Interview 3 – Future Nostalgia Project
Place: Digital Preservation Lab, Cambridge University Library
Interviewer: Leontien Talboom
Interviewee: Will Sheldrake
Date: 11 June 2025
 
Interviewer: To start with a little bit of context. When I started working here floppy disks came into your collection, of course, and I was like, oh, that's fine. We've got a floppy disk drive and a FRED workstation and we've got a bunch of USB-floppy disk drives around , so I thought floppy disks were not going to be a problem. Very quickly I learnt that they were a problem. You probably know this there's quite a lot of quite obscure formats. You can't read everything with a USB-floppy drive. You need an internal floppy disk drive. So very quickly came to a point where I was like, the equipment that we currently have is not enough and started asking around and stuff, but very quickly also realised that within the digital preservation community, so the community that I sit within, so they're trying to keep everything for the long term. We don't really have any best practices or anything around for disks and all that type of stuff. So the idea of the project is basically to set up a resource with people within the community who are in similar jobs to me, work with floppy disks. But also, chat to people who sit outside of our community. So you, for example, who know far more than us about everything, and just see if there's... we can probably learn a whole bunch from you and from just the whole retro computing community because there's so much more out there. So yeah, it's basically about more of the idea of bringing everything together and not having a dispersed... now you probably know as well, like if you type something in Google and you try and find something, you end up in a forum somewhere and someone has asked a question and you're really excited to see what the answer is and no one has answered it.

Interviewee: There's not one yet.

Interviewer: Yeah. It's like, I know there's someone out there who knows the answer. It's that kind of idea. So that's the project. I send you a bunch of questions beforehand. But I was wondering if you wanted to maybe start out with just chatting about how you started being interested in computers and all that stuff.

Interviewee: I guess, yeah. My interest in computers is... I'm old enough that I was just quite small when the... if you're aware of the BBC's computer literacy in schools project came around and this encouraged my parents to buy a home computer because it was educational. Like many parents at that time, it was in the UK, they weren't going to stretch the £400 or whatever it was for BBC Micro, so it was a spectrum. That's how I ended up getting started in computers back when I was quite small. Most of the media on that machine was a tape deck, it was cassette tapes, and I still have many of these tapes. Actually I find that that particular media is quite reliable and most of the problems relating to tape decks these days, the tape is actually the deck rather than the tape. From there, I went on to do... I upgraded that into an Amiga, so that was probably my first home experience with floppy disks apart from school. We had BBCs and 5.25-inch floppy disks and things. That was the first time where I was introduced into copying disks and particularly into the different formats that were around and being a teenager at the time and only having your pocket money, your pocket around money to go with. So you would swap disks with people and then you are introduced to these formats and the tricks that sometimes would be played to stop you duplicating disks.

So from there I did a degree in computing and so yeah, that's more going down the line of my history in programming and things like that. That's still what I do to this day. I'm an embeded software engineer. So I've spent majority of my career either making applications for PCs and things or more recently doing work with Bluetooth.

Interviewer: Oh, cool.

Interviewee: So I worked with Bluetooth for many years and now I work with mobile phone cameras. I've been doing that about 18 months, focussing on image stabilisation and things like that.

Interviewer: Cool. But your interest has always stayed with the old computers?

Interviewee: Yeah. So going back to the old stuff, back in the '90s probably when I was moving one computer to the other, I ended up keeping those old machines or finding... my friend had a Commodore 64 and I managed to find one of those and you end up with a small collection of these machines and one thing leads to another, and that led to my getting involved with the Computing History Museum. So which all sorts of things come across your desk.

Interviewer: They've got a great collection of interesting stuff. 

Interviewee: Yeah. Exactly.

Interviewer: There never seems to be an end.

Interviewee: There is a never an end to it, and always something else weird turns up. So I guess that's my part of history of computers and things like that.

Interviewer: Nice. I like that you've always had an interest in them and it's interesting how, since being a child that that has stayed on. It's very nice.

Interviewee: Yeah. I suppose for me, I've been involved with software professionally, so actually it's interesting. I haven't done a lot of software type things with old vintage computers, but it's interesting for me to not do the work day job stuff and look at the hardware type of things. So it's on my bucket list and maybe I'll do a software project one day, but it hasn't really happened, obviously. On the things I do are, I'm an embedded software engineer. So quite low level, so it leans itself quite well to some of these older machines, so potentially.

Interviewer: Yeah, I can imagine. Nice. Are you happy for me to just go through all sections on the question sheet?

Interviewee: Yeah. However you want to do it.

Interviewer: Cool. So some of them are very specific to floppy disks, but I'm also very happy to just hear in general about sourcing or looking after all the equipment. Because I say, a lot of it is quite similar. You probably know this, there's a lot sourcing material, finding manuals, that type of stuff. It overlaps in a lot of ways. So starting off with sourcing and maintaining equipment, where do you... because you were saying you've got a small collection of computers that you've acquired over time. How do you maintain those computers and where would you source them from?

Interviewee: The sourcing of them, it's been anything from house clearance auctions. So originally, I would get machines where a house had been cleared or something like that, and they would just end up there. Occasionally, you can still find them in these places. But obviously, there's a collector's community around things now. They're more sought after, but a community of things is also quite important. Obviously, the internet has helped a lot with these types of things. I have an interest in looking at old, especially old magazines and things from computers back in the '80s where people... this is pre-internet type things, especially if they're more like a fan type thing and people would write in with their hardware questions to try and get this information. But I suppose that was just a slower, more or less convenient way of accessing community type things. Especially in perhaps back in the UK and the rest of the world in that early '80s time, it was a wild West with computers. It was often very small communities around the computer they'd chosen to buy and when you get into the late '80s with the Spectrums and the Commodore 64s and things, there was heaps of experts around and magazines and books and everything else. But if you'd chosen to buy a Tatung Einstein or something else weird, you are on your own, really.

Interviewer: Like these magazines would be the case that people would go to find their community?

Interviewee: Yeah, and often they would be like a user group, and this would be the magazine that came with the user group like an owner's club, this type of thing. You would see all sorts of interesting stuff that people had sent into these, quite niche type things. Potentially, there's information in these type of things as well that is useful, that is untapped because they're hidden away in the pages of these journals, these books, magazines, and otherwise.

