Oh no, not emulation!

27 Sep

Today amid a flurry of tweets between Andy Jackson, David Clipsham and Euan Cochrane discussing a couple of my problematic PowerPoint 4 files, Euan suggested “Just emulate the PPTs”. It was just part of a tweet, so you have to expand a bit; he was suggesting essentially that we run a Mac System 7 emulator (or similar), and run a copy of PowerPoint 4.0 for Mac on that.

My heart sank!

Emulation is a good solution to some digital preservation problems in many ways, as it brushes away many of the problems of accessing and using old content in today’s environment. Instead, it lets you access and use the old content in its old environment. But this can introduce a whole set of different problems. How well do you now know that old environment? How easy was it to use? How reliable was it? What dependencies are there? How well is the emulation implemented (and in particular are any important elements missing)?

As part of researching my obsolete PowerPoint problem, I’ve been re-reading old email folders, including my Mac email folder from 1995-2000, when I was at Warwick University as Director of the UK Electronic Libraries Programme. Here’s a typical example of a desperate message from 1997:

“Martin, I have un-installed the [version of MS Office] on Michael’s CD, and re-installed (standard installation) the one on your CD. Same problem (open Word document, go to view outline, system freezes). S**t.

“The crashes have been playing havox with the disk & I had to give it a full Norton to fix it: bit map error, MDB error and a dozen or more file errors.”

Or this:

“My reasons for interest in the upgrade stems from the serious flakiness of 7.5.2 compared with the previous version. Since I moved to the Power processor and the new version of the OS, I have had innumerable frezes, often associated with printing, as well as the closedown problem more recently. Life is getting seriously painful at times. There may of course be another reason for all this.

Then there’s a lot of stuff about connecting printers. You get the idea: systems in those days were MUCH more flaky than they are now. The thought of having to try to install a software package on an emulated system base without a strong team of experienced people nearby was… worrying!

I had another reason for concern. In my previous job I had a go at recovering “lost” files for a few folk. One of these was an Amiga diskette with a student dissertation on it, for a member of the DCC staff. The Amiga diskette is an odd thing, but that’s another story. We managed to recover the bits from the diskette, but the only way we could make sense of the contents was to feed it to an Amiga emulator. In that environment, my colleague could see, read and edit his dissertation. He was pretty chuffed about that. But, we could find no way to extract the contents of his dissertation from the emulated environment and get it into today’s environment. So there it was, kind of frozen in the past, visible through this special lens, but not really something he could USE.

So that’s why the idea of emulation was worrying me!


3 Responses to “Oh no, not emulation!”

  1. Euan Cochrane (@euanc) 29 September, 2012 at 14:40 #

    Hi Chris,

    I thought it necessary to give a bit of the other side of the emulation story.

    Firstly, as you later remembered and mentioned on twitter, most of(all?) the text from the dissertation on the Amiga disks was eventually extracted and moved to a modern environment using postscript printing and pdf. This is an important point as it addresses your final (and main?) point: that you are worried about emulation as it traps content in old environments. The example illustrates a scenario that is actually quite a common approach for getting content out of emulated or virtualised environments, i.e. printing to a virtual printer which saves a file which can then be accessed on the host pc (or a remotely connected pc). There are other methods also, many emulators and virtualisation tools enable text to be copied between environments and all of them can have screenshots captured from them. Virtually all emulators and virtualisation tools also enable files to be saved out from them so that emulated software can be used to migrate content to more modern or more open formats (e.g. http://www.openplanetsfoundation.org/blogs/2012-02-20-ufcmigrate-service-now-available ).

    But more importantly in the case of the powerpoint files there was no other option other than using old software to open and migrate the files. I find this important because it is an instance of having no choice but to make sense of the old software in order to preserve access to the thing you were trying to preserve access to. The point I’m getting to is that when you said:

    ” systems in those days were MUCH more flaky than they are now. The thought of having to try to install a software package on an emulated system base without a strong team of experienced people nearby was… worrying!”

    I felt like you were over emphasising a minor problem. The reason I say that is because in both of the examples you cited above someone was (reasonably easily) able to install the right software and access the files. For instance, after we tweeted about the powerpoint files it was a matter of hours before someone had accessed them using an old version of the powerpoint for mac software
    In actuality I think that (for the most part) emulation requires a much smaller/simpler set of technical skills than the alternative preservation methods that are available. It seems to me that the average new archives/librarian grad would have no trouble installing most old software applications on a virtual or emulated environment. Moreover my experience with migration tools has been that they are technically challenging and complicated to configure and even more difficult to verify for efficacy.

    But that is just opinion (though it would be great to get someone to verify that). I think an important point to be made here is that emulation worked in both of the cases you mentioned AND it provided a high quality ,probably complete, version of the original interaction environment that the user had originally interacted with. This is something migration simply cannot do. In cases where information integrity is important then there really is no choice. And in cases where we don’t know if it will be important or not, then surely we should make sure the option is available? And finally we are at my main point: if we are going to need to make emulation available for most things then we need to address the very things you are talking about in your post. While I don’t think the problems are nearly as bad as you have painted them in this post I do agree that some old software is difficult to install, some emulators are difficult to configure, but that doesn’t mean we should just give up though! We should take this as a challenge and overcome it. We should make emulation seamless for the end-user and easy for the average archivist/librarian/consumer/business person to use to access their valued old digital objects.

    There are examples of this happening already and we should build on them and make them work more widely. The Good Old Games Website (http://www.gog.com) incorporates emulators to enable access to old games that it sells for use on modern operating systems and it does this without the end user ever knowing. The remote gaming services that are popping up like Onlive (www.onlive.com) give examples of how we could make remote emulation seamless for the end-user. (e.g. http://www.openplanetsfoundation.org/blogs/2012-05-14-next-generation-access-emulation-onlive-gaikai-dp) . When I use virtualbox on my computer to virtualise modern operating systems I can easily and seamless move my mouse from the host OS (in my case Windows 7) to the virtualised OS, and I can also copy text between them and (if I want to configure it) share folders of files between them – These features should be added to those emulators that don’t have them. Its a matter of will and $$ and nowhere near impossible.

    I posted a while back about how emulation might be incorporated into a generic digital preservation workflow: http://digitalcontinuity.org/post/20400819609/incorporating-emulation-into-a-business-as-usual . I strongly feel that it provides the most viable long-term solution to most preservation needs and I hope I can change your mind about it also!




  1. PowerPoint 4.0 for Mac: the story so far « Unsustainable Ideas - 2 October, 2012

    […] not keen on either of these options! My reaction to Euan’s suggestion prompted a rather hasty blog post, to which Euan has given a measured […]

  2. bwFLA – Funktionale Langzeitarchivierung » OPF Hackathon – Emulation: learn from the experts – Tag 1 « - 13 November, 2012

    […] Espenschied, benutzen alle Macintosh. Klaus Rechert sprach das Beispiel von Chris Rusbridge an oh-no-not-emulation Chris wollte einen Mac 7 Emulator haben und eine Kopie von PowerPoint 4.0 darauf laufen lassen. […]

Comments always welcome, will be treated as CC-BY

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: