Project-Idea Henrik

This is the corresponding issue for the Wiki-page regarding my project-ideas. Feel free to comment your thoughts!

Fine-tune what the historical problems are. Until April. 1. Good History => 2. How can digital methods enable us?

(Text copied here so you don't have to switch back and forth)

The following text is a snapshot of what my current (maybe a little chaotic) thoughts about a possible project are and should be treated as such. If you have any ideas/papers/suggestions feel free to comment in the corresponding issue!

My current thoughts revolve around the idea that the discussion and usage of objects as sources in History are lacking in some regards. At the moment the selection of sources is largely focused on texts. Increasing the usage of objects in historical works could lead to insights into areas (see for example @martin.dusinberre's work), which text-based sources might only be able to touch to a limited degree.

The potential project is trying to facilitate the usage of objects as historical sources. A major point in using objects is quite simply their physical accessibility, which is not always (or even rarely depending on the topic and the location of the researcher) given. In my own experience you develop a whole different relationship with an objects if you have the chance to see it in person, feel the weight, the materials, the clash of your visual expectations vs. the actual contact with the object. I would consider being able to have physical contact with an object the optimum if a Historian plans to use it as a source. Unfortunately this isn't always possible. This might be one reason, why objects have been neglected as sources. I don't think this is a good solution.

Thanks to the my job at the museum I've gotten some insight into how data about objects is stored and specifically which data is stored. In my opinion especially the which data can be improved to facilitate the access of historians to the objects. At the moment most objects (at least in the museum I work in, but I'll assume, that it's similar for other institutions) only receive a very limited description based on size (for storing), damage (for possible restoration), age, provenance (from where the object came into the collection) or place of finding and a rough description of materials used. This is partly because the sheer amount of objects is overwhelming. The medium sized museum I work at currently has alone around 70'000 objects stored. It's simply not possible to cover every object in detail. But I think you could improve it.

For one the kind of information, which is being stored, could be changed to cater to the needs of historians. Besides the current information, which is mostly for archival purposes, specific information could be added for historical research. For example the production location, the exact materials processed and their origin, the object's journey before it entered the collection, etc. could help historians using an object as a starting point for their research or placing it in an existing work. One way to achieve this could be to utilize public science and tap into the knowledge and time of interested individuals. So constructing your data bank like Wikipedia, where anybody can make changes to existing data sets and a second, educated party reviews these changes. The project for the Übung would then be creating a website with a number of objects, on which people can make suggestion on information to add. This will be controlled/researched by second party (me) and then either merged or discarded.

Another idea is to add digital methods to gain information about objects. Maybe if there is something like a image recognition software, which is specialized on materials, one could start taking close-up pictures of materials, which the software compares with a data bank and makes estimates on, what the different materials of the objects are. This is going in the direction @daniel.mcdonald suggested in #6 (closed) but with a more archival and historical focus. The question is, if this is actually achievable in a few weeks. The project then would be take some objects, take pictures of materials, then use the software to add tags to the data set, enriching the information we have about the object.

I'm interested in how historians select the sources they use and how one could expand the diversity of these sources.

added discussion point label

assigned to @henrik.jochum

changed the description

Great write up, all in all. I'm going to think about it through the week in more depth, but definitely, you're onto something regarding

a) the different kinds of data/metadata that could be stored alongside museum pieces etc

b) the digital tools that could facilitate this

A

Rather than just provenance, construction materials and so on, there is definitely a lot more stuff that could be interesting. For example there is always a subset of people at museums who prefer reading the descriptions to consuming the works---I believe I am one of them. I also love reading guestbooks and the like. I really loved visiting the War Childhood Museum in Sarajevo, which consists mostly of various artifacts, old toys, candy bar wrappers, love letters, clothes etc., accompanied by stories from the author about what these objects meant to them during the Bosnian/Yugoslavian Wars. At the end of the museum, people could draw and write stuff on the walls, kind of 'having their say', though it was mostly just kids drawings.

