All About That (Data)Base

Our last post was a recap of the past year, highlighting all our progress and successes. It was a great way to end 2014, and encouraged us to look forward to everything we plan to accomplish in 2015 starting with our presentation at the AIA/SCS Annual Meeting this week in New Orleans.

Our AIA poster!

Our AIA poster!

If you’ve checked out our squeeze collection online lately, you may have noticed that the Athenian Tribute List squeezes are still not up. We really wanted to have it ready before the AIA in January. We always knew the ATL’s were the focus of McGregor’s research and the jewel of our collection; they were the squeezes I started with back in May. All of them were photographed and edited by July. So why aren’t they up and available? Database issues. Blasted metadata and the interminable complications of uploading content. We all know there is always a little (or a large) discrepancy between the presentation of a project and how it really came together. I’d hate to give the impression that we just gleefully skip from one success to another, that every aspect of this project has fallen into place without struggle; not only is it a misrepresentation of the hard work we’ve put it, it goes against our philosophy of openness and transparency. So in the interest of transparency this post is going to look at some of our setbacks, mistakes and the things that are driving us crazy behind the scenes.

squeeze section (1)

So sharp

Lets start with the biggest confession of all: we kind of hate the squeezes.

It’s a love-hate relationship at best; for some its a meh-hate relationship. Speaking only for myself, I love what they represent. I find myself constantly thrilled at the words painstakingly carved by hand with such precision by someone long dead whose work remains, and remains relevant, thousands of years after they’ve died. It’s very easy to drift into romanticizing daydreams about the people behind the inscriptions – the workers carving them, the builders constructing the monuments, the festivals held outside those temple walls…and yet that romantic notion dies the minute you dig into the pile of dusty old squeezes and try to make sense of them.


Not even following our best practices guide.

Not even following our best practices guide.

We have a cabinet housing the squeeze collection, and I’m sure at one point in time it was organized but the occasional mis-filing over the years and my summer spent hauling them all out for photography has played havoc with any order that was ever imposed on them. What this means is that the master list of what squeezes the department had in the late 70’s doesn’t quite match the files of photographed squeezes we have stored on Digital Initiatives’ servers. Heather and I tried valiantly one day in December to sort through the squeezes to reconcile the master list with the actual squeezes so we could see which ones I missed and get those photographed. We underestimated just how tedious the task would be and only managed to get about a third done before we gave up.

It’s not often you read that phrase in any academic writing, even in something as casual as a project blog, but this is a post about transparency.

Lesson 1: sometimes you undertake a task that  defeats you.




Moving forward, I think we’re going to work on a reward based system, no longer kidding ourselves pretending the work is its own reward. We’ve had a month off. With a lot of coffee and the promise of beer afterward we will eventually, reluctantly, return to the task.

I don’t want to give you the idea that we are just spectacularly lazy. Digging through the squeeze drawer is one small task that can be put off to a later date without seriously holding up the online publication of our collection. We’ve been wrestling for months creating the metadata spreadsheet. It has to communicate the various epigraphical identifiers used for squeezes while respecting the necessary categories of information for the Digital Initiatives content management system. I’ve lost count of how many iterations that spreadsheet has gone through. It is not a case of simply entering the data into the correct column, because all too often the information is missing, ambiguous or requires a good bit of research behind the scenes.

Most of our squeeze collection is numbered using the Inscriptiones Graecae, though some have an EM (Epigraphic Museum) number or a BM (British Museum) number. The IG was on edition 2 (IG I2) when McGregor was making the squeezes and is now on the third edition, IG I3, so his handwritten notes are now outdated. Maude spent much of May and June matching each List number to the appropriate IG I2 and I3 numbers so our database would be accurate and up-to-date, but still retain the old IG I2 numbers for people using older articles in research.

However. Timing of the project meant that while Maude was matching up these numbers, I was already photographing the squeezes and it was simplest to name the files according to McGregor’s notation on the squeeze. (Oops). This means that now, when Chelsea wants to upload an image of a squeeze, she gets to do the IG I2 to IG I3 reconciliation that I neglected all summer. (Sorry!) I can only somewhat justify this by claiming I was trying not to tie up the one TTI scanner in the Digital Initiatives lab all summer by working as quickly as possible, but that’s not really an excuse when I could have easily renamed the files when I was editing them in Photoshop.

Lesson 2: cutting corners never saves time.

By far the most frustrating data entry has fallen to Heather. The ATL’s are catalogued by an IG number and a List number from 1 to 40, though often McGregor just wrote the List number on the physical squeeze., so Heather has spent hours at a computer surrounded by the Athenian Tribute List volumes, matching up an image of a squeeze fragment with the correct transliteration of the fragments, making sure all our file names match the official designations for the ATL’s. This tedious task is made all the harder by the fact that the inscriptions are carved in all caps, but epigraphic convention transliterates them in lowercase letters. Matching games are just that much harder when the letters don’t actually match, never mind that the fragments are, well, fragmentary.

ATL transcription from Studies in Fifth-Century Attic Epigraphy

ATL transcription from Studies in Fifth-Century Attic Epigraphy

On top of that, the ATL’s mostly list the demonyms which generally end in the same letters, or numbers representing the tribute paid. Think of it like cash register receipts that have to be painstakingly corroborated. Thankfully, Heather has almost entirely completed the ATL reconciliation and we are reasonably sure we’ve now properly identified all the squeezes and have the correct attributes in the file names.

You’re probably assuming, like we did, that reconciling the file names would be the last step to uploading those ATL’s, but no. Some of them are fine, yes. Some could be uploaded easily. But some of the squeezes contain multiple fragments from different lists. What we have here is a failure to communicate; the metadata spreadsheet doesn’t like the multiple lists, and the content management system doesn’t like confused metadata, and it was coming up to Christmas and we all agreed to just leave this hurdle for the new year.

Lesson 3: Sometimes you just need to take a break. 

I can hear some  of you thinking ‘just how lazy are these people, complaining about organizing paper artifacts and cross-referencing between a published volume and the artifact? Try squatting in a muddy trench in Northern England in the rain, digging out artifacts with freezing fingers. Try underwater archaeology, where every single night your trench will be backfilled with the sand you removed during the day.’ I hear you. Sitting at a computer desk is hardly the most tedious task in archaeological research. That said, wrestling with an excel spreadsheet alone at a computer for hours on end is hardly as exciting as excavating on location with your dig friends all summer. I saw your instagram photos of the blue Aegean, bright sunshine, bottles of wine. There was none of that in the basement of I. K. Barber this summer.

We haven’t even mentioned the Nemean squeezes we don’t have catalogue numbers for, or the 30+ squeezes that have no notation whatsoever on them. Trying to figure out how to deal with all the squeezes we have that don’t fit neatly into the necessary metadata categories is the last hurdle. We’ll get there, and we hope its soon, but keep in mind that our combined technological prowess is about what you’d expect from a motley crew of archaeologists, philologists and historians. For now, even though the DI website may not show it, we are hard at work trying to get the squeezes online and trying our best to make sure that they are catalogued in a reasonably user-friendly way. It helps that we have only the vaguest notion of epigraphy ourselves, so we ask all kinds of stupid questions while we’re working through the information and our final product will be as accessible as we can possibly make it.

About Lisa Tweten

As one of the project mangers, Lisa is the heart of the project. She works with Digital Humanities to photograph the MacGregor squeeze collection.


  1. Pingback: Spot the Classicist – Bikers, the AIA and Beignets in the Big Easy | From Stone to Screen

  2. Pingback: FSTS à la Maude: Finding that Work-Life Balance as a Grad Student - From Stone to Screen

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.