Interviewer: Because where would you find those magazines? Because I know the Internet Archive has quite a lot.

Interviewee: Yeah, a lot of them are there. But oftentimes, I've got hold of them. I don't have a vast collection of these things, but sometimes if I've bought a machine especially if you can find a machine that's been owned by that owner and they've just taken it down from their loft because they're moving house or something, and they'll say, here's the computer that you bought from us off eBay, and here's some stuff that went with it. Then here's a load of disks and here's a load of books and things that you didn't know you were getting and often, it's those things that are more interesting than the machine.

Interviewer: Wow. Okay. That is really interesting. Because I think that's what we lose out on quite a lot. So like floppy disks will come in with collections, but we will have no context around them. So the thing that we go off is, even if they have a label, is the label. So that will give us a little bit of an indication of what it's been formatted as and if we're really lucky also and what machine it's been used. But it's really interesting to hear that, it's sometimes the context that is more valuable than the actual machine.

Interviewee: Yeah. Those bits and pieces that come with the machine. Some of it is just social history. I remember being given a disk and it was a man talking about his holiday, a letter he'd written about his holiday. My wife and I, and such and such. Again, especially at the museum, some of these disks, they might have people's personal information on and we want to be careful with it and not just give it away or put it in the wrong hands. But a lot of that stuff exists on those type of disks. What do you do with it? There's endless amounts of it, but it's there. But in a way, you could argue that that's more interesting than another copy of Windows 95 or something on a floppy disk. 

Interviewer: Exactly. Yeah.

Interviewee: I see you have video recorders and stuff around the place, there's people who are more interested in a mint copy of Star Wars. They're more interested in something that someone recorded because they've got an advert on it, or they're looking at the Ceefax that came off it or something like that.

Interviewer: Yeah, exactly.

Interviewee: So yeah, there is that social aspect to it.

Interviewer: Yeah. That's, of course, the aspect that we're really interested in it as like memory institutions, it's like the personal data that people have put onto it. Because normally, they come in with collections of, maybe it's a writer or a poet and that type of material or scientist or someone who works at the University. That is the stuff that we want. So we don't really care if they've got... like you say, it's cool if they've got a copy of Windows, I don't know, whatever. But it's actually really cool when we find like early drafts of manuscripts and that type of stuff which they've been working on. Because for us, that's the valuable stuff. But yeah, it's very cool to hear about the magazines and how useful they can be as well because that is something like... I think you're right there. It's quite an untapped point of information for a lot of these types of systems. Would you expect for these magazines, and you say certain user groups would sit around that, what we see a lot is that, the computer systems that were used were quite businessy. So it's word processors or it's stuff like research machines, that type of stuff. Would they also have magazines like that?

Interviewee: Probably less so, I would think. Once you get into the business end of things, lots of machines were on service contracts with firms. If you had a problem, you spoke to them. That being said, lots of documentation was produced for those type of machines. They were very meticulous about documenting everything. If you can get hold of those manuals, then yeah, they could be a mine of information usually technically about the system. If you can make sense of it without being the trained technician back in the '90s and '70s or so, it was.

Interviewer: But I would assume that in those manuals they would also give information on how the floppy disks were formatted and that type of stuff?

Interviewee: Yeah, potentially.

Interviewer: Potentially. Okay.

Interviewee: Because a lot of this information was not shared with the end user. So a lot of that gets back to the company and when they've written off that computer, it ends up in the skip and if they're not supporting it anymore. So there's possibly a potential for a lot of that more to be lost because it's never been put in anyone's loft.

Interviewer: Yeah, exactly. That's a big worry for us, basically. Like you say, a lot of it may have already ended up in a skip because like you say, it's not seen as useful anymore.

Interviewee: But equally, it's strange how some of this stuff comes out of the woodwork and I think it was a year or so ago, like most collectors like me, I bought something off eBay and the chap was like, "Do you want to see any of this other stuff that I've got?" Turns out that his father had passed away and his father used to work in one of the air bases and he used to maintain the computers that were there and in his office that he... obviously his father was obviously repairing video recorders and computers and he was tinkering. In the corner was a machine from the 1970s with all the manuals and all the big 8-inch floppy disks and the disk drive. Where do you find that? All the manuals and everything.

Interviewer: Everything with it. Probably not realising how valuable that is as well.

Interviewee: It's in the museum now.

Interviewer: Good. That's where it belongs.

Interviewee: That's where it is now, But that is just by chance.

Interviewer: But it also needs someone like you to be able to walk into that loft and recognised that it is something that's really valuable and really important.

Interviewee: Who else knows where it might have been and they were clearing out the late father's equipment and he might have just gone, oh, we've sold what we can, we'll just skip the rest of it.

Interviewer: Yeah, exactly.

Interviewee: So who knows?

Interviewer: Yeah. It's good that you were there.

Interviewee: Yeah, that's just one example. That's an example of a machine that was a business machine that was not something that somebody would've bought at home and tinkered with. So this was not a hobbyist machine or a small business machine. This was a commercial thing.

Interviewer: I think we will probably see more and more stuff coming from business machines as well. It's a weird mix of everything that we get because it depends where people have worked, where they've come from basically. But yeah, it's good to know that they're still out there.

Interviewee: Yeah. They're still around.

Interviewer: They're still around. But like you say, it's something people just keep, do they? People like you just need by chance have to come across it one day.

Interviewee: Yeah. It could have sat in his workshop for years or something. So who knows?

Interviewer: Yeah, exactly. But it's good that it's at the museum now. That's its found a new home. In terms of imaging floppy disks. Do you do that a lot? Imaging floppy disks?

Interviewee: Yeah, sometimes. Often it is the reverse. We're trying to make a floppy disk rather than read it because we want to run some piece of software or something like that. So yeah, backwards from a lot of your aims. It is often the other way round. That's probably do more of that than other way round.

Interviewer: So someone has already made basically the disk images and then you try and get them back onto disks?

Interviewee: Well, yeah. Well, I suppose there's an element sometimes with duplicating things, which is both steps or we're trying to recreate, there is already an image and we're trying to recreate it into some real media for use using the same machine.

