship” works the same as the one you can use for annotating abstractsIf you are not the author of the paper (you are not claimed the paper as yours at RAS) you will see in the list the “Professional opinions and evaluation” taxonomy values:

  • responds positively to
  • innovative result
  • very interesting result
  • turning point for the science development
  • best, most relevant on the subject
  • responds negatively to
  • unscientific approach
  • potentially dangerous effect
  • result based on confusion
  • suspected plagiarism.

Please send your comments/suggestions on developing this list to the SocioRePEc admin.

The main window below the “Relationship” list allows you to enter an annotation text. It has the usual formatting options and includes the LaTex notations.

By default, your annotation are public. When you clicking on the white down arrow on the button “Make public” you can set the default to private.

When you create a private annotation, it will be available to you only. This private annotation will be displayed when you are logged-in into SocioRePEc and open the PDF.

When you create a public annotation it will first go to the SocioRePEc administrator. If your annotation is accepted, it will appear within the public content of SocioRePEc (other RePEc systems can also show annotations). It will be visible for readers of the paper approximately in a day.

Your public annotation will be shown with the PDF file by clicking on highlighted zone. When you click on the left arrow at the right top corner you will see a list of all annotations. –




Annotations in the list can have following colours:

  • white – it is your private annotation.
  • blue – it is your public annotation that has not been accepted by the administrator yet;
  • yellow – it is a public annotation created by the author of this paper
  • pink – it is a public annotation created by somebody else.



This tool is based on the open source software provided by Hypothes.is. See some additional information at the initial instruction from Hypothes.is. At SocioRePEc we don’t use some options of the Hypothes.is software.

Known problems of the SocioRePEc annotation service:

1. Some PDF files present content as an image not as a text. It can not be annotated.

2. Selecting fragments within the old PDF files can be a problem. E.g. some white spaces between words can be missed.

3. Currently this tool does not properly work at tablets (e.g. like iPad).

Signaling system

The SocioRePEc signaling system daily collects, process and publicly visualizes data about:

(1) the research outputs “production” as indicators of the new/updated input flow of papers from RePEc archives;

(2) the research outputs popularity as indicators of the papers’ views-downloads data from the SocioRePEc server http logs differently aggregated by time and by the object types: organizations, authors, archives, collections of papers, JEL codes, etc.;

(3) the research outputs relationship-usage characteristics as indicators of interconnections and interdependency between information objects.

These data are aggregated and presented in the system for the current day and for a variable period of time in different forms, including statistical portraits for: an institution, a person, an archive, a collection of papers, a single paper and a research area (by JEL codes).

1. Production data

The task of this section of the signaling system is to display what outputs the research community recently produced, who did this
(organizations and authors) at what collections and in what discipline areas (JEL codes). Additionally, users should see the most active “producers” (organizations and authors) and subject areas with the largest new papers input flow.

Currently the system maintains a daily display of the new/changed papers’ input flow to the RePEc dataset and presents these data as follows:

1.1. The daily updated SocioRePEc main page displays all collections with new/changed papers received by the system for the previous day by the Central European Time (CET) zone. For each collection with updates there are numbers of the new and modified documents for the current day.

1.2. At the main page there is also a daily graph of the “New documents for the last 28 days”. It displays the number of new additions by days with hyperlinks to the lists of updated collections on the selected date. It allows users to inspect collections with updates day by day.

1.3. Daily numbers of new/modified documents are available for each archive, section and collection of papers in their content lists. See an instruction on common features of the content lists at SocioRePEc.

1.4. Numbers of new documents for a specific author or an organization are available for each year. See instructions on
features of authors’ profile pages and organizational profile pages.

Additional Features: 1a) users should also have ability to select a time interval for displaying collections with updates, e.g. "-7" - for the last 7 days. Required work hours/budget: 5/$0. Current status: in the workplan. 

1b) Make "production" indicators by subject areas (JEL codes). Required work hours/budget: 50/$2500. Current status: waiting on a budget. 

1c) The signaling system should provides the "production" statistical portrait for authors and organizations including creation of non-traditional outputs (relationships, artifacts, etc.) on a daily base. Requred work hours/budget: 40/$2000. Current status: waiting on a budget. 

See the current ToDo list, vote for changing priority of tasks or support its implementation.


2. Popularity data

The task of this section of the signaling system is to display what the research community reads and how the community is active while reading. It means the system should display what papers and discipline areas are the most popular for the papers’ viewing and downloading. It also should show who are the most widely-read authors, organizations, collections of papers and subject areas.

