N.N. (mouse over)(on the subway tracks)

When I come back home after an evening out, I eventually end up on the platform of the Subway Nb 9 at the Zoo. Between 10 and 11 P.M., I always expect people to be as half asleep as I am, but there usually is so much activity that you might think it is 5 P.M. So I head to the front of the platform and check the tracks for mice. In the summer, there usually are 3 or 4 dancing around. I only saw one yesterday evening, one lonely mouse carefully checking each and every bit of the track. I suppose people just throw away so much food on the tracks that it’s worth looking for it in between trains (3 to 10 minutes). At least it is worth for a mouse to pop up.

Just like our popups appear by mouse over to show the explanation to a hand or an unclear passage (enough intellectual food to check it over).(or almost just like it, for the metaphor’s sake)

While watching the lonely, hungry mouse, I was on my way back from the DH Berlin Stammtisch, a monthly institution consisting in a wise balance between beer and interdisciplinary discussions on diverse digital topics. Although we were only 6 yesterday evening (or maybe because we were only 6), we discussed so many things that I will not even closely be able to render all of it. Instead, I will egoistically concentrate on the aspects that are most relevant for this blog.

Alexander Czmiel (who, together with Markus Schnöpf, attended our presentation of the 15th) asked me very rightly why our edition does not have a decent IP address. And I am extremely grateful for what happened after I had explained that I had had the possibility to do so for over 3 months now and was not able to settle on a name. At that point, everybody tried their best to help me find one. After giving it some thought today, the first suggestion (intellectuals.hu-berlin.de) seems the best one to me. It is certainly not optimal. You don’t expect something called “intellectuals” to deal with the time around 1800, where it was not used. The anachronism is a problem (also the fact that it does not give any relation to Berlin proper). The concept of  “intellectuals” has been my premise for what seems to be forever, but I have started doubting it a year ago. Maybe, after all, there is a better way to analyze as a group scholars, writers both male and female, publishers, in that time around 1800? – a way and a concept I would not know of yet, but that would dawn on me as an epiphany on the last funding day of my project. “intellectuals”: it is only a work hypothesis (one which I was also very grateful to be given the opportunity to explain at length yesterday at the risk of boring everybody), so should it really be an IP?

Although I said the contrary yesterday, I think that the controversial dimension of this work hypothesis and the historical context in which it is usually used (20th century social-historical studies) are the main objection to its use in an IP for our digital edition. I was strongly advised again adding a date yesterday (my beloved “around 1800” would have done so nicely though!) in an IP.

So what was it that I yesterday considered as the main problem and today do not? Mainly the answer to the question “What is this edition bound to become?” And when asked that, I was torn between the pessimistic answer (a series of XML files disseminated in the libraries and archives who host the digitizations of their manuscripts with it) and the optimistic answer (a big portal on literary and scholarly discourses in the late 18th-early 19th century that will be used by scholars, students and just everybody). In the end, this edition can probably become either one of those, but I can only in part influence its future destiny. So  it seems after all wiser to take it as it is now and as it will be in the two years ahead: a digital edition dealing with intellectuals in Berlin around 1800. In that sense, “intellectuals” would be an adequate identity. I might even file for the IP tomorrow.

N.N. did not just concern the digital edition’ missing IP. Part of the conversation also dealt with biographies and their norming. What shall be done with all those historical figures one encounters in each and every corpus? You would be surprise to see how long N.N. happens to have lived, how many books N.N. has written, how many battles N.N. has won. The way from anonymity to identity – a coarse-to-fine of some sort, as my statistician of a husband would put it – should in fact be accompanied by a general synchronization with a database like the GND (or the PDR). Now will we ever know of all the palaces, battles and elephants? Certainly no. But we should be able to sort out who is worth coming out of anonymity. A nice experiment: make scholars of all disciplines define what is a relevant historical person to them…


I will finish writing my mid-term report for the German Research Foundation this afternoon. This and the presentation of last Monday lead me to formulate more clearly our achievements and our goals. The good thing is: I was not overwhelmed and depressed by the to-do list when I saw what the already-done list of the last 2 years contains.

I spent last winter and spring running from a meeting to another, trying to figure out how to organize cooperations with different institutions in order to present an edition that does not pretend to re-invent things that already do exist. I have the feeling that this is finally going to take a real shape especially with the Boeckh corpus. Finding a satisfying presentation of our Boeckh metadata collection combined to the edition of interesting Boeckh documents is the next big thing on the to-do list. I have been looking forward to reaching that point for over a year. I am totally excited about working on that. Plus, it is made easier by the fact that we don’t need anyone to move on: we have the data, we just have to think about how to organize and present them.

