Information System on Digital Humanities: an experiment of describing the Internet resources of Russian archives

Introduction

INION RAS and the Institute of Digital Humanities Research (DHRI) of the Siberian Federal University have signed an agreement on the joint creation and support of a Reference and Information System on Digital Humanities (SIS TSG).  This system should contain information about CG objects created in the world and Russia — resources, services, standards, tools, projects — in order to reuse them and avoid duplication during development. It is assumed that such a CG system can be used:

· when developing R&D programs;

· during the preparation and examination of applications for grants, projects, research topics;

· to evaluate the results of research, as well as institutions and research units;

· to coordinate the digitization of CG objects;

· for information support of scientific and educational activities on the use of computer technologies in the field of humanities and social sciences.

The CG system should ensure increased visibility, citation and reuse of CG objects created in Russia.

The long-term goal of the SIS TSG should be to form the basis for solving the problem of determining the optimal level of integration of resources and collaboration of activities in the field of digital humanities and more broadly, when creating the infrastructure of digital science in Russia.

The thematic boundaries of the infosphere, which should be reflected in the SIS TSG, are still being discussed and will be clarified in the process of designing and practical filling of the content of the SIS TSG. These issues are partially considered in ^[1]. However, it is obvious that in the institutional context, the resources of libraries, archives and museums are included in the infosphere of the TSG. Therefore, it is these resources, and mainly of a socio-humanitarian orientation, that should be investigated first.

 The results of the presented research should be used in the formation of the content of the CG system.

The purpose of the study

The purpose of this study, to put it briefly, is to conduct an inventory of the resources of Russian archives presented on the Internet and to form a database of descriptions of these resources. This database can be considered as a pilot version of the content of the designed system. Obviously, at the same time, it is necessary to systematize resources and determine the structure and composition of the database. 

Background of the issue

The study of the Internet resources of domestic archives has its own history. A special contribution to it was made by E.V.Bobrova, who conducted several studies of these resources, including in 2005 ^[2], as well as in 2007 and in 2018 ^[3,4]. E.V.Bobrova analyzed in detail the state of archival sites, both their content and technological features, as well as, what is especially important is the user aspects.  Her works also provide statistics on the state of Internet resources and the dynamics of their development.

Many important issues of the organization of archival Internet resources are considered in the fundamental monographs of Yu.Yu. Yumasheva ^[5,6]. The first of them, devoted to the informatization of archival business in Russia in 1991-2015, examines in a broad context the issues of creating thematic archival databases and information resources, the history of archival site construction, etc.  The second monograph by Yu.Yu.Yumasheva is devoted to the problems of digital transformation of traditional audiovisual archives, the translation of audio, photo, and film documents into digital form, the creation of electronic resources based on created copies and their presentation on the Internet. The monograph is based on a database created by the author, which presents information resources available on the world Wide Web based on digitized bibliographic catalogs, scientific reference apparatus of archives and museum records, as well as electronic copies of objects.

We should also note the work of I.N. Kiselev ^[7], devoted to the analysis of reference books on archival documents posted online on the websites of state and municipal archives, as well as authorized executive authorities of the subjects of the Russian Federation in the field of archival affairs and on the Archives of Russia portal, the degree of their accessibility to Internet users, formats and some other properties. All data is collected by the author by continuous browsing of sites as of October–November 2017

An interesting experience of systematization of historically-oriented information systems was undertaken by a team of Perm historians led by S.I. Kornienko ^[8]. The generalized results are presented in the collective monograph "Historical information systems: theory and practice", in which historical-oriented information systems are considered as an important element of the information environment of the humanities ^[9].

The features of the international experience of describing resources and tools used in digital humanities are factually presented in a collective article devoted to the merger of the DiRT (Digital Research Tools) catalog with the TAPoR (Text Analysis Portal for Research) portal ^[10].

However, the cited authors did not address methodological issues of accounting and cataloging of Internet resources, which became the content of this study. 

Scope of research

In this study, the resources of federal archives, state archives of the subjects of the Russian Federation presented on the Internet, as well as departmental archives in a small number, if they have socio-humanitarian content, were studied. The administrative bodies of the archives of the subjects of the Russian Federation were included in the database if they have their own website hosting resources of interest. About half of the subjects of the Russian Federation turned out to be such.

Archives with dominant technical or natural-scientific content were not included in the study.  Also, at this stage, the resources of municipal archives were not considered.

The main source of information on the composition of state archives, as well as archival management bodies, was the Archives of Russia portal ^[11], where there is a special section "Archival management bodies and archives of the subjects of the Russian Federation".

Objects of research

In principle, the following types of information objects belong to the objects intended for inclusion in the CG system:

· Institutions – i.e. management bodies, organizations, or subdivisions of institutions that have their own Internet resources.

· Resources – i.e. arrays of documents or data (sites, sections or pages of sites, databases, datasets, collections, documents or their descriptions, catalogs, reference books, etc.).

· Services – available automated processes for the implementation of user tasks.

· Standards – standards, methods, indexing languages and other means of creating resources or implementing services.

· Tools – software products used to create resources or implement services.

· Projects – objects that cannot be attributed to the listed types, but are of interest from the point of view of the objectives of the CG system.

These types were identified based on the analysis of foreign experience described in ^[1]. In the study of archival resources, only three types of objects were practically analyzed: institutions, resources and services. Standards and tools were considered at other stages of this study. As for projects, in relation to archival activities, the distinction between projects and resources turned out to be difficult and irrelevant. 

The list of types of institutions included in the review was listed above.

As for resources and services, two methodological problems were solved during the study. Firstly, it is the task of determining the optimal level of detail for allocating a resource as an accounting object. Secondly, it is a distinction between typical resources and services, the inclusion of which in the CG system was deemed inappropriate, from specific ones that are of interest to an external user from the point of view of the goals of the CG system. Further, these problems are discussed in more detail.

The problem of the level of detail of resources

Determining the optimal level of detail of accounting objects is one of the most difficult methodological problems when creating navigation systems in the information space. For example, in the metadata application guide for the well-known OLAC linguistic data repository ^[12], this problem occupies one of the central places.

Recall that we call resources named (allocated) arrays of documents/data, which are the objects of accounting and search. Since we are talking about network resources, the array identifier is usually a network address or URL. Other methods of array identification, such as DOI, are not widespread enough yet and cannot be used as a selection tool.

 On the one hand, the detail of the allocated arrays should not be too detailed so that the navigation scheme remains visible and the arrays of homogeneous data have a common name. Therefore, navigation usually does not descend to the level of a separate document or dataset.

On the other hand, it is necessary to give an idea of the composition of complex heterogeneous information objects, such as websites or portals. 

On the third hand, when choosing the level of detail when forming a resource, developers could be guided by a variety of ideas or even technical reasons.

Consider an example.  On the website of the National Archive of the Republic of Sakha (Yakutia) ^[13] there is a section called Information and educational section on the history of Yakutia, which is of interest to the SIS TSG. This section has the following structure.

• • • • • • • • • Ancient and medieval Yakutia
                • Yakuts
                • The generic composition of the Yakuts
                • Yakutia is part of the Russian state
                  • • • • • • • The entry of Yakutia into Russia
                              • Yakutsk and Russian geographical discoveries of the XVII century
                              • Yasak
                              • Yakutsk
                              • city of Olekminsk
                              • Vilyuysk
                              • Verkhoyansk
                              • Srednekolymsk
                • Yakutia in the XVIII-XIX centuries.
                • Yakutia at the beginning of the twentieth century. (1900-1917)
                • February and October Revolutions
                • Civil War
                • Yakutia during the Great Patriotic War of 1941-1945 .
                • 1922-1929
                • 1930-1935
                • 1936-1941
                • Yakutia in the post-Soviet period. 1992-1999

However, this section entirely does not have its own identifier in the form of an address, which is due to technical reasons. Also, there is no separate address for categories that have a second level, such asYakutia is part of the Russian state. Therefore, there is a dilemma: either to limit yourself to the level of the entire site, or to include in the database the CIS TSG as separate resources of the rubric at all levels.

Another example. The scientific reference apparatus of this archive is presented on most archive sites. As an example, here is the section  Reference and search tools of the site The National Archive of the Republic of Tatarstan ^[14].

^{GUIDEBOOKS AND REFERENCE BOOKS ON FUNDS}

^{Central State Archive of Historical and Political Documentation of the Republic of Tatarstan. Travel guide.}

^{Directory of personnel documents on the funds of the State Archive of the Republic of Tatarstan (2011-2018)}

^{State Archive of Personnel Documents of the Republic of Tatarstan: Quick Reference (2005-2010)}

^{A short guide to the funds of the National Archive of the Republic of Tatarstan (2010-2014).}

^{A short guide to the funds of the National Archive of the Republic of Tatarstan (2005-2009)}

^{HELP-ORIENTATIONS}

^{Form lists}

^{Household books}

^{Reference guide about the census forms for the first general population census of the Russian Empire in 1897}

^{List of evacuation hospitals located on the territory of the Tatar ASSR during the Great Patriotic War of 1941-1945}

^POINTERS

^{Index of drawings and plans of Kazan (volume 1)}

^{Index of drawings and plans of Kazan (volume 2)}

^{Index of drawings and plans of Kazan (volume 3)}

^{Inter-fund index of clearing statements}

^{Interfund index of maps and plans of settlements, land and forest plots...}

In this case, on the contrary, both the entire section and the arrays/documents of the second and third levels have independent names and identifier addresses and can be included in the database of the CIS TSG as independent resources. However, the question arises about the expediency of such detailing for the tasks of the CG SIS.

It is obvious that the question of the optimal level of detail in the description of resources will be finally resolved during the operation of the CG system, taking into account the interest of users. So far, for the described study, the principle of the highest level of identification among homogeneous data with a common identifier has been applied. I.e. for this example, the section  The reference and search tools were included in the database entirely. It is assumed that with the practical functioning of the CG system, the optimal level of detail can be changed in one direction or the other.

The problem of typical and specific resources and services

It is easy to see that the sites of Russian archives are generally very saturated with a variety of information. However, it is unclear at this stage to what extent this information will be useful for users of the CG system.

The main criterion for the inclusion of resources in the database of the CIS TSG was adopted as follows.  Resources specific to this archive were selected, and resources typical or standard for many archives were not included. At the same time, as mentioned above, individual documents were not included unless they were lists of data, such as guidebooks, indexes, bibliographies, etc.

Lists of projects, exhibitions, services, resources and other information objects have taken an important place among the selected resources. These lists should serve as a basis for the possible expansion of the content of the CG system.

Of course, only digital resources were included, but presented in any formats, both text and graphic. Descriptions of traditional resources, in particular libraries of archives, were not included if they did not have electronic services or resources

A special question about the inclusion of services (services) in the database of the SIS. Recall that the task of the CG SIS is to reflect, first of all, the digital infosphere of digital humanities.  On the other hand, most archival services are typical for the archival industry. Therefore, the database included mainly services performed using the database, as well as lists of services.

Conferences, seminars and other events should also be mentioned.  conducted by archives. At this stage, this information was not included, because it is often of a non-permanent nature. In addition, there are special resources that reflect information about events, so their direct duplication looks impractical.  However, the issue of the reflection of scientific events in the CIS TSG requires a separate discussion.

As a result, the following approximate composition of resources selected from archive sites for inclusion in the CG system was obtained. The list is represented by keywords reflecting the type of information resources.

• Archival collections
• Audiovisual resources
• Databases
• Bibliographies
• Virtual resources
• Exhibitions
• Genealogy
• Geographic information systems
• Geodata 
• Internet projects
• Information and search engines
• Catalogs
• Museum collections
• Multimedia resources
• Open data sets
• Scientific reference apparatus
• Inventories
• Lists (lists) of persons,
• Lists of projects
• Lists of resources
• Periodicals
• Personal collections
• Portals of a wide profile
• Travel guides
• Dictionaries 
• References
• Thematic collections
• Thematic collections
• Pointers
• Photo Galleries
• Chronological indexes and calendars
• Digital editions and archive publications
• Electronic libraries
• Encyclopedias

To complete the picture, we will also give an approximate list of typical sections of archival sites, which, as a rule, were not included in the database of the CIS TSG.

• Archive structure
• Archive history
• Composition of archive documents
• New arrivals
• Sources of archive acquisition
• Scientific Council of the Archive
• Expert verification Commission of the archive
• Archive Partners
• Archive Honor Board
• Archive Reports
• Normative documents of the archive
• Anti-corruption
• Events
• Reading room

The structure of the CIS TSG database

The initial version of the database is a structure consisting of the following fields of the object representation:

o An acronym of an object that also performs the functions of linking the resource with the owner;

o          Full name of the object in the original language;

o          Russian Russian translation of the name or the name of the object in Russian;

o The web address of the object;

o Object type (according to the Research Objects section.);

o          The type or functionality of the object (currently 1-3 keywords separated by commas);

o Category of the upgraded GRNTI (currently 1-3 codes of categories separated by commas).

