A historical source in the digital Age: towards the publication of a monograph by Yu. Yu. Yumasheva

Zlobin Evgenii Valentinovich ORCID: 0000-0002-9292-6083 PhD in History independent researcher 125252, Russia, Moscow, Sorge str., 20, sq. 10 zlobinev@mail.ru

And if I'm restless and can't sleep

Or is there no face on me with a hangover? —

I'll open the codex on any page.,

And I can't, I'm reading to the end.

V. Vysotsky.

The new monograph by Yu. Yumasheva ^[1] has been read by me three times. And every time I discovered something new. This book is multi—layered, and you can really read it from anywhere, because literally every paragraph is a small, complete source study miniature. Yulia Yurievna is distinguished by her scrupulousness and meticulousness as a research scientist. To paraphrase another famous poet: "She wants to get to the bottom of everything..." And in principle, she succeeds. The list of sources and literature used, which includes more than 1,600 titles, can serve as a confirmation of the enormous preparatory work done. A significant part of which is not in Russian. As well as 755 footnotes, the total volume of which is comparable to the volume of the main text of the work. Another thing that surprised me when reading the monograph was the synchronicity of our thoughts with the author. When reading the book, questions repeatedly arose, and literally on the next page I found the answers to them. The appearance of this work became possible due to the unique work experience of Y. Y. Yumasheva in both the archival, museum and publishing fields. An experience that hardly anyone else who studies sources has.

The relevance of the work can hardly be overestimated. For more than half a century, historians have been actively trying to work with various kinds of electronic documents (ED), or electronic historical sources (EII), this term is introduced by the author, but a generalizing source work has appeared only now. The scale and essence of the problems that historians face can be judged by some quantitative parameters of the EII, which have been deposited in the General Assembly of the Russian Federation. For example, based on the materials of the All-Russian Population Censuses of 2002 and 2010 and the agricultural censuses of 2006 and 2016. Thus, 250 million sheets were scanned during the processing of primary materials in VN-2002, and 226 million documents were scanned in VSP-2006. The volume of files exceeded hundreds of gigabytes (hereinafter Gb). These arrays were transferred to archival storage on DVD-R discs with a capacity of 4.7 GB. According to VSP-2006, the transfer took place in 2010. 694 DVDs with a volume of more than 3 terabytes (hereinafter Tb) were accepted. Discs with census data have been uploaded to robotic libraries installed in the archive building on Berezhkovskaya Embankment. The volume of the VPN-2010 information array was 1,150 DVDs, that is, about 5 TB.^[2] And for the VSP-2016 only in the Altai Territory, one of the 85 subjects of the Russian Federation — 258 GB. That is, in total, the total volume of such a source in the Russian Federation is about 20 TB. At the same time, in addition to the actual VSP-2016 materials, software has been adopted for archival storage for viewing various file forms and checking checksums. ^[3] The storage of such ED continues, and entire "electronic files" are already being formed. ^[4] This is a relatively new term for archival science ^[5]. Working with such historical sources (AI) requires certain special skills from the historian. This is probably why, according to information from the archive, none of the researchers has accessed these data arrays since they were accepted for storage.

There are two possible levels of approaches to the description of the published monograph. The top or most general level is an assessment of the general concept of defining a new type of historical source — the EII. The lower level is some private remarks that arise when reading the work. As for the general idea of the book about the identification of EII as a new type of historical source. Despite the harmony of the presentation and argumentation, supported by numerous references and the detailed classification scheme developed by the author [1. p.127], it seems that the new type of sources is unlikely to be accepted by classical sources who are quite conservative and at least a couple of steps behind technological progress. For example, in our opinion, the classic and voluminous textbook of the Russian State University of Economics, which was published at the turn of the third millennium, the authors only mention "... the technical means of modern times, [which], on the one hand, facilitate access to information, shorten its path to the consumer, on the other hand, contribute to even faster to increase the total amount of information, create new types of [sic! EZ] sources. The object of source studies — the historical source — is becoming even wider." ^{[6, p. 29]} However, what these new types of historical sources are, and in which direction the historical source is expanded in the future, the authors do not disclose. With the exception of a brief mention of "technotronic sources" ^{[6, p. 124]}. We also note that the Internet is not mentioned in this textbook for humanities.

In comparison with the introduction of a new type of EII sources into scientific circulation by the author of the monograph, two other possible options are closer to the reviewer, although they are somewhat opposite in essence. First, instead of the definition of ED given in the work, use the wording "historical source in electronic form", derived from the official definition of ED [1. p. 360]. And apply it to all classical types and types of AI. Or secondly, using the author's logic, introduce a definition of "technotronic historical sources" (TII), and then consider EII as a subset of this type.

One of the most interesting and promising sections for a historian is the section of the monograph devoted to the application of artificial intelligence (AI) ^{[1, pp. 166-171]}. The use of AI technologies in various fields of human activity not only increases the efficiency and productivity of intellectual work, but also threatens to completely replace natural intelligence, its degradation and the inability to generate new knowledge. After analyzing some real-world examples of the introduction of AI technologies in historical research, the author makes a pessimistic conclusion: "... currently, there are no reliable methods to uniquely identify the participation of artificial intelligence in the creation of AI" ^{[1, p. 169]}. It remains to be hoped that further development of technologies will allow this problem to be successfully solved.

The author of the monograph paid great attention to the description and analysis of metadata, as the most important element of EII, without which it is impossible to fully involve them in scientific circulation. Changing at different stages of the EII life cycle, metadata itself turns into "... a kind of historical source..." ^{[1, p. 197]}. With various transformations of the EI, editing, and transfer to new media, metadata is constantly increasing in volume. Metadata, having increased in volume to some critical value, according to the author: "... will make working with EII almost impossible ..." ^{[1, p. 199]}. In principle, this could be the end of the description of EII as a new type of source due to the potential impossibility of working with them. However, in our opinion, this conclusion is unnecessarily pessimistic. The available technologies and capabilities of computer hardware and software make it possible to process fairly large amounts of historical information, examples of which we have previously given.

As for some particular remarks that do not detract from the many advantages of this scientific work, but are intended only to highlight some of its points, I would like to share the following considerations. Within the framework of one monograph, "it is impossible to embrace the immensity" (although Yulia Yurievna partially succeeded in this), but it would probably be correct if at the beginning of the work at least a brief definition of information was given in the author's understanding. It is clear that this topic is endless, and there are many more such definitions than 150 for ED [1. p. 354], which are mentioned in the monograph, but such a description of the author's position would make it clearer to use the terms "source information", "historical information" and, especially, "electronic information" [1. p. 25].

Describing EII, Yulia Yurievna equates all electronic documents/sources to digital ones. However, this is not entirely true. Some of the documents deposited in the archives are electronic in nature and do not fall into the EII due to the fact that they are analog. These include, for example, reel-to-reel tape recordings of N.S. Khrushchev's memoirs preserved in the Russian Academy of Sciences, and similar recordings of interviews with academicians in the archive of the Russian Academy of Sciences. As well as numerous videos on BETACAM analog video cassettes (and VHS) in the video archives of media companies. The reception of such documents for archival storage continues to this day. In 2018, the State Security Service of the Russian Federation received: a videotape recording the conference of the magazine "Sowing" in New York; audio cassettes with recordings of O.S. Polyakov's memoirs and recordings of Russian folk songs performed by the Volga Choir and the Sveshnikov Choir. ^[7]

Defining the grounds for the classification of EII, the author of the monograph writes that the rapid development of technology "... has turned the criterion of the carrier into inapplicable ..." [1. p. 123] for this purpose. To some extent, we can agree with this. However, it is the media that determine those EIAs for which the Digital Dark Era or "black hole" described by the author has already begun [1. p. 148]. These include all EIS on punched cards, punched tapes, magnetic tapes (partially), floppy disks (or floppy disks). In the near future, due to the decommissioning of optical discs, the same can be said about DVD-R media, on which, according to the instructions of the Federal Archive of Electronic Technologies, they must be deposited. There are examples of successful migration and recovery of lost information, but they are rare, require high costs and the availability of read/write devices. Thirty years ago, the author of the review managed to transfer the materials of one of the censuses from punched cards to magnetic tape, and then to a floppy disk. Currently, due to the widespread disposal of reading devices, this is hardly possible. Although, according to information from the scientific director of the RGA NTI, G. Z. Zalaev, magnetic tapes with telemetry information were successfully transcribed into DVD-R discs in the archive. In turn, these optical disks, when archived, are copied to the storage system on hard drives (hard drives) organized into RAID arrays.

Magnetic tapes with materials from the general department of the Central Committee of the CPSU are stored in the Russian State Library and have not yet been declassified. While the EU 1066 computers on which they were recorded were disassembled several decades ago. And it is not known which version of the software was used when they were created. Thus, transferring information from them, and actually reading them after such a long period of storage, seems unlikely.

As for floppy disks, which have a guaranteed shelf life of years, information was announced at one of the scientific conferences about the disposal of several million floppy disks by an institute that collected and stored computer programs due to the impossibility of their further use.

The monograph briefly mentions some network information resources (SIRS) that existed in the pre-Internet era. In our opinion, they deserve a separate, at least brief description. These include the so-called teleconferences (tele in this case means remote) FIDO (Fidonet is an international amateur non-profit computer network built using point-to-point technology), the number of which continued to grow after the advent of the Internet: from 30,000 in 1997 to 60,000 in 2010. [my calculations, E. Z.]. At the same time, several hundred of them were Russian-speaking, and several dozen of the last ones were historical subjects. An example of such a thematic conference is "The White Case against the Red one." The encoding of text information in such sources was carried out in the most economical way, 1 byte - 1 character, which made it possible to easily save several thousand messages from several conferences to an archive on the user's computer.

Separately, I would like to focus on computer programs as a possible component of AI. In the monograph, this part is described as "...the texts of computer software ..." [1. p. 131] In this case, for some reason, executable program files remain outside the EII. For Windows operating systems, which the author of the monograph focused on, executable program files mainly have the extension COM and EXE. At the same time, the language of a computer program can be completely different in level, properties, volume, and functionality — from the language of machine circuits (JMS) and the assembler of specific processor models to high-level languages (JAVA) such as Fortran, Algol, C++, Java, etc. Computer programs as an intelligent product have their own GOST standards for descriptions. For example, the algorithm of the program should be disclosed, the types and values of arrays and variables, including control points, should be described. Usually, in practice, the full description of the program is comparable in volume to the program itself or exceeds it. Such complete descriptions were not always carried out. If available, they should also be accepted for storage.

With software as one of the elements of EII, the overall situation is quite specific. For example, back in Soviet times, huge arrays of real-time software were developed for missile defense systems (ABM), missile attack warning (SPRN), and space monitoring (SCCP). Only in the Research Institute 5 of the Main Rocket and Artillery Directorate of the Ministry of Defense, later renamed the Moscow Research Institute of Instrument Automation (MNIIPA), according to the historian of Soviet and Russian software V. V. Lipaev ^[8], more than 2000 programmers worked on the creation of programs for territorial air defense systems (a topic, unfortunately, archaic for the present). Currently, the institute has been disbanded and demolished, its territory has been transferred to residential development. All the software developed in it can be considered lost.

It seems to us that the refusal to delve into the technological features of the creation of AI led to the fact that, unfortunately, the book did not reflect the global trend of universal "mobilization" or "mobile revolution" ^[9]. Namely, the ubiquity and use of mobile devices in society. A smartphone has now become an indispensable attribute of everyone. It provides basic communications within society, interaction with government agencies (Public services), banks, insurance companies, mockups, etc. ^[10] The number of available mobile applications is in the hundreds of thousands. A generation has grown up for whom the smartphone is the first and often the only source of information and communication. How EII will be created in the smartphone era, what features will distinguish them, how they can be identified, preserved and studied (and whether it will be possible at all) is an interesting and promising problem for modern source studies. Meanwhile, for the Russian archives, the task of creating mobile applications was not even set, and the term "mobile" itself is not found in the monograph.

In general, Yumasheva's monograph is of exceptional value not only from the point of view of introducing a new type of historical source into circulation, but also as a truly storehouse of information about various kinds of regulatory documents, GOST standards, and projects implemented both in the Russian Federation and abroad. In the future, I would like to see this monograph revised as a section of the author's website, in which all footnotes to regulatory documents would work in the form of "non-rotten" hyperlinks, according to the terminology of the author of the monograph [1. p. 185], including the possibility of reading the full texts of those sources to which the author refers. As well as access to the websites of these projects/organizations. This would create an extremely capacious and useful electronic resource on the problems of information source studies.

Of course, AIS in the form in which they are described and presented in the monograph have the right to exist. The prospects for the application of this term may be different. It is possible that they will fall out of scientific circulation, as happened with technotronic documents after the death of V. M. Magidov and the disbandment of the relevant faculty at the Russian State University. Or with machine-readable documents, the oblivion of which was facilitated by the development and improvement of information technology. Nevertheless, we must admit that a very important step has been taken in historical source studies in an attempt to make it relevant to the digital era.