There will be further meetings, of course, on the three other works in progress (synchronizing with the Personendatenrepositorium, connecting text and image more finely and comparing our manuscript transcriptions to the first prints hosted by the Deutsches Textarchiv). What will make those easier than it was in the last winter during the first meetings is the fact that I now have a more precise idea of what has to be the profile of our edition. It has to be performant in both quantity an quality. In quantity because we want, in the end, to analyse intellectual networks on a representative scale and need at least a mid-sized corpus to do so (not a small one like it is now the case). This aspect was clear from the beginning. The question that was harder to answer was the one of the quality in detail, especially considering that it can be contradictory with the quantity: if you encode every other word, you need months for a single document and never get to the point where you have the whole convolute encoded.

Still: It seems that the text genetic part of our encoding is not only a scholarly requirement shared by all in the group, where nobody considers it an option not to render a correction, a missing letter, a hole on the page. Our two views of the transcription (diplomatic version and reading version) really is an originality of our edition and should be a strength of it. There is still some work to be done on the definition of text genesis and its implementation on the encoding and the visualization (encoding on the level of a single letter makes it difficult to generate a clean visualization). But these are interesting questions as well and I think we can define a clear frame within this winter for sure.

When we will have more text (hopefully in the spring), we will be able to think about what can be automatized in the encoding. And then we will all relax and encode with the toes, letting the computers do all the work…


That’s what I call Digital Intellectuals!

The Berlin Academy of Sciences just launched a digital edition of the minutes of the work sessions that were held there between 1746 and 1786. You can consult it here: it is awesome. Not just because it looks awesome (great design, intuitive orientation, no information overload), but also because it is an actually well conceived resource – and a totally helpful one for my further work too.

The originals are in French (French became the official language at the Academy under Frederick 2nd and remained it until 1810), but you can find a selection of German translations. You can search either through the minutes (“Protokolle”, chronologically ordered) or through the papers presented at the Academy (“Akademie-Schriften”) or by looking for occurrences of a person (“Personen”). I was thrilled to see that my guy Beausobre (see this post on Beausobre) appears 912 times and am looking forward to cross-reference the occurrences with the periods for which I have letters waiting to be read. I so  like it when you find out that the official version of an information is presented in a radically different way in private writings.

The only thing that I think could significantly improve the edition would be a more precise work on the biographies. At the time being, they are pretty scarce. Hopefully those good people writing minutes and attending sessions at the Academy will at some point be integrated in the database of the Personendatenrepositorium. This is also a project of the Academy of Sciences and it aims at bringing together the biographical data emanating from the different Academy projects. One could argue that the PDR has a focus on 19th century and already a lot to do with just one century. But 1760 is really almost the 19th century, if you consider the major turn Berlin took with Frederick 2nd as the beginning of an intellectual era in Berlin (like I do).

It somehow confirms a kind of aporia I have been discussing with Claudia Sedlarz of the Berliner Klassik last Monday:  even if your aim is to seize network effects or connections, it all comes up in the end to the way in which you will be able to connect biographies. You always need a few “lighthouse” persons in your research horizon, like orientation marks, in order to be able to shed any light at all. And with that, you are either stuck in the canon or constituting your own canon. Either way, as soon as you are moving in the field of biographies, a block of normativity falls on your shoulders. But you can’t properly operate without it. Sound like Sisyphus to me…

CC-BY (bye-bye, sissy!)

No joke today: I plan to publish the link to the edition tomorrow. We could still postpone for a while, but right now, what we mostly need is users’ feedback. Even if I have not prepared (by far) as many letters as I wanted to, the few I have prepared have already allowed me to broaden noticeably the scholarly horizon of the “intellectual networks” that are at the core of the edition. The unexpected boom of the index of persons has made me develop new questions on that topic: How can we explain the hierarchy in the historical recognition of some categories of persons vs. the non-recognition of some others? To what extent does it have to do with politics or with communication strategies, how does it relate to the constitution of a historical and/or literary canon? What would be an appropriate way to represent the position women took in those networks despite the lack of information we have on them, how can we gather such information? Why is it so hard to gain information on publishers, even though their names and the works they published are so well known? I hope that the cooperation with the PDR will help me structure this over the spring and summer.

One thing that had to be decided before going online was the license under which we will publish the work. We finally settled on a non commercial CC-BY. Since I had not really intended at first the “non commercial” part, let me tell you how it came to it.

The way I put things “out there” on the internet feels to me at some extent like recklessness. I don’t particularly trust human nature; I am certainly exposing myself to distortions of my work. I accept it for myself (like I do in this blog): I consider that there is a certain solidity to the contents I publish which is, to my eyes, sufficient to make it intellectually self-sustainable. Even if someone abuses the contents I put online, re-uses them under his or her name, I am confident enough that the right people will know what is actually my work.

