This article aims to lay to rest some of the gripes that many people have with the Portable Antiquities Scheme database. Since I joined the Scheme in April 2003 from an Investment Bank, I have been responsible for the development of the Scheme’s IT infrastructure. Compared to the budget of the projects on which I previously worked, we’d spend the same amount as we’ve spent on the Scheme’s IT on just 60 personal computers (a trading floor has thousands!) The majority of my work is based on web technology and in this I am entirely self-taught from the web and various books. I currently build and maintain the Scheme’s main website (www.finds.org.uk), the children’s website (www.pastexplorers.org.uk), the All Party Parliamentary Archaeology Group website (www.appag.org.uk) and several others that are numismatic based. The other aspect of my job involves project managing the development of our much-criticised database (www.findsdatabase.org.uk), which is maintained by Oxford ArchDigital.
The database has now been online since the 1st week of April 2003 (indeed it was formulated before I was appointed, and therefore a legacy was left for me to deal with.) It is designed to:
• Allow for centralised collection of object and numismatic data
• Allow public access to these data
• Allow for transfer of these data to the relevant Historic Environment Records
• Store high resolution images
• Manipulate spatial data to allow for study
• Provide a lasting record of highly mobile artefacts – quite often it is the one and only chance the Scheme has to record them.
How much did this facility cost and is it cost effective?
Cost per unit added to the database, when plotted against total IT spending demonstrates a fall with every object recorded. Figure 2 below demonstrates the annual IT outlay by the Scheme between 2003 and 2006 and the cost per find recorded.
||Number of finds recorded
||Running finds total
per find recorded
|2003 – 2004
|2004 – 2005
|2005 – 2006
|2006 – 2007
The falling cost for recording items on the Scheme’s database.
The usual cost efficiency argument that is discussed in many quarters, factors in the costs added for recording each object by using staff salary and associated costs divided by the number of finds recorded. I, for one, believe that this is a false measure of the Scheme’s workings. How does one attach a cost element to the extra work that the average FLO carries out? For example, the local society talks, finds days, school visits, gallery talks and local Museum or authority duties. There is never 100% of the FLO’s time spent on pure data entry; nor should there be. Therefore, I believe that the above figure demonstrates that the Scheme is performing a very cost effective recording process. Are there comparable figures for the rest of the Heritage sector? Not that I can find!
The database is my biggest headache in my working life, and I get various complaints regarding the workings of the interfaces. Below is a sample of the usual comments that I have to contend with:
I, and the Scheme itself, recognise that the user interface is not especially user friendly. However, over the coming weeks, steps have been made to rectify this and a new search facility should be unveiled shortly. This is not an easy thing for us to implement. The annual budget for IT improvements is miniscule. With an extra £40,000 a year, we could have an interface as slick as any on the Internet. However, the new improvements should go some way towards making your experiences better.
Before I discuss the tips for using the database, I think it is best to detail what improvements have come online since the Scheme launched its online presence.
1. High-resolution image viewer (this allows you to zoom and pan around the image and view details.)
2. Expanded record details – more on this in the tips.
3. Workflow indicators – as a non-recording user, you see 2 traffic lights (yellow for a unchecked record, green for a checked record.) A finds adviser has checked not every record you see with a green light. However these do have a warning message on them about their quality.
4. Finds of note list – this makes our staff decide whether finds should be drawn to your attention.
5. Larger thumbnail on records.
6. Redesign of aesthetics (I would still like it to look nicer.)
7. Simplification of periods used (removed late and early and now just use broad terms such as Roman.)
8. Implemented a thesaurus to aid searching.
9. Introduced RSS feeds to allow you to reuse or be alerted to our latest records in pre-defined searches.
10. Introduced self-registration (this is extremely important, more later in this article.)
11. Allowed HER officers to download records directly and use them for Development Control.
12. Introduced Reece recording periods and enhanced numismatic recording facilities to lessen mistakes made on data entry.
What’s still to come?
In the next few weeks, we expect the following to be put online by Oxford Arch Digital (OAD):
1. New search engine (for further details visit our blog and read this article – http://www.finds.org.uk/wordpress/index.php/274
2. New mapping client – based around either Google maps or the open layers client.
3. Enhanced visibility of RSS feeds.
As and when we can access more funds, the system will be migrated to the newer and far more intuitive (it is believed) second-generation system that OAD have created.
But I still can’t use it!
I hear this from various users, including our staff (Michael Lewis especially!) and it is something that needs some explaining for many. To begin with, I believe the best thing for you to do is register with the database. This opens up some of the more useful features to you. To do this, visit:
Fill in your details, and then by return email, you will be activated. This email expires after 24 hours, so if it doesn’t arrive, check your spam box. Once activated, you can now access the following:
1. Save your searches (either for reuse on the system each time you visit, or as an RSS feed that you can use as an alert system – this needs some special feed reading software, or for reuse in your own website – see Corinne Mills’ excellent website http://www.ourpasthistory.com)
2. Change preferences, such as viewing thumbnails in search lists, changing the number of thumbnails that you can view, changing the number of records per page.
