A rainy weekend looked just perfect for doing some maintenance work for my website: making backups, upgrading below the surface, reading and using a new handbook on the software involved. It turned out to be the perfect time for something you dreadfully fear: it did not work anymore. I will not inflict on you all the painful details, but yes, I had to put a notice on my site apologizing for any inconvenience caused by my virtual absence.
Almost two days after the first signs of a breakdown of the site installation the good news is that the backup of the information is safe and unharmed. And I have learned and learned again some lessons: upgrading equals releasing a new version; documenting of the exact structure below the surface is very useful; making a site depending very much on one central piece of software is nothing but creating the weakest link in a chain…
Of course I have often been warned that these things can go wrong, but this year I had taken new precautions. Of course I had met just a few days ago somebody who told me about the trouble with his e-mailbox. And best of all a few weeks ago I had read again by chance about a website with historical data which have triumphantly survived all major changes in computing since the original data were collected. The data for the Florentine Catasto of 1427 were collected and studied in the sixties and seventies by the late David Herlihy (1930-1991) and Christiane Klapisch-Zuber. Herlihy and Klapisch-Zuber published in 1978 their study Les Toscans et leurs familles: Un étude du catasto Florentin de 1427. Robert Litchfield and Anthony Molho have taken responsibility for the version since 1995 online at Brown University. This university hosts also an online database with records of the tratte, the elections of the Florentine office holders from 1282 to 1532. More technical details about the transformations of the raw data can be found on the website of the University of Wisconsin with the complete data set of the 1427 Catasto, supplemented with ecclesiastical records from Florence. The harvest of online sources for the history of Florence and Italy is rich and varied, with for example Florentine charters and the Fondo Datini at Prato. Records on medieval taxation can be searched online for other countries, too, for instance the Taxatio Ecclesiastica Angliae et Walliae from 1291, a document on church properties in England and Wales. The Netherlands Economic History Archive in Amsterdam presents a fine overview of historical datasets. Do I need to remind you that these datasets yield their secrets only when you consider them in the light of their context, taking into account such things as the historical limitations of the data collected and the niceties of their online representation?
Perhaps you learn something in depth only when you make mistakes, when things break down, when errors and lacunae take away received wisdom and unmask your bad habits. I have to swallow this interruption of service as a fact of life, and to accept my own small addition to the march of human folly. Meanwhile I will try to re-establish my services to the visitors of my website as soon as possible. However, putting the necessary security devices and tuning the appearance will yet take some time. Most important, next time I will carefully prepare procedures before upgrading to something which looks promising and useful, but is also potentially dangerous. As for now I had better resume working on it, and leaving you hopefully a bit more informed about my site’s crash, perhaps a bit uneasy about your own website, or just amused – or annoyed! – by this unexpected mixture of daily misery and some historical information.