There’s a peculiar kind of bitter-sweet comedy in reading old print articles about the Internet, isn’t there? A volume I was looking at for other purposes has a state-of-the-situation piece in it from a conference in 1997. This sits well with me as I was just about getting the hang of the Internet by then, having graduated from Mosaic on a Mac SE30 to Netscape on a Quadra (Netscape Navigator 3 Gold, no less, whose Word-a-like HTML editor is still one of the most straightforward ever created) and having first encountered mentions of The ORB and so on. But of course, and the article itself anticipates this, the web changes a lot very fast, and those were the slow days compared to now when really, it was still quite strange to have an Internet connection unless you were a big business or, paradoxically, a student. Some things don’t change: it’s amusing, for example, to see soc.history.medieval being cited even then as an example of how much low-quality information there is on the ‘net to sift through (and I see from a quick look at the Google Groups page that D. Spencer Hines is still, a decade on, doing his level best to make it so). However, pretty much every other resource it mentions is gone, or at the very least, not where they left it. In some cases this is because it’s actually been worked on, updated and moved, like Carrie, the full-text library of the University of Kansas. The whole University has changed its domain since the article was written, which twits both the print link here and even that given from the ORB (or at least from the cached version of the ORB since poor old ORB, already becoming outdated and underloved when I first met it, is down at the time of writing). But it is still out there, though the name changes have made it difficult to happen on it without cunning websearchery. Another case is Virginia Tech’s Bibliography on evaluating web information, which is actually really useful and deserves wider publicity, but still isn’t at the URL or under the name this article gives any more and has to be Googled. Labyrinth, at least, is still where it was and serving more or less the same content. I mean, it’s not where it was, actually, but Georgetown University’s IT department have at least been kind enough to ensure that the old URL redirects to the new one, which as you can see from the above very rarely happens. (Honourable mention here to the University of Leeds’s International Medieval Institute, now the Institute for Medieval Studies.) Even when it does, as with the Online Medieval And Classical Library at Berkeley, it can still be screwed up.
The article also covers search engines, from a period when I’d just about woken up to the fact that really, anything was better than Yahoo but that this still basically left one with Lycos or Altavista. Thank goodness Google haven’t yet proved to be evil. I didn’t even know about MetaCrawler, despite it being old enough to get mentioned here, but since it returns me as second result for « tenth AND medieval » whereas Google returns me now (I’m childishly glad to see) as top result for « tenth medieval », which was long ago the challenge for this whole exercise, I think I know who wins…
It’s also salutary to see a short and simple explanation of why, in 1997, the WWW was already going to drive out Telnet (inaccessible both because of its authentication and its slightly arcane command set) and Gopher (like the WWW only less capable), because although I still maintain an online diary on what began as a Telnet server (now SSH, for which reason no, you can’t find it) I don’t know of any library catalogue that still uses a Telnet interface even though they were often quicker and more transparent than the OPAC front-ends that now proliferate—hey, hey, no, the School of Advanced Studies in London is still running its one. Once you’re in it warns you “this service is no longer in use”, but since it turns up two pamphlets I gave to the IHR last year I’m not sure what ‘not in use’ means in this context. (The cite the article gives for a list of Telnet catalogues is still linked from the University of Madrid site that they give but itself returns a 404.) And, anyway, Gopher is all gone isn’t it. The WWW doesn’t necessarily do things as well but it does it without the learning curve. You can tell it was 1997 however because the authors don’t appear to realise that there is more than one programme available for browsing the web, in fact they don’t mention programmes at all.
But you know what’s missing? This here bit of the medium. I remember that in 1998 or so, 1997 was still a bit early but by 1998 it was beginning in my part of the world, people were starting to use this thing called Livejournal. And I suppose Blogger was also up and running by then? But I didn’t meet it until later when a casual web-search for I-now-forget-what took me to the Blogenspiel of Another Damned Medievalist and that alerted me to the fact that there were in fact people blogging in my field. No hint here of this informal educational medium that’s so challenged, well, newspapers and so on, for informed commentary and, well, education. I mean, even a couple of years ago `blogger’ was still a term of ridicule, Cafepress are still selling “I’m blogging this” t-shirts from when that was an empty threat because reading blogs was something only misfits did. And now, well, for heavens’ sake, the popularity of this here site has been going down for some months—it’s hard to tell because of the substantial but varying input of the Český Krumlov queries, which persist, and rather drown an apparently declining traffic from other sources—but I still draw ten thousand hits a month, and I am not exactly mainstream media, you know? In 1997, anyway, this appears to have been off the radar, which given that it may be happening right now is perhaps forgiveable. But even this is recognised in a small way by the fact that it is because of this change that one of their links, once located via Google, the old journal Le Médiéviste et l’Ordinateur, ceased publication in 2007, seeing a sea change in Internet use that made it effectively redundant. Their page now directs readers to Digital Medievalist and Ménestrel, among others. I wonder if I’ll be able to do a post like this from a print article in 2019?