Interviewer: What would you use for that? When I've been over at the Museum, it seems to be a mix of sometimes the floppy disk drive is completely taken out and there is a little controller that basically acts as the floppy disk. Would you have a preference for using an actual floppy disk drive?

Interviewee: It depends on the context sometimes. Some machines would take this drive out and you're aware of the Greaseweazle type things. We use that quite a lot to connect things up. We've connected up to the 8-inch floppy drives to those. We've adapted those to the Greaseweazle so we can write an image.

Interviewer: Can I ask you a little bit more about the 8-inch and how you went about connecting that up? Because that's one in our community that not a lot of documentation is around. Have you got any tips?

Interviewee: Yeah. Probably, I've got some spare PCBs that do the adaption.

Interviewer: Oh, that's cool.

Interviewee: Yeah, because you have to order them in fives, don't you? So yeah, I can certainly sort you one of those out for you. But they're mostly compatible. They usually have a 50-pin interface on them. But I believe that it is generally just a rewiring exercise. There's no controller needed to... you don't need to do anything clever. You just need to rewire it basically. So the adaption board that we have is just got 50-pin connector on one side and put one on the other side and just connect a ribbon and cable up.

Interviewer: And it would just work. Is that because it's quite low level that it's quite easy to read? Because that sounds a lot easier than some of the other stuff.

Interviewee: Yeah. The Greaseweazle happily talked to our drive and the concepts of stepping over tracks and things are still the same for that drive. I'm not fully clued up. There are different sorts of disks and formats that go with 8-inch floppy drives. There is lots of these disks at the Museum and drives. We have one there now that handily came in a big external box. So we use that one. Some of the earlier drives can be a bit scary because they have mains voltage, spindle motors and things. So the later ones are generally a bit more easy, nicer to work with and more reliable. So we use those for preference.

Interviewer: Yeah. So you would have a preference for... how much voltage have the later ones got?

Interviewee: They usually a 12 or 24 volt.

Interviewer: That's not bad actually. Better than a main.

Interviewee: Better than just a main voltage motor. They're also little smaller and lighter, so they can be quite heavy, old ones.

Interviewer: Yeah. I can imagine.

Interviewee: Yeah. Because there's a lot of cast structure in the actual physical when it's 8-inch. That’s just one of them and a twin drive can weigh like 40 kilos or something.

Interviewer: That is very, very heavy. I could be completely wrong here, but the 8-inch drives that I've seen, I've not seen that many of them. I've seen five or six of them. They're all very, very different. Is that just because that was still in that period where people were experimenting?

Interviewee: I think so. Personally, I've never experienced one professionally or personally outside of the museum. They were long since history, when I was active in computers. Apart from a few exceptions, I think there's a TRS-80 or something that uses... one of the particular TRS-80 models uses the 8-inch disk. But mostly, they would not have been put on home systems.

Interviewer: Yeah, it's the same. I don't think we will come across a... I know that some people have them in their collections, but we don't. I think, like you said, there's a very small chance because they weren't really used on home computers. Maybe if it's a business computer, but even then, a lot of them are just backups from servers and stuff. It's not necessarily the data that we would be after.

Interviewee: Yeah. Obviously once you're into a PC or something, you're not going to see it.

Interviewer: Yeah. Which is interesting, isn't it? Gives us one less worry, it's nice to know that people have got them to work with and that the Greaseweazle is also happy to talk to an 8-inch drive.

Interviewee: I think maybe you can even connect them to an older PC, you can connect them to a PC floppy controller and tells it an 8-inch drive.

Interviewer: And it will be happy.

Interviewee: It'll be happy.

Interviewer: Yeah. Which is nice, isn't it?

Interviewee: They're generally good.

Interviewer: You touched upon this when you were talking about the history around computers. But how do you handle like write-protected disks and stuff like that? Because they can be a real hassle sometimes.

Interviewee: What do you mean by handling write-protected disks?

Interviewer: So some of them seem to have some type of write-protection on them and they can be quite difficult to actually get any data from them. Or is that not really a worry because you're just replicating them?

Interviewee: Normally, I would consider that a right protected thing was been a problem if I'm writing to the disk and usually, you're aware, it's obviously just to put some sellotape over the notch and then we've solved the problem. In fact, that's probably an anecdote. People used to have obviously single-sided, dual-sided drives and some people used to... there used to be a tool that you could put the notch in the...

Interviewer: Flippy disks. Have you come across a lot of flippy disks?

Interviewee: Not so much, but I'm aware of them. People used to have their little notches thing to notch things out. Even back when I was at school, when the high density drives came in, it was a really bad plan, don't recommend it. We used to get a standing... can I use one of these disks [picks up a 3.5-inch double density disk]?

Interviewer: Yeah, of course.

Interviewee: So we would cut this hole out with a knife.

Interviewer: Why?

Interviewee: So you could get more on the disk. Cut the little hole out.

Interviewer: That would give you more room on the disk?

Interviewee: Yeah. Because you'd fool the drive into thinking it was a high density rather than a double density disk. Then you get more on it. It wasn't a great plan because often, not very reliable, but it would work.

Interviewer: Wow. I didn't know that. That's for the 3.5-inch that it worked with?

Interviewee: Yeah. Certainly, the later ones, if you find a double density 3.5-inch disk, there'll be a little indent. Like this one. There is a little indent, you see. So you can just cut that out and hope for the best.

Interviewer: You would cut that out and then put it into your high density drive and be like...

Interviewee: Hey, high density disk. Reformat me high density. I'll get more on.

Interviewer: That is fascinating. I didn't realise people did that.

Interviewee: It's not a great idea, but as a student or something... well, I've got more on my disk now.

Interviewer: Yeah. I find that really interesting about floppy disks and just hardware and how people use stuff in general. We've got a number of... I think they were software disks, but when I was looking at the stuff on them, they're just people's files on them. It's just like, well, someone that's just made the decision to overwrite whatever was on it before and just put whatever they want on it, which I think is fascinating because nowadays... or when I was growing up, I was at the end of the ear of floppy disks, but a lot of stuff came on CDs and you couldn't overwrite CDs. It's just fascinating that did happen in floppy disk.

Interviewee: Yeah. I suppose even though you could buy writeable CDs, that wasn't what people were shipping software on, was it?

Interviewer: Exactly.

