Longevity, incept dates and that sort of thing

I watched c-beams glitter in the dark near the Tannhäuser Gate.
All those moments will be lost in time, like tears in rain

As tempting as it is, this isn’t going to be a piece on Blade Runner. Instead it’s going to be about horrors facing your digital creations as time goes by.

Recently I saw a discussion about when ‘web design’ became part of the curriculum in design schools. For me that was around 1996 which means I have a bunch of archived digital files from that period. Now they haven’t seen the light of day in a longtime but I still have access to them. Thankfully Every few years I’ve been moving them off a dead format so they aren’t trapped on floppies or Zip drives (or even CD-ROMS), so that now they are on external hard drives and in the cloud. However, in my diligence I completely forgot about some of the nuances that working on a Power Mac running OS 7.5.3 gave you. Little things, like not needing file extensions. So when I decided to take a look at these 20+ year old files, I had a bit of a shock. The best thing was that anything I’d created for the browser, basic HTML/JS pretty much worked. Thanks to the strict naming and simple assets of early web design most projects displayed almost as they did back then. There were a few issues with text sizing and layout as too early for CSS and the “<basefont>” tag is no longer supported, but on the whole the web stuff worked. Of course if this had been more complex like a PHP or Perl thing, I think I’d have given up.

That said, things got more tricky when I started looking at the files for the print based design work. In my super smart, all knowing macOS Sierra finder, all the files without extensions added to the end of the filename were unknown ‘unix executables’. That clever ‘simplifying’ UX decision in OS 7.5.3 was no longer understood in the modern mac OS. So that meant trying to work out what type each file was, and remember what the hell I’d created it in; was it Aldus Pagemaker, Adobe Illustrator, Photoshop, QuarkXpress, Macromedia Macromind or Director. Was it a pict, pic, tiff, gif, jpg, bmap or bmp, or was it just a text file? Where I had extensions (usually because I’d needed to open them on a PC) some of the modern day applications associated and even manage to open things. Adobe was particularly good at this; once an EPS, always an EPS. I’ll talk about Flash a little later.

I had one particular thing I wanted to ‘resurrect’, it was my Master’s thesis. The printed version got ruined a few years back so thought a digital (maybe PDF) version would be a good idea at least I could then get it reprinted.

I had a hunch that the main files were QuarkXpress 3.2 (or something close to that). Thankfully I’d named them reasonably clearly so I could tell the difference between them and the assets in the folder. First up, add the file extension, after a bit of Google I found it ‘qxd’. A bit more Googling and it appeared that Adobe InDesign could open QuarkXpress (pre v5) natively, no questions asked. This could be my lucky day, but alas, no longer works in Adobe InDesign CC 2017, or it didn’t like my files.

So this is what it took:

• Guess the file type and extension,
• register and download a 7 day trial of QuarkXpress 2017 (yep it’s still going), which alas also doesn’t open 3.2 files,
• download and use a QuarkXpress convertor (version 3 to version 10),
• open file (woohoo!), missing assets/images, missing fonts (boooo!),
• close file,
• spend half an hour trawling through ageing personal font library, find one of them (just as old — install using Fontbook, fall off chair when it works), fail to find the main font,
• find an online source for the missing font (fonts.com as it was a Monotype Font),
• register for the Monotype Library trial on Fonts.com, select the font to start my 1 hour trial of it!
• download the proprietary font management app Skyfonts,
• install, set up and make sure font is available,
• open QuarkXpress file,
• use the names of the missing assets to work out which other items in the directory (and nearby) are images and eps’ etc, guess/hope that the images are all ‘tiffs’,
• rename the image files with correct extensions,
• update/relink the files,
• give the document the once over to check all is as it was and there aren’t any reflow issues or text anomalies,
• realign one illustration (feel stunned that everything else looks ok),
• export to PDF (with just a few minutes to spare, phew).

Now despite all that, luck was definitely on my side. For a start QuarkXpress still exists (and offers trials). Without that I would have been hosed. The other options did not look very appetising. One option was to buy a plugin for InDesign for $200 (as if, the cheeky bastards), another was to go into the files in Textedit and manually fish out the text from all the other ASCI crap and recreate the document (nope, don’t care that much about it), another was to build an emulator (Chubby Bunny or the like) of a mid 90s Power Mac (I had the 7200) and see if I could find original install files for QuarkXpress (no comment), or as a last resort (and not the worst idea) buy an mid 90s Power Mac with the old software off Ebay.

With all these trials and tribulations fresh in my mind, Adobe announced that it was finally ready to kill off Flash by 2020 (“no really, we aren’t joking, we will kill it this time”). What was a little weird was the sheer amount of media coverage that generated, but I guess fitting after the years of banner ad hell it encouraged. Then I saw Jeffrey Heer post about how we better find a way to save some of the groundbreaking data visualisation.

This got me thinking about the longevity issue for digital files again. I’ve discussed it in the past, using the BBC’s Doomsday Book (1986) laser disc project as an example. How that couldn’t be widely (in fact barely at all) read after just 16 years, where the actual Doomsday Book is still usable over 930 years after it’s creation. The mix of format (laser disc) and system BBC Micro, were the main culprits as neither were widely used and thus had a small market footprint.

These days the archivists and technologists have got wise to this and save entire systems (machine, OS, storage/files) along side any artefacts, but that will require people who know how to use them. The holy grail is some other form of software that can read and emulate any software previously created (but don’t hold your breath). And then what’s that running on? May be it will be the new bogey man AI that Elon’s so worried about. Even then we then get into the debate of what’s worth saving? Artefacts of historical, commercial or cultural value are top of the list, but what about the stuff you or I have or even some of the people you most admire, perhaps even leading luminaries of your favourite obsession; is their stuff on the list to be saved?

Physical artefacts, require housing/space, environmental controls and of course are at risk of damage. I once heard an archivist (on a TV show) say that he always tried to get 3 of a thing, one to be shown/admired, 1 to be held untouched in another safe location and 1 to potentially sell. Physical artefacts are limited by reproduction and some of them may use components or fuel that’s no longer available but on the whole if correctly kept they last, as they mostly aren’t fully dependent other mechanisms. Digital on the other hand is infinitely reproducible (at least as data) but it’s always dependent of other mechanisms. The software to view, open, read or run it. The system to run that software and the hardware to run that system.

The best work/hope today seems to be from the Internet Archive. They are storing tonnes of scanned documents, media, files and have a range of software and system emulators, although I’m not sure about running your own files on their emulators.

From a creative/creators perspective it makes me wonder about how much of what we’ve created over the last 20/30 years will still be accessible in the next 20/30 years, let alone in 200/300 or 2000/3000 years. I guess as with the things of the past, those of ‘value’ survive. Enthusiasts look after them, collect them, transform them. But the archivists and collectors will need to find new ways to save our collective history. New methods that will require more active engagement with the content, probably by transforming its digital state from one thing to another (like I did for my QuarkXpress file), which is something that doesn’t sit well with archivists. Either that or the software we make needs to always be keeping an eye on the past, and whenever we make the decision that this version will not open older versions we need to think that through, and be willing to open source what’s now missing, so that the enthusiasts can step in.

After all what’s the future equivalent of finding a shoe box full of photos in your grandparent’s attic, a firewire portable hard drive with an Aperture library on it? Good luck with that.