3. Enable tracking of your experience. I can see what people want to search upon and therefore identify problems. You cannot be identified by name as it just logs searches by auto-registered groups.
Once registered, you might need some help searching the database to get the best out of it.
There are 3 main areas for searching the database:
• Quick search – via the box in the header or the single search box from the search menu.
• Advanced search – most popular across all user groups and requires a modicum of effort to construct a search.
• Image search – perhaps most useful for quick searching for visual recognition.
The fourth type of search – “visual” – is to become obsolete soon as so few people use it. The quick search box in the header is limited on the fields that can be searched within the database. These fields are:
• Object type – for example coin, brooch, axe
• Period – for example Roman, Early Medieval
• Ascribed culture – this is only used for Early Medieval objects
If you use the quick search menu option, check the radio button for which type of object you want to search for. For example if you check coins, then you could enter the ruler’s name and get back all coins for Trajan.
How do I make an advanced search actually work?
Well this is a frequently asked question, and quite simple to answer. With the current interface you need to do the following as shown in the examples below. Before you begin, determine what you would like to retrieve. One caveat that has to be remembered; we don’t pluralise objects types. We record everything in the singular; therefore coins are incorrect – coin is correct. Brooches are incorrect; brooch is correct.
Let’s do a search for Roman coins from Lincolnshire recorded by Adam Daubney our Midlands rock star.
Step 1: choose object type is exactly coin and then press the more button.
Choose AND as the operator term.
Step 2: choose county is exactly Lincolnshire and then press the more button.
Choose AND as the operator term.
Step 3: choose broadperiod is exactly ROMAN and then press the more button.
Choose AND as the operator term.
Step 4: choose created by is exactly adaubney.
Now press search.
This will allow you to find all records that meet these search terms. If you have registered and logged in, you could save this search and access this from the saved search menu every time you login to the site. Another hint to save you getting page expired message is this: right click on a link and open in a new window.
When the new advanced search facility launches, the above steps should be rendered obsolete, as the steps will be far more obvious.
How do I search the image bank?
This is very simple and requires very little input to gain lots of output. However, it is hampered by the labelling terms entered by our recorders. To find coins of Trajan, enter Trajan in the filter box and press filter as shown below. You can then scroll through pages of anything attributed to Trajan. More often than not, you’ll get mostly coins back for that particular search.
How do I use RSS feeds?
These are perhaps the most powerful feature of the database and also the most underused feature. RSS stands for Really Simple Syndication, and it is a way for one to share content easily. You need to download some software (see http://www.finds.org.uk/news/rss.php for more information) and then create and save a search as an RSS feed. I currently use some software called “Feedreader”, simple and easy to install from http://www.feedreader.com.
Once you save the search as an RSS feed you can paste the address you are presented with into your feed reader. Every time a new find that meets your search criteria is added to the database, the software alerts you to its existence. Lazy man’s surfing! If you don’t fancy creating your own feeds, try some of the ones that I have already created. You can also get RSS feeds from a whole range of other sites, for example the BBC, Google’s news pages and many more.
The other advantage of RSS, is that these feeds can be reused in your own website. If you want to do this, email me at the British Museum for more details. It could invigorate your web site with finds for your parish, county or even your own finds!
How are these data used?
These data that we collect are used all over the globe by archaeologists and the layman. Example projects include:
1. English Heritage funded Night-hawking study conducted by Oxford Archaeology
2. VASLE based at the University of York (AHRC)
3. The Tribal Hidage project at University College London (Leverhulme)
4. AHRC funding for 3 PhDs in 2006-2007
5. Over 50 individuals completing research for higher level degrees worldwide (Japan, Australia, USA and of course the United Kingdom)
As more people start to study these data, our understanding of the archaeology of England and Wales will be radically altered. The set of maps below demonstrate how quickly the Scheme has started to collect data all over the area that it covers. In sequence, these maps demonstrate all finds from 2003, 2003 – 2004, 2003 – 2005 and 2003 – 2006.
However, these data are useless when studied in isolation. Therefore it is imperative that these PAS generated data are compared with other institutions data. The next two maps show PAS data compared with data from Oxford University’s Celtic Coin Index and then PAS Roman coin finds from Peter Guest’s recently published database of Roman coin finds in Wales against the PAS generated Roman coins data. This highlights discrepancies in recording areas and allows archaeologists to try and identify biases and trends. The analysis of these data and HER data will be fundamental in writing the early archaeological texts of the 21st Century.
Can I get more help?
Yes of course! Speak to your FLO or call the Central Unit on 0207 323 8611 or email us at email@example.com Don’t forget to register! You’ll get more from the database!