Interviewee: So you couldn't, but yes, you'd often see, especially magazines would include disks. People would just stick a label on them or pen over them and say, oh, I'll just copy this thing onto there. Nowadays, we're swimming in disks and it doesn't seem like we've got too many, but I think back then it was like, well, disks cost money. You'd reuse them and improvise.

Interviewer: Are there any other ways that you knew how to extend the capacity of your floppy disk, like the flippy disk and notching?

Interviewee: Yeah, probably the most common ones. Certain disks were designed to be flipped. Like the 3-inch drives were designed to be flipped and things like that. But obviously the 5.25, and 3.5-inch ones, you can't do it.

Interviewer: Yeah, the 3-inch ones are interesting, aren’t they? Because the single-sided ones, you could flip them of course and then the double-sided drives came. I find that fascinating because, of course, you can read them at the same time and stuff, but just the whole idea of making it already reversible was really interesting to me.

Interviewee: Yeah. The BBCs often were a bit strange because they have different drive letters per each side of the drive, so each side is independent.

Interviewer: Oh, that's interesting.

Interviewee: Yeah. So zero is the bottom side of the first drive and two is the top side of it, and one is the bottom side of the second drive. So one is zero and two and one is one and three.

Interviewer: Why did they make that decision?

Interviewee: I don't know. Maybe it's because they were single, so they originally were single-sided drives.

Interviewer: Then it would make sense, wouldn't it?

Interviewee: You don't need to fit the disk. You can just assign, just go drive one and put drive one and two and then drive two and three of the other side of the other disks. That means that they appear to the machine as like separate disk drives.

Interviewer: But these types of things are so fascinating to know and that context around them is so important for some of the stuff that we find.

Interviewee: I don't know. I imagine that's some historical thing.

Interviewer: Yeah, it has to be, hasn't it? It makes total sense when I find out, but the floppies drive being of course drive A and therefore on modern computers, it starts with C.

Interviewee: Makes no sense.

Interviewer: Makes no sense. But when I realised that, I was like, yeah, okay. Doesn't make sense. Not really, but it's like, okay, that's where drive A comes from. Yeah, it's interesting how some of that historical stuff stays in place and others just like gets repurposed or redone. It's cool. So you were saying that you don't do that much floppy disk imaging because it's mainly writing it to the actual disks?

Interviewee: Yeah. Obviously we have a lot of disks to do. But yeah, often my job at the museum is to try and get machines working. So yeah, where were we at, I can't remember. But I think most people's first job would be to try and use it. If we've got a drive, try and use a Greaseweazle or something like that. Sometimes there's a machine that's a bridge machine that's quite useful if you're using the real hardware. If you've got that PC that you can connect a 3.5 -inch and 5.25-inch too, you can read that disk and then you can copy that to your 3.5-inch disk and then you can put that in your USB drive.

Interviewer: Which is nice, isn't it?

Interviewee: It's nice. But also with like Macintoshes. The early Macintoshes had their own particular format and the early Macintoshes were single-sided disk 400k, single-sided drives. There's a certain era of them that will read those disks, but they'll also read a PC floppy.

Interviewer: So there are ones that do both?

Interviewee: There's a sweet spot of machines that will read the early disks and a PC disk. So then you can copy your Macintosh disk file onto a PC floppy disk.

Interviewer: That's cool. I didn't realise that there were more ones that did both. Because like you say, that being able to go from Macintosh to a PC floppy disk, makes life sometimes a little bit easier.

Interviewee: Yeah. That's something like a Macintosh SE/30 or something. It's got that superdrive in it, whatever they call it, that will do those earlier drives or if you've got an external drive, you can plug that into it and you can still read the early ones, but it will also connect to PC floppies.

Interviewer: So you were talking about the '70s and the early '80s. There's a little bit of a mess because there's so many formats out there. Are there any machines that are able to go from one format to another? Was that a thing back then? Because I can imagine that and it's just really annoying when you want to swap disks with your friend.

Interviewee: Totally. It probably only happened to a large degree either with PC things because they were so ubiquitous. People were advertising PC compatibility, they wanted that type of thing. I'm pretty sure there was various... well, the Atari ST could read PC floppy disks, 720K PC floppy disks out of the box. I'm pretty sure that Amiga could do that as well. Amigas had their own format. So you could use if you were to get to Amiga into floppy disk. So there was a few, but if you're talking about other things, it's probably only within if the manufacturer made an update. So then they would go, oh, well maybe yeah, you can read your old files because we've added that in. I wouldn't think most other people would bother.

Interviewer: So there wasn't really an interest in that having... only if the manufacturer basically put that into place.

Interviewee: Yeah. I'm not aware of it, but not unless a manufacturer was trying to take business from somebody else or something like that, it's the only thing and I'm guessing now really. I don't really know. Mostly it was like, well, PC, that's what we want to be compatible with. That was already obviously being quite common. The actual compatibility of the file with the machine was not part of the advertising claim.

Interviewer: That's the next problem.

Interviewee: That's the next problem all the way along.

Interviewer: Because that is the big issue for us. It's nice to get a flux stream. It's nice to get a sector image and be able to play it in an emulator, but then the next step is, can we actually get the software in place to be able to open or make these files accessible? That's the next step.

Interviewee: Yeah. I suppose that's the modern equivalent if you may be... I don't know, pick an example, if you had an Amiga and if you had one of the more big box type Amigas rather than the smaller ones, you could get for example, a PC card to go into it. So you could run  PC software and you might be able to use that to transfer files through the hard drive between the Amiga world and your PC world that was also running on that machine. These days you'd have an emulator.

Interviewer: Yeah, exactly.

Interviewee: But then it was, the machine could only do limited emulating. So you had a PC on a card that you slapped in and put in it and that did it for you. So it was a hardware solution.

Interviewer: Yeah. That is interesting that that was a hardware solution.

Interviewee: Yeah, because the machine wasn't powerful enough to emulate.

Interviewer: Okay, that makes sense.

Interviewee: So they put a PC and an expansion card with a PC on it.

Interviewer: That's so interesting.

Interviewee: Yeah, there's some of those at the Museum. We've seen them in the machine, but I don't think we've really got them up and running, but they are there.