Will I in fact be able to do a post like this in 2019 at all? I’ve mentioned all this mainly out of nostalgia but also by way of giving weight and perspective to a rather apocalyptic-sounding comment I recently made at Modern Medieval and an older one I made at In the Medieval Middle that didn’t make it through moderation, to the effect that anyone who thinks putting something online makes it permanent is fooling themselves. Where’s Simon Keynes’s Anglo-Saxon Studies Bibliography? Even when, as with Carrie, it is still there, you can’t necessarily find it because the domain has changed and so has its name so all your search terms are uselessly broad. So keep offline copies and contribute to the Internet Archive, I guess, and remember that we work on sources that survived a thousand years or more because someone wrote them down in ink on skin and someone else, most likely, packed them between wooden boards and then generations of someones else kept them somewhere dry. Your CD-Rs will not last that long, or probably even as long as you do. Also, manuscripts (or books or print journals) don’t need mains power to be read. The low-tech will still be worth thinking about for a while.
Celia Fernández Corral & Enrique González Alonso, “Latin Medieval e Internet”, in Maurilio Pérez González (ed.), Actas, II Congreso Hispánico de latín medieval (León, 11-14 de noviembre de 1997) vol. I (León 1998), pp. 449-462.
It is terrifying how quickly modern data becomes inaccessible. In a previous existence I worked for a Big Oil Company and the office I worked in spent all its time piecing together data about oil fields. Not just data from the 1960s, when oil engineers were real oil engineers and everything was printed on very long rolls of paper, but data from this century too. Data which had been transferred from program to program, tape to disk to tape again, corrupted, misunderstood, lost, found, miscatalogued… Even after such a short time it required trained specialists to make this data meaningful again.
(And what did we do with the data when it had been fixed, labelled, archived, stamped, sealed, and approved? We saved it to CD-ROMs. CD-ROMs, which, shortly before I left, were turning out not to be as permanent a storage medium as people may once fondly have hoped…)
I’ve told this to many people, but when we first digitised the Museum’s registers of donations, the first thing that was done with them was to print them out again, on archive-quality paper, and stash them in a controlled-environment cupboard. After all: whatever system it’s running on, future OCR is probably going to be pretty good, and we can already database formulaic texts like catalogues pretty effectively. Getting it back into digital format if that’s ever necessary should be simple enough. Getting it out of Word format after the eventual meltdown of Microsoft in the Teraport Wars, however, maybe not so easy…
Pingback: Quid plura? | "Frame me, and hang me on the wall..."
an older one I made at In the Medieval Middle that didn’t make it through moderation
Sorry! I’m guessing that was user error.
I’ve pretty much duplicated it elsewhere now except that I recall asking rhetorically why we think PDFs are such a safe format when they’re proprietary to Adobe and therefore commercially vulnerable. So it doesn’t matter, but thankyou for coming over here to respond anyway.
Yes, I’ve ben thinking about this for a while. The context in which this issue first came to mind was when my University library began dropping print subscriptions in favour of online ones: extremely handy in the instant you decide you want to read something, but much more vulnerable since if they ever decide to stop subscribing, there will be no hard copy to fall back on. (Not to mention depending on things like servers being functional, and there not being a power blackout, etc…)
At Kalamazoo last week there was an angst-ridden hour of discussion about ‘preservation’ of digital resources like the digital doomsday: this basically boiled down to saying ‘yes, backing up the data is important, but what about the software and hardware necessary to use it?’ There was no substantial answer in the room other than that lots of people agreed there was a problem. It left me with the distinct feeling that the old technologies are still the best (she types with a knowing sense of irony…)
So, can this problem be resolved or should we just regress and be content that this is, in the end, the safest way to preserve information? I mean, we could put it on paper, but then vellum is probably better and lasts longer, but then stone is even better…
I’m only partially trying to be a pain. Granted technology nowadays requires power but, Battlestar Galactica ending notwithstanding, we’re not going to summarily give up electricity and start anew. And if we HAD to, then I think that we’d have arrived at an apocalyptic moment and it wouldn’t really matter anyway.
The electricity is an extreme problem, for sure: loss of accessibility, either through obscure file formats (reading, for example, Commodore word processor files now) or storage media (5.25 in floppy disks? Hell, ZIP disks!) is a lot more serious. And if Blogger was shut down by a data centre fire or bankruptcy, how much would you lose? Actually: I haven’t backed up this blog for months…
So, you say, what’s the answer? Well: the Internet Archive, for academic journals gone online initiatives like LOCKSS (Lots Of Copies Keep Stuff Safe), periodic updating of non-plain-text files to new open-access formats, and an acceptance that all these are only short-term measures that will need new short-term measures in due course. Also, don’t just make backups; refresh them, and every now and then, check on them to make sure they haven’t rotted. If you have a load of old CD-Rs from years ago, bow’s a good time to put them on a flash memory stick or external drive; my housemate discovered he’d lost all his 1990s games the other day because his CD-Rs were no good any more. Basically, don’t be complacent, and maybe some things would be worth printing out too.
I think vellum is probably more durable than stone, too; less brittle. But, I realise, you were joking.
This is a particularly relevant discussion given the H-Net appeal. I know they have a plan to archive their stuff, but I don’t think it’s got anywhere yet.
Indeed, well, there is what I hope does not become a perfect example of how such a resource can disappear…
Pingback: Airminded · History Carnival 77
Pingback: Seminars CXIII & CXIV: Vandals and burning houses | A Corner of Tenth-Century Europe
Pingback: Metablog XI: link clear-out | A Corner of Tenth-Century Europe
Pingback: Links of hopefully-still relevant interest | A Corner of Tenth-Century Europe