Proszę o wsparcie dla Krzysia Bulczaka, największego bohatera jakiego znam.

Web archiving for memory studies

In 2009, a former employee of the Grodzka Gate – NN Theatre Centre in Lublin named Piotr Brożek published a Facebook profile of a Jewish boy: Henio Żytomirski, born in 1933 and murdered around November 1942 in the German death camp of Majdanek. Brożek has been publishing posts containing fabricated quotes of Henio speaking to his parents, reflecting on his days or writing letters to his family. The readers were participating in this project as though it was theatre, emotionally engaging in communication with a virtual actor. But was it only a virtual experience?

In 2010 I have published a short text about that project, presenting it as a case study for the challenges which face studying memory in the digital sphere. Alongside the theoretical background, the idea of postmemory by Marianne Hirsch and some perspectives from the theory of photography and digital photography, my article contained a few screenshots from Henio’s Facebook profile. They were published in the printed version of the article, but in the digital version they have been removed due to copyright-related issues.

Now, after almost a decade, only a few traces of that project are available online and my article has become a unique source of screenshots showing users commenting on Henio’s profile.

How we can study digital memory without born-digital sources?

In our work at the Digital Humanities Laboratory at the University of Warsaw, among other projects, we aim to support a serious discussion about the importance of archiving the Web and social media content for the purposes of research, citizen activities and national institutions. It is of course also an important issue for memory studies.

First of all, can historical digital resources, published a decade ago or even earlier, be accessed today? Our research projects shouldn’t be based only on relatively new content published and still available on the Web. The Web is a continuous and constant medium and we cannot just cut its social impact only to the latest few years just because we have no older sources available. This is a serious methodological issue.

Secondly, if we want to study grassroots initiatives and phenomena in the field of memory, we must remember that resources which arise within such a unstable and unofficial social activity are usually inconstant and easily removable. There is no institutional support which could guarantee their preservation over the years; they can disappear very quickly without any signal or any reason. The entire domain or hosting account can expire and be removed; the software used to publish content can become obsolete. But if we also should search for specific URLs, a particular discussion board comment, blog post or Wikipedia entry, they too can be radically changed or simply deleted.

Thirdly, today, social media are the most promising of all resources for memory studies, due to their popularity and the ease and directness of communication which can be observed there. Twitter, Facebook, Instagram can be seen as mirrors of the social relationship to history and as a field of memory games. But the truth is that every social media site is also a “walled garden” – a commercial system in which data and resources can’t be freely accessed. It is not easy to collect and preserve data from these sites not only because of the limitations of APIs (programming interfaces), but also because of each site’s terms of use.

For example, if I wanted to prepare a collection of tweets relating to death camp museums located in Poland, I would have to:

  • prepare an open list of hashtags and users whose tweets are to be accessed;
  • search for tweets’ IDs by using API queries of particular hashtags and users – where access to tweets beyond a certain historical point would be impossible and where the results would likely be incomplete – in Twitter we just can’t get ALL tweets with a specified hashtag from all the time;
  • using the tweets’ IDs, collect their metadata containing text, user information, date of publishing, geolocation, etc., available within a JSON format.

And since Twitter’s terms of use prohibit republishing the full metadata of tweets, if I then wanted to make the resulting collection open for other researchers, I could only publish only a list of IDs, so they can be used in future queries with API. Despite the fact that Twitter is an important social sphere and a “source of sources” in research, we can’t just archive it or share it with others. There is no way to legally ensure that social data is safe and independent from the commercial enterprise.

Going further, the Twitter queries should be run on a regular basis to preserve tweets also for the purposes of future research – our archive or database should always be open to collect new sources, as they frequently disappear or change. The organisational, financial and legal costs of holding such a dynamic and large collection are high.

There are, of course, tools to do this, but it can’t be done as a six-month project. Memory research on social media needs permanent preservation, even without knowing the questions which may be asked in future research projects.

Such archival activity should also be undertaken to preserve the collections of classic Web pages. The Internet Archive foundation shows that this can be done also within crowdsourcing methods, with the help of an ordinary internet users, as it is very easy to suggest a URL to be preserved in the Wayback Machine. But with this method comes a serious question about the shape and neutrality of the collection.

Post-digital perspective

The problem is not only with digital sources, but also with using and interpreting them. To what extent are they really ‘digital’ and not “real”? Were the comments on Henio’s profile real, digital, virtual or live? We have to stop thinking in a way which splits reality into the online and offline spheres and stop using the term ‘digital’ as a common metaphor. Florian Cramer suggested even the term ‘postdigital’ to undermine this division:

So, according to Cramer and his concept of the post-digitality:

  • the term digital today is only an aesthetic buzzword and is not suitable as a useful concept to use in research or media criticism;
  • post-digital describes society after the digitisation era, when each of our activities is mediated through digital communication, and digital media and tools are a natural, even transparent element of everyday life
  • the simplest definition of post-digital describes a media aesthetics which opposes digital high-tech and high-fidelity cleanness, so no online/offline — just hybrid reality, and no new media — just hybrid media, like a screen from the YouTube printed in a paper magazine.
  • software and Web interfaces have their own meaning (for example. the Facebook friend shaping the general social idea of what it means to be a friend)
  • software and Web algorithms have a social agency of their own (e.g. Power Point which has formatted the didactics).

For memory studies, the post-digital perspective can be implemented in:

  • using old media sources to document Web activities (for example, reading press reviews of the first museum webpages built at the beginning of the second half of the 1990s);
  • using oral history methods to research grassroots virtual communities in the absence of born-digital sources;
  • studying the history of Web interfaces in the context of statements of memory (e.g., the use of Facebook’s like button in reaction to posts published by the Museum of Auschwitz)
  • transforming the research language in the field of memory studies.

Przeczytaj także:

Udostępnij na Twitterze | Udostępnij na Facebooku