Will it last? How secure is the longevity of archaeological data?
Download as PPT, PDF0 likes107 views
The document discusses the longevity and security issues surrounding archaeological data, particularly focusing on the challenges of preserving small data systems. It highlights various databases that have faced obsolescence or lack of support, and contemplates potential hosting solutions that would allow for enduring access to data akin to traditional materials like vellum. The author seeks recommendations for free or sustainable hosting options for cave archaeology records and substantial microCT scans of mummies.
Will it last? How secure is the longevity of archaeological data?
- 1. Will it last? How secure is the longevity of archaeological data? Ahmad Alam Supervisor: Professor Andy Brass Co-Supervisor: Professor Robert Stevens Bio-Health Informatics Group
- 2. Will it last? • Major recurrent issue • Competition : Vellum, life span, 1000+ years. • Loss: Several near misses – some systems ‘stillborn’ • Information systems ‘decay’ in progress • Totally unlike ‘big data’ – this is ‘small data’ • Longevity is key
- 3. Hanging by a thread • Almost lost once • No paper published version • Currently held as web pages by a volunteer • No suitable ‘free’ host • Hence a good ‘use case’
- 4. Skeletal db: Dying – slowly • 100K + detailed skeletal remains • Written Delphi (obsolete language) • No source code • No data dictionary • Sybase Back End (a SAP minor product) • Dated interface, on the verge of breaking
- 5. IADB : Contemporary, but orphaned • Javascript – My SQL • Comprehensive system • No wide scale adoption • Several users now off-line • Can data be trusted with it, when there’s no support?
- 6. Manchester Mummy Tissue database : stillborn • PHP - MySQL • Caught up in deployment ‘issues’ • Original team of authors ‘gone’ • Sat in an inbox for a year (!) • What is its long term future?
- 7. Linked Open Data • Evaluated • More effort than relations DB with web framework • Geared to towards electronic data • Doesn’t help classical information
- 8. Ask the Question! I have an interesting challenge, and just want to hear your two pennies' worth. Put simply, where can one host a few dozen records (cave archaeology) in perpetuity, i.e. where it would last the longest without being deleted, or becoming inaccessible (e.g. dBase, 8" floppies disk etc.), for free? 'Longest' in this case means competing with vellum, calf skin, proven life span of at least 1000 years, in the news recently due to a review of costs as Acts of Parliament are recorded on it. If you're still interested, the bonus question is, where could microCT scans of mummies (70 GB each) be similarly hosted, this for the least amount possible, free would be unfeasible for this volume of data! Vellum is out of the question, that would be a lot of calves.
- 9. Hang on, 70 GB Mummy Records?! • ADS York : £120,000 – A very big number! • Zenodo CERN : data only, no manipulation, volume an issue • Morphosource : Not EU based, breaches funding condition • UK Data Archive : Said No, as wrong type of data
- 10. Challenges • Are huge and diverse • … and largely down to funding • Preserve the artefact, the data, both? • Is the rush to the ‘Cloud’ safe, as services come and go? • But small steps can help • Why not start with data capture at the earliest stage to help with long term ‘longevity’?
- 11. Using spreadsheet programs for scientific data • Good data organization is the foundation of any research project • Most researchers have data or do data entry in spreadsheets • Spreadsheet programs are very useful graphical interfaces for designing data tables and handling very basic data quality control functions
- 12. • Good data entry practices - formatting data tables in spreadsheets • How to avoid common formatting mistakes • Dates as data - beware! • Basic quality control and data manipulation in spreadsheets • Exporting data from spreadsheets • Overall good data practices • Much of a researcher’s will be spent in the 'data wrangling' stage. It's not the most fun, but it's necessary. This can help teach how to think about data organization and some practices for more effective data
- 13. Resources • http://www.datacarpentry.org/spreadsheet-ecology- lesson/00-intro.html • www.ahmadalam.net • http://www.presentpasts.info/articles/10.5334/pp.58/