29 November 2023

Digitisation is death to data

A few weeks ago my laptop computer was stolen, and since the insurance company asked me what it would cost to replace it, I've been stuck, because it is beginning to appear that it is irreplaceable.

The computer that was stolen was about 13 years old, a Toshiba Satellite laptop running Windows 7 32-bit. It came with 64-bit Windows, but with a set of discs for a 32-bit version. Using it for a couple of hours showed me that it would not run many of my older programs, which I used to access research data I had collected over the last 25 years,  so I quickly installed the 32-bit version of Windows and all was well. 

Now, however, it seems that 32-bit versions of Windows are extremely difficult or impossible to get. I can still access my data on my desktop computer that runs 32-bit Windows XP, but what happens if that dies?

Referring to a related issue, director Guillelmo del Toro pointed out the dangers of streaming serivces when he said:

Physical media is almost a Fahrenheit 451 (where people memorized entire books and thus became the book they loved) level of responsibility. If you own a great 4K HD, Blu-ray, DVD etc etc of a film or films you love... you are the custodian of those films for generations to come

But there is another bigger problem. Even if you possess the physical media, they are quite useless if you have nothing to play them back on. And for many of these things the hardware and software to read such physical media is becoming rarer all the time. 

For a long time people have been recommending the digitisation of paper documents, with or without the destruction of the originals, as a means to better preservation, but that depends on the availability of the hardware and software to access the digitised versions. Someone wrote a book a few decades ago called CD-ROM: the New Papyrus, but how easy is it to get a computer that can read a CD-ROM? Better stick to the old papyrus!

One thing that would go a long way towards alleviating this would be for historians, librarians, archivists and others who are concerned about preservation of information from the past to push for international agreements and legislation to ensure that whenever an operating system, or version of an operating system is no longer supported by its manufacturer, it should be put into the public domain, without copy protection, and possibly also made open source, so that people can adapt it to run on new hardware.

Something similar should be done with application software (apps) such as word processors and the like -- how many people can read a Multimate document nowadays?

A personal example: 

Since the age of 11 I have kept a diary, originally written in pen and ink in a series of notebooks. In 1985 I began digitising it. I typed out the entries for 1969 in Wordstar on an Osborne Executive portable (luggable!) computer running CP/M3, stored on 185k single-sided floppy disks. I did it because I wanted to collect memories of my grandmother that I had written back then when I had seen quite a lot of her. 

Later I realised that quite a lot of what I had written might be useful to historians of Namibia, so I continued to transcribe it beyond the period relating to my grandmother. In 1987 I got a newer computer running MS-DOS, and a better word processor called XyWrite. I converted the Wordstar documents to XyWrite (I still have the conversion program on my computer today) and carried on transcribing. In about 1990 I printed out an edited version of the Namibian portion and sent a bound copy to the Windhoek archives. 

In 1992 I began making notes for the current version of my diary using a then-popular "terminate-and-stay resident" program called Sidekick, which I would then use to write up the hard copy version. In about 1995 I started using a text database program called askSam, and stopped keeping up with the hard copy version. In 2001 I started using a different text database program called Inmagic, and began converting all the remaining Wordstar, XyWrite and askSam versions to that, and since 2006 have kept it in a single file. A couple of years later I had more or less finished transcribing all the hard copy ones going back to when I had started at the age of 11, and every morning I look at it to see what I was doing in the past going back at 10-year intervals. I can do that on a computer running 32-bit Windows, but not on one running 64-bit Windows. So 64-bit Windows is quite useless to me. 

That is just one example, but there are many other things, like research notes made from books, interviews with people, with research data that I've now been collecting for 35 years (and older data that I have digitised in a similar manner to the diary). but the planned obsolescence policy of software companies like Microsoft would require that I must give up all that. Perhaps I need to do a "Go Fund Me" appeal for the funds to print out all the stuff on my computer on hard copy in order to have continued access to it.

Digitisation as a means of preservation only makes sense in an open source and public domain environment.


No comments:


Related Posts with Thumbnails