Interviewer: Because I've seen a few for the Amstrad online, when you look up, there seems to be certain cards that you can buy. I can't remember. They've got a really distinct name as well that seem to make it possible to go from certain Amstrad stuff to PC. Now it makes total sense why it's a card.

Interviewee: Yeah. But it was literally running. Some of these, it's running in a whole different context and you can just switch over to it and then switch back to your Amiga because they're separate machines running really. The Amiga, for example, was a quite creative machine. It was in graphics and a lot of video work and stuff and so on it. But also maybe they needed to use a PC app, so they had one of these cards. I guess there was ways of transferring data between the two.

Interviewer: Yeah. That is cool. Because that can be maybe a little bit of a solution for some of our stuff.

Interviewee: Other things, there's quite a lot of these hard drive solutions now for SCSI things or otherwise. So if you put a compact flashcard in your PC and copy a file to it, you can then take that card out and put it in a card reader and plug it into your modern PC.

Interviewer: Which is very cool.

Interviewee: Yeah. These type of things. But also with SCSI things as well.

Interviewer: Yeah. I see that there seems to be quite a number of cards that have come onto the market for SCSI interfaces and stuff and more like the last five years, which is very interesting.

Interviewee: Yeah, and they've got more accessible now because hobbyists have got involved in it, they are cheaper and things like that.

Interviewer: Yeah, it's always great when the hobbyist get involved because everything becomes open source and there seems to be a community around it. It's open source, it's cheap solutions to stuff. Which we as a community also very, very much like.

Interviewee: Yeah, we are very appreciative to all of these things, and some of them are very niche as well.

Interviewer: Which I love. Some of them can get so niche, but then there is still a market, there's still people and a community around that and everything, which I really, really like. So when you make disk images, I know you said that you only do that in a small number of cases. Are there any steps or procedures that you put into place to ensure that you get the best disk image? How do you know that you've got the right disk image in the end like that?

Interviewee: Usually, the proof in the pudding is, can we write it back and get the same thing?

Interviewer: That is a good one.

Interviewee: If we can write the image back and then it works, then we can be sure that we've got a good image.

Interviewer: Yeah, that does make sense, doesn't it? Once it works, it's going to be fine.

Interviewee: Yeah, that's usually our litmus test for that. It depends on the value of the data that we perceive to be on that disk. Normally, we'll have a look at it and see whether it's mouldy or something and things like that. Unless we believe that there's something very valuable on there, then we'll just give it a try.

Interviewer: Nice. If it is mouldy, what do you do?

Interviewee: We probably just wouldn't do it, put it to one side.

Interviewer: Yeah. Put it to one side. Even if it's something really valuable that you need?

Interviewee: Well, maybe if we need to read it. But we'd probably do a bit more research on that, which is probably not something I've got into a lot, but there are other volunteers.

Interviewer: Exactly, yeah.

Interviewee: They have done a lot more work with that media that's quite problematic.

Interviewer: It's quite fragile.

Interviewee: But yeah, it depends on what we believe to be on there.

Interviewer: Yeah. Because we get floppy disks that have... now these are actually quite well documented, but we get in a lot of these without any labels.

Interviewee: No labels whatsoever or just a number on it.

Interviewer: Yeah. Or a label that states ‘notes’ on it and that's it. It's like, wow. Great that you've got notes. How would you evaluate how useful or valuable something is?

Interviewee: It would probably be in the context of the donation.

Interviewer: Yeah. Okay.

Interviewee: If we found just a random pile of them, maybe if someone was curious they might read one and see what was on it and see whether that applied to the whole batch. But unless we had a reason to believe that there was something valuable on there, there's just too many to go through and document every single one. Also, as you say, we don't necessarily know what format that disk is in. It amuses me sometimes that we have a box of disks and it's like, these are all just disks. Obviously at the time, someone had an Atari ST or whatever they had and they were like, oh this is disk? They didn't think to write on it, Atari ST.

Interviewer: Would love if you've actually done that.

Interviewee: Historically. There's no reason for them to have done that. But now looking at it, that would've been really useful if you'd done that. Thank you very much.

Interviewer: Would've been great if they just wrote the system or what type of drive they used to create it, that type of stuff.

Interviewee: It would've brilliant. But no such luck, I'm afraid.

Interviewer: It makes sense, doesn't it? Because if I create something, I don't know, it doesn't really matter anymore nowadays, but if I put something on a USB stick and I put a note on the USB stick, I would put what's on the USB stick. I would never be, oh, I copied it over, it was Windows 10 pro. I wouldn't do that.

Interviewee: You wouldn't do it because you know what you're using at the time.

Interviewer: Yeah. It makes sense, doesn't it? But yeah, it would've been great if they actually did that. But it's also just really fascinating seeing what's on them and seeing how misleading the label can sometimes be because they've been reused in some way.

Interviewee: I remember sometimes, if you're using a disk maybe temporarily, maybe you are backing something up, the reason for me anyway, you get a disk with no label on it is because people had stored them in their disk file in a certain place and they were like, oh, this is my backup and I keep all my disks in order when I'm backing things up and that's where they are. But down the line, they've moved around and it's just in a mess. But that's why you end up with things like that. It was like writing on the label was being committing to something.

Interviewer: Yeah. Which is true, isn't it?

Interviewee: It's committing that this is my coursework or this programme or something like that. Where are you going to write it in pencil or pen, depending on whether you're going to reuse that floppy disk or not. 

Interviewer: Because there's also enough of them, and there's probably a few in here as well that have something written on them and then crossed out and then something else written on them. Like you say, it's committing to what's on there, isn't it?

Interviewee: Absolutely. Yeah. You only got one label per disk in your box.

Interviewer: Yeah. You have to use it sparingly.

Interviewee: You have to use it sparingly. That just me, I don't know. They didn't necessarily commit to what they were putting on there because they were going to reuse it.

Interviewer: Are there any other tools or pieces of hardware that you would recommend? Of course there's the Greaseweazle, but is there anything else in floppy disk realm that you would recommend?

Interviewee: The Greaseweazle, that's the primary tool we use nowadays. I believe there's been some other ones around that were like PC card things and stuff, but I don't think there're any better than that nowadays for the usual disks. So yeah, we usually either use a Greaseweazle and you've seen our machine that we have at the museum. It's got all the drives in and things like that.