YouTube comments are garbage, but occasionally for (older) songs/music videos, you'll see comments about what the song means to someone, how it played during a special moment or whatever. These can be quite moving, even if they are fake, and even if you don't know the person at all. Adds something to the enjoyment of the art.

B

In the context of digital history, the main logic of git is enough to accomplish this (though users likely would need a simplified interface and set of terminology). Git itself is fast, scalable, stable and so on, and there are I think some open source frontends like GitLab but less confusing. SOmething UZH has spoken about already is using such systems for dataset storage and versioning, e.g. in the linguistics faculty.

Final points

I agree your topic could quickly get too large and you won't be able to complete on time. But you could sketch out a roadmap, complete some features by due date, and continue on with it afterward as a hobby project if you liked. Can discuss.

@henrik.jochum Thanks for these initial thoughts. I agree with Danny on (b): my immediate thought was that this is what git could enable, although the challenge would be to make an interface attractive and simple enough for citizen scientists to use. (Citizen science, incidentally, is one of UZH's key slogans at the moment; see the new Competence Centre website.)

I also agree that much of the information on objects presented digitally is insufficient; and even if the information is there (the object is 27cm high, for example), the user doesn't really feel the information. For a good counter example, listen to a few of the radio programmes (now available as BBC podcasts) that eventually became the bestselling book, A History of the World in 100 Objects. Neil MacGregor is brilliant in describing how an object feels in the hand, its texture, its weight, what its shape might remind you of. If you could somehow incorporate this sensual element into your object descriptions, you would be doing a great service.

added project label

Thanks for the input. I think I will look into ways how one could use Git to achieve this. Also the Middleware article @martin.dusinberre linked mentions something called Omeka, which is a content-management system for digital collections. If there is the possibility to suggest changes to existing data-sets as an online user, I could use it as an established interface and focus on how to construct the information/databank.

EDIT: In the article several other similar systems are mentioned, and the limitations of Omeka are being described (see §35). So maybe Drupal might be a better solution, but I'll definitely look at both.

I'll give the podcasts a listen and maybe purchase the book. Finding a productive way to describe the sensual element in a digital way probably is the key challenge in this project.

So long as the licenses say so, you're usually free to modify and rerelease other people's open source code, even sometimes for commercial purposes.

For example, the code for GitLab itself is here on gitlab. You can fork this, change some button names, and bam, you've got a version-controlling database simplified for the humanities. I mean, don't do this, but the principle is there. If you can find an open source project you like, but which lacks documentation, or could in some other way use a little love, you can go ahead and do dhat, without knowing how to code even. I'm sure we could also develop an 'April project' in such a way (i.e. find a resource you love and contribute to it somehow, make a site/markdown file that helps new users get started with the tool, etc). I'm sure the devs of many small open source projects would happily argree to mrege good documentation into their project

@henrik.jochum And to follow up on what I said yesterday about Citizen Science: perhaps the last two talks in this "Brown Bag Lunch" series will be of interest? And someday, someone must explain this term "Brown Bag Lunch", which I never heard until I came to Switzerland. Any linguists among us, @daniel.mcdonald ?!

I'll probably dedicate some time next week to looking into, which system might be best to structure a project around. But there still is quite some time until April. I'll keep an eye on the talks about Citizen Science, they might be useful.

@martin.dusinberre There is a Wikipedia page for the concept of Brownbag. Interestingly enough there is not an English one and the events seem to be held primarily in the German language area.

It's a talk plus a medicore free lunch. All the theory of pedagogy stuff be damned, let's call a spade a spade and say it's a talk where you also get a sandwich and a bottle of water.

added Mutual help label and removed discussion point label

changed the description

I uploaded a passage of Don Norman's Design of Everyday Things. It's a text that was originally conceived as a book for designers of physical objects, but it has been expanded to include digital objects as well. This is really interesting, because it implies that physical and digital objects can be described and designed with the same vocabulary.

Since you're interested in objects, this might be interesting for you, as it develops a precise toolset for describing and analyzing an object.

changed due date to March 09, 2020

mentioned in issue #26