It turned out that for the younger people in my team, they see things quite differently. It has to do with the fact that they are less established scholars than I am on the one hand, and with the corpora we edit on the other hand. While I go with obscure correspondences or lecture notes, they happen to be the ones that are responsible for the “big authors” in the edition: they are the ones who publish major literary works after the manuscript, like it has (for some) never been done before. As for me, I only considered it a great chance for them to be the first ones to do so. It did not occur to me that – because these works belong to the canon – it would expose them to plagiarism or denial of the work done. This is how we settled on the non commercial license: even if their work is bound to be subject to criticism (which we will thankfully be able to integrate through updates – this medium is so great, really), at least nobody should be able to earn money with it.

I don’t actually believe it will change much. But if it gives them the feeling that their work is more protected, why not. The problem is, your work is never protected, it is always a risk, an exposure, a hope for a dreamed audience that can just as much turn into a nightmare. And yet: so few things are as enjoyable as putting a message in a bottle. Go figure…



Word of the Day: Workflow

I started noticing the recurrence of the word “worklflow” about a year ago. This is also the time that I had begun to get what metadata are (I come from a galaxy far, far away, as you see). At first, I imagined a nice little boat gently floating  on a river: me and my team along the workflow. I took me a while to accept that the problems of workflow involve thinking in terms of non-linearity and arrhythmia.

In the traditional way my discipline works, the most you have to coordinate is topics for papers when you plan a conference or exchanging corrections between the author and the editor for a book. The idea to work in direct connection with institutions and not just shake hands and put their names as “cooperation partners” completely changed the way I do my job. Half of it is, in fact, another job.

At the time being, our cooperation extends to 5 institutions we are working out a workflow with:

– the easiest workflow to implement so far – and also the only institution we already have a signed Memorandum of Understanding with – was with the Stiftung Stadtmuseum Berlin. We received their digitizations and used them on our edition; anyone wanting to re-use them has to ask for their permission; it is all settled and implemented on the edition page (which I promise you will get to see at the beginning of April)

– the Staatsbibliothek zu Berlin-PK: the simple idea to implement in our digital edition a link to the digitizations hosted in the StaBi’s digital library involves a series of work steps before it comes to actually realizing this linking. We are now working on that specific workflow, which supposes a mutual effort in coordinating the granularity of information we put in the metadata (which both the StaBi and us want to use). Slowly but surely, I am getting acquainted with librarian data formats;

– Some of the texts we edit are major works of German Romanticism, so that the comparison between the manuscript and the first print is a major scholarly issue. In order to realize that, we are working on a workflow with the DTA (Deutsches Textarchiv, at the Berlin-Brandenburgische Akademie der Wissenschaften). The ball is in our camp right now, we have to realize a document that will be comparable to their document. It is not so easy since it does not just concern local amendments, but for instance whole paragraphs missing in one version – and the ending of the text is completely different, too;

– For the letters, I have emphasized often enough how important the index of persons is to me, especially with the aim of reconstructing intellectual networks. We are working out – guess what: yes, a workflow! – with the PDR (Personendatenrepositorium, also Berlin-Brandenburgische Akademie der Wissenschaften) in order to retrieve some information from their database and for them to be able to use some of the specific information we provide through the unedited manuscripts;

– Within the Boeckh project, the cooperation with the University Library is a key element. Like for the StaBi, we have to find a way to mutually enrich our metadata. Unlike with the Staatsbibliothek though, it is not a limited number of folios we have as a common ground but thousands of books to be digitized. The difference of scale involves – well, a different kind of workflow. I am a little bit overwhelmed by the figures on that part of the project, I must admit. But we still have some time to figure it all out properly.

I hope to come to concrete cooperations like those also with the Archive of the Berlin-Brandenburgische Akademie der Wissenschaften and the University of Chicago Library / Special Collections. And hopefully others too. At some point.

One thing that I can say for sure is that I am very happy and proud of the way we use XML. The fact that we, the editorial team, the people who transcribe, comment, write papers on the literary value of those texts, encode them directly in XML/TEI makes it easier to put up all those cooperations. In fact, it makes it simply possible to work out direct workflows. That is one decision I will never thank Laurent enough for.

My biggest worry is still, of course, the long time archiving. On that point too, I think working in XML/TEI will be a major strength of the project. Still, I get more and more the question from the cooperation partners: “and how long does the project funding still last?”. June 2015 is coming closer, especially considering the time it will take to implement those workflows. But never would I have dreamed to be able to go that far in these 5 years, and I keep my fingers crossed that I will be given to pursue this work after this funding phase.