This structure has already been tested when creating some navigators functionally similar to the CG system ^[15,16], and it turned out to be quite workable.

Here we will briefly discuss some of the problems arising from this structure. First of all, this is an ID problem. The idea was that the acronym of an institution should serve as a key to link resources, services and projects belonging to this institution with the owner.  This will allow you to show all the resources of this institution when searching, which is very convenient.

In principle, in most cases this is quite feasible and the ownership of the resource to the owner is determined unambiguously. However, there are cases when some difficulties arise.

Firstly, it is not always possible to use the official acronyms of the organization as acronyms, since they may coincide, therefore it is necessary to use artificial acronyms.

Secondly, some resources may belong to several owners, i.e. the ownership relationship may not be unambiguous.

Thirdly, there are resources for which the owner institution cannot be established.

It is obvious that in the practical implementation of the CG system, technological solutions to these problems should be provided.

Indexing languages

Next, we will consider the possibilities and prospects of using some indexing tools for the CG system, based on the main task of this system, which can be technologically defined as the search for analogues for projects, resources and services in the field of digital humanities. In more detail, indexing tools for digital humanities as a whole are analyzed in ^[17], and here we will consider the application of  indexing languages for describing archive resources.

In the described study, in addition to the general division of information objects into types, 1) indexing of objects was used to determine the type of object using keywords; 2) determining the subject using the upgraded GRNTI rubricator.

Keywords for determining the type of object Information resources are mainly given above, in the section   The problem of typical and specific resources and services

The following keywords were used to determine the type of the Institution object

·         Archive of the subject of the Russian Federation

·         Association

· Libraries of federal executive authorities

· Libraries of branch academies

· Depository Archive

· Research institutes

·         Archive Management Authority

·         Industry Academies

·         Federal Archives

Keywords were used to determine the type of the Service object:

· Information service

· IPS

· Genealogy

·         List of services

Such a modest list of types of services was explained above: services were included in the database in a very limited volume.

 In general, determining the types of information objects using an open list of keywords that can also be used as search tags does not cause problems.

The situation with determining the subject of resources using GRNTI looks more complicated. In cit. The paper ^[1] described the use of the modernized GRNTI for indexing foreign projects in digital humanities. This experience can be recognized as quite successful, probably due to the fact that foreign projects had an obvious link to a certain topic and/or to certain information technologies. This is not the case with regard to archival resources. With the exception of archival collections related to military history and a small number of thematic collections on other branches or eras, most archival resources are not thematically colored and they had to use the general heading 13.71 Archival business. Archival science. Recall the division of this heading in GRNTI, and the heading 13.71.85 Digital Archives was introduced by us in the order of modernization of GRNTI.

13.71 Archival case. Archival studies

13.71.01          General questions

13.71.07 Theory and practice of archival business

13.71.09          History of archival business

13.71.15          Organization and management in the field of archival affairs

13.71.85 Digital archives

13.71.91          Archival business in individual countries

13.71.96          Natural science and technical support of archival business

Obviously, this division does not allow thematically structuring most of the archival resources, the composition of which is shown above. At the same time, it should be based on the general methodological principle of using classification for navigation. According to this principle, the categorization should be such that the number of objects in each category is visible for intelligent viewing, i.e. not exceed 30-40 points.

It follows from this that for the tasks of the CIS TSG in relation to archival resources, it is necessary either to develop a fundamentally different scheme, or to abandon the thematic classification, limiting it to subject tags.

The main quantitative results of the study

In total, during the study of the sites of the Russian archival industry, taking into account the limitations and methodological principles described above, about 1.5 thousand information objects were allocated, including resources - 1250, institutions – 256, services - 28. Table 1 shows quantitative data by type of objects.

Table.1 Number of information objects

Conclusion

The conducted research made it possible to present quite clearly the structure, quantitative and qualitative characteristics of Internet resources posted on the websites of Russian archives. Of course, this is not the final data: in the process of actual filling and operation of the CG system, both the composition of information objects and the means of describing them can and should change and develop.  The study also identified the main methodological problems of accounting and cataloging of Internet resources that need to be solved.  Above, we pointed out the need to solve the problems of resource granularity or thematic classification. Obviously, an important role in the way of organizing data will also be played by the interface of the CG system, which has yet to be developed. In the near future, the information system will be available online for pilot use by those who wish. You should also keep in mind,  that the resources of archives are only part of the content of the projected digital humanities system.