Currently SocioRePEc builds the popularity indicators by processing the papers’ views/downloads statistical data from the SocioRePEc server http logs. It is based on the LogEc scheme with accurate exclusions from the data any bots’ activity, like Google and other crawlers. The main difference from LogEc is that at SocioRePEc the data collecting and indicators updating are organized as a daily routine. This makes for some kind of incompatibility with the monthly LogEc data. And it doesn’t allow a smooth intergration of LogEc data into the SocioRePEc statistical database.

The SocioRePEc popularity indicators at the moment display only a local picture, i.e. the SocioRePEc users’ activities. It does not consider the same papers’ popularity data from other RePEc-based information systems, because of their by-monthly base. But SocioRePEc provides a wider set of the popularity indicators with different aggregations by time, object type and relationships between objects. In particular this service produces daily popularity indicators for organizations, authors, archives, collections of papers, JEL codes, etc.

2.1. At the bottom section of the main page there is a table “List of the 60 most popular objects …” on the current date. It provides
links to lists of objects by types:”paper“, “author“, “organizations“, “collection” and “JEL code“, and by different aggregation time intervals: the last day, last 7 days, 30 days, 90 days. At the right of this table there is a graph of total numbers of hits by IP addresses, information objects, numbers of views and downloads for the last 60 days.

2.2. Each personal profile page [example] has links to: a) the most popular papers for the last 90 days claimed by the author as her/his own [example]; b) aggregated downloads/views data [example] for these papers; and c) the author’s statistics at LogEc.

2.3. Each organization’s profile page [example] has links to the 90 day popularity data of papers that belong either (1) to the
collections provided to RePEc by the organizations or (2) to the authors who claimed this organization as their workplace. In both
cases a user can see the most popular papers [examples by collections, by staff papers] and aggregated downloads/views data [examples by collections, by staff papers].

2.4. For each JEL code a user can see 90 days popularity list of papers which have this code in their metadata [example] and aggregated downloads/views data for all papers with this code [example]. Since JEL codes have some hierarchy the data of parent codes aggregates data all of its child codes.

2.5. Each collection [example] in its table of content has links to 90 days popularity data for the papers belonged to it [example] and aggregated downloads/views data for all papers from it [example]. A table of content of any archive [example] also includes links to 90 days popularity data of papers from all collections of this archive [example] and aggregated downloads/views data for all the archive’s papers [example].

2.6. A daily updated downloads/views data is available for each paper. It allows to see paper’s popularity in time [example].

Additional Features: 2a) To make visible e.g. 10 top objects of each popularity lists mentioned in 2.1.  This will show, on one page, all the most popular objects of all types. Required work hours/budget: 20/$1000. Current status: waiting on a budget.

See the current ToDo list, vote for changing priority of tasks or support its implementation.


3. Relationship-usage data

The task of this section of the signaling system is to display what classes of scientific relationships exist between research outputs. Data showing that one output was used (cited) while making another one is of particular interest. In addition, the system should display statistical distributions of outgoing and ingoing relationships for selected information objects. E.g. the system should visualize who (authors and organizations) produced what relationship classes, and what relationships are coming to outputs of these authors and organizations.

Currently at SocioRePEc most information objects already have relationships.

3.1. The table of content of each collection, section and archive has a column with the total number of outgoing and ingoing relationships for any objects. The content list can be sorted by numbers of the outgoing or ingoing relationships to see on the top of the list the objects with the largest number of relationships [example].

3.2. Most of personal profiles of authors [example] have a link to a relationships table with numbers and a structure of outgoing and ingoing relationships for the papers claimed by the author [example].

3.3. Almost each profile of organization [example] has a link to a relationship table with numbers and a structure of outgoing and ingoing relationships for papers belonged to the collections provided by this organization and to the authors claimed this organization as their workplace [example].


Additional Features: 3a) to make one page that allows to compare the "production", the "popularity" and the "relationship-usage" data. Required work hours/budget: 40/$2000. Current status: waiting on a budget.
See the current ToDo list, vote for changing priority of tasks or support its implementation.


Please share with us in comments what you think about the proposed tasks of the signaling system and its current implementation at SocioRePEc. Please also suggest for the ToDo list how we can improve and develop the usability of this service for the community.

Strategic goal and the concept

The SocioRePEc initiative has an ambitious strategic goal. We want to build an innovative research information system for the Economics research community by joint efforts of the RePEc users (researchers and institutions) and the RePEc developers.  The system should improve global research communication and cooperation compared with the traditional system provided by academic journals and publishers.

Traditional global scholarly communication scheme

In the traditional system researchers find and read papers. While reading they select from the paper’s text some interesting fragments – research artifacts. Researchers mentally manipulate the new artifacts to discover scientific relationships with the artifacts they already know about. This discovery can be successful. If it is successful, the researchers found a set of new artifacts and relationships between them that looks like a new research result. In that case the researchers organize these materials as a canonical research paper with the proper structure, citations, list of references, etc. They submit a paper to a publisher. After publication, researchers typically are monitoring the papers’ citing and waiting some occasional comments on it. And this cycle as at Figure 1 repeats by researchers again and again with new reading papers.


Fig. 1. Traditional scholarly communication scheme through academic journals and publishers infrastructure

Scholarly communication supported by the traditional scheme at Figure 1 has obvious weaknesses and shortcoming.

While manipulating artifacts researchers make with them some experiments or “trials and fails” actions. As a result of such “trials and fails” usage actions researchers either accept some artifacts by connecting them with their own research assets or reject them. At the end in the both cases researchers get some individual experience about the scientific value and/or the potential impact of these tested research artifacts. This “empirical” information can be very useful for authors of the tested artifacts and for the community at large. But currently only a small part of this information becomes public and reaches, e.g. though the citations, the authors and the community.

In the traditional scheme of the global scholarly communication most authors get hardly any feedback on their papers. They usually don’t know when someone tried to use their research results and failed. If someones really used the authors’ results and cited them in their own papers, the authors had no any chance to assist them on the process with proper using of the results, to correct a wrong usage or to get a better effect from using their results. From the point of the general cooperation efficiency inside research labor division system this situation looks like a really inappropriate.

Generally, researchers don’t manipulate research papers but the research artifacts contained in the papers and the relationships between them. Why they could not share these artifacts and relationships as their micro research outputs with the other researchers as is? Scientific circulation of such micro research outputs in parallel with canonical papers can give the community multiple benefits, like a rise of the traditional papers’ quality level and so on.

For example, it would allow to initiate immediate scholarly communication among the “neighbor nodes” of the research labor division system. Such communication on the “trails and fails” stage could involve them into live cooperation with mutual benefits. Researchers after some interactions with their “neighbors” (authors of the used artifacts) can update and improve their micro outputs. As well, they can improve their canonical research papers produced after such pre-publish interactions.

SocioRePEc strategic concept

The SocioRePEc can give researchers ability to create and share micro outputs, including relationships. The system can collect and visualize data on “trials and fails” actions which appear when someone create scientific relationships. We can imagine SocioRePEc users, which create relationships between own outputs and papers/artifacts of other author as “neighbors” of the labor division system. The system will help researchers to find out their “neighbors”. It will initiate scholarly communication between the “neighbor nodes” to get better collective intelligence and to have benefits from direct research cooperation.

From the technology side, the central idea is to embed the SocioRePEc system into the traditional communication scheme.

Fig. 2. Scholarly communication scheme as a combination of SocioRePEc facilities and academic journals/publishers infrastructure


SocioRePEc gives researchers the following additional abilities (see at Figure 2 points at the circle around artifacts):

a) to save and store research artifacts from the paper being read and artifacts created by the users;

b) to share research artifacts with the community and to share knowledge or hypotheses about scientific relationships between artifacts and/or papers by linking them using a taxonomy of the applicable relationship classes;

c) to allow authors of the linked artifacts/papers to be immediately notified why, where and by whom their outputs were linked and how their research results were used;

d) to allow authors of linked artifacts/papers to publicly express their attitudes to these actions. They can state what they think about using their results.

e) to allow some coordination between a researcher who created an artifact and another researcher who is using this artifact. They can adjust and adapt their “supply” and “demand” to get better mutual effect from their cooperation.

The basic “author<–>user” communication provided by the system is public. It can creates additional competition within general research cooperation. Researchers who are initially are out of the specific “author<–>user” communication can trace the communication and can compete with the “author” by offering the “user” better research artifact or more efficient solution for her/his “demand”.

A combination of the these SocioRePEc facilities with the existed academic journals/publishers infrastructure creates the research community some new opportunities.

Please share with us in comments what you think about the SocioRePEc goal and the concept.


SocioRePEc team


SocioRePEc builds on the RePEc dataset. In addition, we use: an approach and services of the Semantically Enrichable Research Information System SocioNet [1], CERIF research data model [2], Open Annotation data model [3] and the protocol [4], Open Archives Initiative Protocol Metadata Harvesting [5]. We share the Open Science ideas and work to support it by developing the SocioRePEc facilities.