Interviewer: Yeah, it's very cool.

Interviewee: It's got my computer in it, but yeah, we use it to store drives in and we swap the things around in that. We either use real machines or we use that.

Interviewer: The drives that go into that machine because I've walked past that machine before. Are there specific drives that you would say, get one of those, that brand or avoid this brand or anything like that?

Interviewee: I don't know. It depends a little bit. There are probably ones that we've just found to work all right. Then we've kept them there. 

Interviewer: Yeah, exactly.

Interviewee: Sometimes we've put a drive in there, they look the same, but they're not. So there might be a 40-track drive in there and an 80-track drive in there because it's better to read a 40-track disk on the 40-track drive.

Interviewer: Yeah. Would you say that? Always better to read whatever it was formatted on?

Interviewee: Yeah. Certainly, with the 5.25 inch drives... Actually, writing is better that way around.

Interviewer: Interesting. It makes sense because then you don't have to put in any of the step functions or stuff. 

Interviewee: Well, yeah, because when you're reading back with the 8-track drive, it's doing a double step, but it's got the smaller head. So when you're writing that back, it's only writing a smaller section. So when you're using the 40-track drive it, it's writing it more completely. 

Interviewer: Oh, that is cool.

Interviewee: So yeah, that machines have got 80-track drive and 40-track drive in it and things like that. There are certain drives are terrible for rubber bands, drive belts from perishing and stuff. That's not even to say some drives, the later ones are better, but I certainly know with 3-inch drives, the early ones were better. They were direct drive TEAC units that are like... they work today and don't need anything doing to them. Whereas the ones that came in the Amstrad machines that were cost reduced and made of plastic and they've got drive belts are perishing and they can be much less reliable. So it varies quite a lot.

Interviewer: We've got it as well. We've got an early single-sided Amstrad disk, which is great. I am convinced that I could throw that thing out the window. It's bloody great. Then we've got a newer one, which is a double-sided one and that just been a bloody nightmare. The drive belt went on it and now there's something wrong with the little hatch that goes down, like the little thing that goes down so the reader hits the disk, and it's just, like you said, it's plastic. The other one is metal. It's like all those things where I'm like...

Interviewee: Yeah. It's cost, isn't it? Been reduced. Cost reduced over the years probably.

Interviewer: Does make sense. So the drives can like be a little bit different in quality and stuff. Do you see that in disks as well? Floppy disks? Have you seen some that would last better than other ones?

Interviewee: I suspect there are, I wouldn't want to claim which one was better than others. I would certainly think historically that if I had a branded disk that was a Sony or TDK or something like that, then that would be better quality. Whether that assumption is last in the test of time is a different matter.

Interviewer: For our collections. Yes. It's funny, isn't it? It seems like the brand stuff does seem to last.

Interviewee: Yeah. What you do expect it to do, it costs more money.

Interviewer: Yeah. That's it isn't it? It costs more money, but there's the quality that keeps it. Because you're talking about there's certain disk images that you try and put on or write onto floppy disks. Does the museum keep some type of system or backup of a whole bunch of, I don't know, software disks or boot disks?

Interviewee: It does. I don't particularly deal with it a lot, but yeah, it does have an archive of type of things like that of images. But also like digital things like ROM images and things. So that's another whole different software problem.

Interviewer: Yeah, that's probably the next problem after floppy disk gets to ROM disk and the data tapes. Because that is interesting. You did say you don't deal with it in that, but are they labelled in a certain way that you can easily find the right disk?

Interviewer: Hope so. When Tom comes and sees you, you ask him. He'll probably know more than me about that bit of it.

Interviewer: Cool. I'll ask him about that.

[Unknown A1] Because I can imagine that the museum has some really interesting boot disks or software stuff in that collection which are probably quite valuable and quite useful.

Interviewee: Yeah. As volunteers, we try and maintain a collection of disks that are useful for machines to work with.

Interviewer: Just out of interest because it's not that... how long does a floppy disk live for if it's constantly in a computer?

Interviewee: Sometimes a long time and there are machines in that gallery that are running a disk every day.

Interviewer: And they're fine?

Interviewee: They are fine.

Interviewer: It's so funny.

Interviewee: Every single day, it boots up from that floppy disk and runs it.

Interviewer: It seems so random at times why some stuff fails and other stuff doesn't fail.

Interviewee: Yeah. There's no one or more reason to it.

Interviewer: Which is funny, but also would be nice.

Interviewee: Interestingly, again and maybe there's an issue with reading and writing because I've been talking to some people who were software developers back in the day and this is just one particular anecdote and they were using Tatung Einstein with 3- inch floppy disks. But you could connect to that machine a 5.25-inch drive or 3.5-inch drive and they would... I think they were using the 5.25-inch drive and they would say, oh, we saved all our data so that we plug in 5.25-inch drive because it was more reliable than that 3-inch drive.

Interviewer: Interesting.

Interviewee: These machines were off... they were hobbyist machines, but they were often used back in the early '80s as a machine to programme on for more popular other machines. So they were using 5.25-inch drives externally to the machine as a storage medium because they believed they were more reliable than the internal 3-inch drive that was in them. But I know that that is one of those machines sat at the museum that it boots up off that 3-inch drive and runs and loads and works perfectly fine. But maybe that's because they were saving things more to that drive and overriding things and doing things with that disk that maybe the other one didn't work so well.

Interviewer: Yeah, interesting. Like not the older drive, like you were saying, the newer older, but the newer 3-inch drive is not as reliable as our older one. But I would say that that was one of the most reliable drives that we have. But like you say, we only read from it. We don't do any writing from it. So that's maybe where it...

Interviewee: Yeah. In that particular machine, those 3-inch drives are high quality TEAC drives with direct drive and there's belts and all business like that and they're generally very reliable, but that was their experience.

Interviewer: That's really interesting, how different that can be for different people. Interesting how they use the 5.25 external drive then as a solution to that problem. Because that is fascinating, actually. But what do you do, you're not going to save your disk if it's not reliable. You're going to find a solution to it.

Interviewee: Luckily, the machine had a disk controller in it that you could just plug those in and they worked and it was fine, but that was their solution.

Interviewer: Yeah. Very cool. Then talking a little bit about troubleshooting and stuff, how do you go about, say one of... I think I already know the answer, but say a floppy disk drive in the computer that has all of a... say one of them dies or is not reliable anymore. What would you do? Would you try and fix it or would you just go and find a new one?

Interviewee: Well, if you've got a live selection of them, change it out. But no, I think the first thing everybody always tries is they always do a head clean.

Interviewer: Do a head clean.

Interviewee: Do a head clean and reasonably good success with head cleaning diskand secondly dismantle it and to clean the heads.

Interviewer: Because that is something that I've learnt by experience, but there's still ones where I'm never really sure. How do you know what is failing? So how do you know if it's a Greaseweazle that's actually not... how do you know, not that the Greaseweazle acts up, but maybe your commands slightly off. How do you know if it's the drive that's acting up or how do you know if it's just a bad floppy disk.

Interviewee: Some of it's depends also on context. Sometimes you listen to these things, don't you? You see how they work and if you have another one that... or often with everything where you're repairing them, if you've got another one that works, this is the best thing, isn't it? If you've got one that works and another one that doesn't, well, you can see why that other one isn't working. But sometimes, you've only got one of them and you have to just deduce that based on your experience. Normally, if you had experience that the drive was working fine and then all of a sudden it stopped working fine, but it seemed to be working and you're getting a bad data read, you could probably assume that one of those disks that you put in there shred a load of muck over it and it needed a clean, but otherwise it was okay.

Interviewer: Yeah. But that's good. I don't know if you've reached that, probably you have, but it's like a weird party trick to have. But now I can listen to drives and be like, oh, that's not reading correctly.

Interviewee: Yeah. You can hear it. Sometimes that's just the way it's stepping and it's not regular or you can hear it retrying or things like that, you get to know.

Interviewer: But sometimes you can just hear it drag the dirt across the...

Interviewee: Oh dear.

Interviewer: Oh no.

Interviewee: So yeah, that's the first one. Grease in those drives is probably a problem as well. Can dry up and be hard. These are similar type of problems, similar to any mechanical thing.

Interviewer: It's very similar problems that show up.

Interviewee: Similar type problems. Perished rubber type things and they're not just drive belts, but in stops and things can be made of rubber and they can perish and be bad. They're all similar things. Certainly other things like less critical, but eject mechanisms can get sticky or if you've got fancy Macintosh drives and gears can break and stop the eject working and things like that. They're the mechanical things. I suppose apart from cracked plastics and rows and stuff, they generally, they're just stuck up and they just need a clean.

Interviewer: Which is nice, isn't it? Compared to my mobile phone nowadays that I can't even open anymore, these things are quite mechanical, like the floppy disk drive. So I wouldn't say every repair is easy, but a lot of them are quite straightforward. Like you say, it's a little stopper that's gone. It's the belt that's gone. None of it is rocket science in that sense. Whereas a lot of stuff nowadays, it's a lot more intimidating to try and fix it. Which is really fascinating.

Interviewee: Yeah. Past those mechanical things, if it was an electronic problem in the controller, it would need to be something special to warrant investigation in that area.

Interviewer: But it's cool. It's really nice.

Interviewee: Yeah. Normally if we thought the controller was broken, then we'd just swap it unless it was something desperately needed for a machine. Then it's not about reading the disk, it's about having something that works with a particular machine for preservation.

Interviewer: If you swap a part for... so say it's like an easy fix, do you tend to get a new part? So for us, because we only have a set number of drives, it's just going onto eBay and finding a specific belt or a specific thing. Do you do the same or do you go into the collection and find maybe a drive that's broken or something else and swap parts?

Interviewee: Probably a bit of both. So it depends a little bit on what it is. If it's a drive belt or something, you're better off buying a new one because you find one and you might replace one, have another drive for speed, but it's still going to have a lifespan of six months. But if something more bespoke is broken, then you might well be worth trying to make good one out two. We talk about yes they are mechanical, but also there they have been made in an environment where a lot of alignment of disk drive heads and things and try not to fiddle with things like that. It's going to make problems you can't fix.

Interviewer: That seems to be one that's broken.

Interviewee: Yeah.

Interviewer: I have seen people online that have fixed them, like the misalignment, but for me, that's one where it's probably just easier to get a new drive.

Interviewee: Yeah. It's too much. You also might end up in a situation where you might have aligned it and you might be able to write and read a disk on it, but it won't read anything else because...

Interviewer: Yeah. So only for that specific... because that is something that I've seen mentioned quite a lot as well, where people have written stuff on misaligned disks and then for us to read them, it's just really difficult because it's on a misaligned disk.

Interviewee: It worked perfectly for that, but yeah, nowadays it's just out.

Interviewer: Yeah. Which I think is also fascinating and a whole other world of trying to get that to work. Another interviewee actually mentioned that, that sometimes the first track, which is of course the most important track because there's all the information on it can be so misaligned that you actually want to go back a track, so you want to go to minus this one. I find that really fascinating. But if it's really crucial to be able to read that disk, then something like that can be really, really helpful. But it's mainly just a pain. So continuing on to the last section around knowledge sharing. Because I know that you're of course like part of the community at the Centre for Computing History. What happens if no one there can answer the question? Are there other communities that you would talk to?

Interviewee: There are other communities and things. There's also a problem that sometimes this knowledge is A, locked up in people's heads and and B, even when it's written down, it can get lost. So I think we think that everything's posted on the internet is there forever and it's not necessarily the case. But like you say, a lot of times you search for it and it was on some forum somewhere and someone had looked at it and many of those have vanished over time and that information is lost. I tend to find nowadays a lot of this, what there are still these forums around, but some of that is now transferring over to Discord platforms and things like that. That's even worse because you're not on a certain Discord, you can't really search it and it's great to ask somebody and to get your question in front of somebody, but the answer is lost. Often when you've fixed your problem, you've forgotten about it and...

Interviewer: Yeah. It seems like it's just a repeat of the same information, isn't it? Because you could come around one week, post your question on there and it gets answered and then maybe six months later on a very similar forum and someone asks the exact same question... or on the Discord server, someone asks the exact same question gets answered again. But there's no interoperability or there's no searching.

Interviewee: So yeah, it depends where these communities exist, but they're mostly online now.