SocioRePEc team

The core SocioRePEc team has been working together more than 15 years. The current team includes:

  • Sergey Parinov. He is the project leader. He does conceptual development, system architecture, XSLT, testing, and some admin work.
  • Victor Lyapunov. He is the chief system administrator. He also does software architecture, writes applications in Perl and shell scripts, etc.
  • Roman Puzyrev. He codes in Perl, Javascript, XSLT, etc.
  • Mikhail Kogalovsky joined the core team in 2007. His main contributions are conceptual development, taxonomy of scientific relationships and testing.
  • Thomas Krichel, the creator of RePEc, has been advising the project since 2014.

The team is open to new members. Please contact us at admin@sociorepec.org.

We will add here names and roles of all who will contribute to the SocioRePEc project.



[1] Semantically Enrichable Research Information System SocioNet

[2] http://eurocris.org/cerif/main-features-cerif

[3] http://www.w3.org/TR/2014/WD-annotation-model-20141211/http://www.openannotation.org/spec/core/

[4] http://www.w3.org/TR/annotation-protocol/

[5] http://www.openarchives.org/pmh/


SocioRePEc is a community project

SocioRePEc is a RePEc-based public research information system. It is an open non-commercial community project funded by grants and donations. Organization can sponsor this project.

The project and the research information system are maintained by the SocioRePEc team.  The team intends to further develop the system based on requests by users and in collaboration with them.

SocioRePEc users can submit ideas/request to improve and develop the system facilities. They can vote to change priorities of the current SocioRePEc ToDo list. They can donate funding and allocate it to the specific tasks. Donations without relation to specific tasks are allocated to the top priority tasks at the current ToDo list. 

The users can watch how funds arrive. They can control and publicly discuss how the SocioRePEc team implements tasks from the ToDo list and spend the funds.

The SocioRePEc team publicly discusses with the users their proposed ideas and requests for the system development. The team makes public draft evaluation of necessary funding for the proposed new tasks and updates the current SocioRePEc ToDo list. The SocioRePEc users and the team can publicly discuss the claimed costs for the tasks from the ToDo list.

The SocioRePEc team performs funded tasks first.

Developers interested to join the SocioRePEc team can propose their help with tasks implementation from the ToDo list.

Without necessary funds, the SocioRePEc team is maintaining the SocioRePEc system and provides the helpdesk support for its users.

Please share with us in comments your vision how we can improve this organizational mechanism.

What is new?

SocioRePEc is a RePEc-based open research information system. It is driven by join efforts of two communities. We have the researcher community as users of the system. And we have the developers as builders of the system. The latter group forms the SocioRePEc team.

Compared with already existing RePEc-based systems, such as EconPapers and IDEAS, the SocioRePEc provides its users with following main added value facilities:

1. A statistical signaling system. It gives users daily updated pictures of “production”, “popularity” and “usage” activities behind changes of the RePEc data. This system can help users with tracing what is going on in the economics discipline and with
optimizing their professional activity.

2. Visualization of and navigation by all types of scientific relationships between RePEc information objects. As an example: institutional profile <–> personal profiles <–> staff papers <–> cited papers. We produce a picture of and a navigation by interrelations and interdependency over RePEc data.

3. A new “dimension” for a public scientific creativity over RePEc data. While reading papers users can create and share with the community non-traditional forms of research outputs. These include relationships, assertions, quotations and artifacts. Such outputs are small and easy to create. They are better suited for digital re-use than traditional papers.

4. New opportunities for more productive research through the specific scholarly communication among RePEc authors who makes scientific relationships (e.g. citations) with papers for the RePEc content. Such communication is initiated by the information system when e.g. one author used (cited) a research output of another one. The system services both authors as neighbor nodes of the global division of labor in economic research to get them a chance for better collective intelligence and to have benefits from the direct research coordination and cooperation.

In the long run, the SocioRePEc proposes an initiative for the Economics research community with an ambitious strategic goal. The goal is to build a more efficient research environment for the community researchers in Economics than the one provided now by the traditional system of academic publishing. We try to provide better conditions for information orientation in the profession using SocioRePEc signaling system. We introduce researchers to a new dimension for scientific creativity, i.e. research micro outputs including scientific relationships. We hope to significantly improves the communication between authors and users of research outputs.  Overall we support and develop the Open Science ideas.

Please express your opinion in the form below. Let’s see what the community of RePEc users thinks about SocioRePEc –

Please also use comments to share with us what you think about the current SocioRePEc and the proposed strategic goal for its further development.

If you like, you can be involved and/or support our work with the SocioRePEc project.