Interviewer: I hadn't thought about that. Yeah, I was aware, of course, the forums are starting to disappear and stuff and that in itself I find a really interesting project trying to save that type of knowledge and the web archiving and that type of stuff. But then what do you do about Discord servers? Because how do you get that? How do you capture that material? Because like you said, some of them are really difficult to search through. You have to know about them.

Interviewee: You have to know about them. Exactly.

Interviewer: Yeah. You have to know that sometimes you're probably the same. It's like you chat to someone about something and they're like, oh, there's a Discord server for that. But it's not the same, isn't it?

Interviewee: It can easily get lost and has done so.

Interviewer: Yeah. Which is not great.

Interviewee: It's really good for putting you in contact with those people. But yeah, the information itself is a problem.

Interviewer: What in your ideal world which you would you like to see in sense of that type of information?

Interviewee: I don't know. I think there needs to be some like massive wiki somewhere.

Interviewer: With all the information?

Interviewee: Some people maintain these. For certain computers, there's lots of information out there and things. But some of it is so bespoke and weird. It's like, where'd you put it?

Interviewer: Yeah, exactly. It can be really niche problems and then it's a bit like, do you create another page on this wiki for it? It doesn't just live here. There's also something when I started using the Greaseweazle, there's a lot of language around floppy disk imaging. It's very niche. Some of it is very niche. Now, I get it because I've been looking at it for far too long, but I can remember looking at the documentation the first time just being like, this could be in a different language and I would understand just as much of this. So I find that really interesting as well. People coming into the communities or being interested in it just being a little bit like, what am I supposed to do? So yeah. Whereas like places like a Discord server are really helpful in that sense because you could ask a question and be like, where do you start? What does this mean? Again, that information is a bit lost, isn't it? Because it disappears with that person answering that question. Solution for Discord server should maybe be found?I think that were all my questions. I don't know if there's anything that you were like, oh, we haven't touched upon that and I would like to mention that.

Interviewee: I don't know, we've covered a lot of stuff.

Interviewer: Yeah, we have. We've covered all the floppy disk stuff.

Interviewee: All the things. Yeah.

Interviewer: But if there's nothing, I'm happy to stop the recording.

Interviewee: Unless, anything else you want to ask?

Interviewer: No, I don't think there's anything specific. It's just been really, really helpful. It's really good to talk to different people about floppy disks and how they... and I know that our use cases are quite different in the sense that you're more interested in writing floppy disks than...

Interviewee: Yeah. I was just thinking as you talk about image formats as well, we came across some quite recently and it's like some of these images had been made historically on some old format. This was a Macintosh thing. So these were files that we could only read on a tool, on a programme that was running on a Macintosh. So this was where we needed real hardware to make it work because we had to use the disk image. We had to find the bridge back to the bridge machine, which was like, some find that tool that could run that bit of software.

Interviewer: That can be really tricky, isn't it?

Interviewee: Yeah. Also reading the files as well. So you can copy that file to a machine, but how do you read it? Especially if it's compressed or something like that. Again, sometimes a real hardware or the emulator is the best way to shift it off.

Interviewer: Yeah. It seems like for us that an emulator... was the Amstrad stuff, the emulator has been the best. There's a number of niche tools out there that can create or that can basically get the hex values and know exactly where file start and Amstrad machines and stuff because first off we were like, oh, we can do that ourselves. But there's quite a weird interleaving on Amstrad disks which is awesome, but not very good when you're trying to determine where the specific files are sitting. But yeah, emulators seems to be the way to go for a lot of things.

Interviewee: Yeah. How do you deal with unknown files that you don't allow? Sorry, I'm asking questions now.

Interviewer: No, but that is good. How do we deal with unknown files and we don't know what they are? Currently, we just preserve them and hope that something will come along that will help with that. We've got a number of floppy disks that we do not really know the formatting of. Currently, we just have a flux stream of those. So the low level reading from them. I think that's the best that we can do. But I'm a little bit worried about the fact that there's a little bit of a... just created this image and then, I don't know, in decade or two decades along, we may be able to read that stuff. I'm a little bit worried about that because the knowledge of the people who know how to read it are slowly retiring and passing away and I don't think waiting two decades...

Interviewee: Too late.

Interviewer: It's maybe a little bit too late. So it seems more of a, we should get a disk image and do our best now to try and read it instead of hoping that in the future there will be a tool that comes along.

Interviewee: Yeah. Some of this comes back to there's other clues, isn't there? If someone says there's a label on the disk saying, well, it might be, or there's a file extension or something. That can give you a clue, but otherwise you could be completely stuck.

Interviewer: Yeah, exactly. Then you've got on some of the old floppy disks, file extensions, they aren’t there yet.

Interviewee: Yeah, there's not there.

Interviewer: They didn't even use that. Then of course you can check the hex value and that type of stuff to see if you can get any of that information. But that's also not a guaranty because...

Interviewee: No, that's probably what I would do open it up and see if there's any clues in there.

Interviewer: Yeah. So it's fun, but also I do worry a little bit about that because it means it would be such a shame, especially the '70s, some of that stuff would be lost because we didn't spend the time trying to get the context around it.

Interviewee: You end up with a lot of data and you don't know what you're looking for. You're looking for a needle in a haystack often. Unless you have access to the data then maybe there's ways of searching it or passing and looking for things.

Interviewer: Can you imagine going through all the trouble of trying to read this stuff and then you can finally read it and you find out, I don't know, it's someone's shopping list. It could be anything. It makes it fun, but also it's a little bit of a worry. But then on the other hand, I would still not say that that's wasted time because you still have discovered how to basically read those formatted disks. So in a sense, maybe someone else may have use of that.

Interviewee: So the process the has been useful in its own right.

Interviewer: Yeah. So at least that's good. 

Interviewee: Yeah, I want a machine where I can just put a stack all those disk in, it reads them all and then I can search to see if anyone's put the answer to life universe and anything on them.

Interviewer: Exactly. That's what we want to know. Is there anything that we should be keeping from this that would be... well, if you could go and build that machine, that would be great. But yeah, I'll stop the recording there, but thank you.

Interviewee: You're welcome.

[END]
[Unknown A1]I probably should be suggesting you order Tom around :)