Olia Lialina & Dragan Espenschied
Cory Arcangel: You’re all Linux now? Should we start there? Is that okay to start?
Dragan (Drx) Espenschied: I think so. We are almost full Linux now.
Olia Lialina: Yeah, we’re full Linux, but we have to run virtual machines.
DE: With Windows installed.
OL: With Windows because—you know why?
CA: No.
OL: It’s tragic …
(Laughter)
OL: We didn’t want to see Windows ever again.
DE: Ever again.
OL: But then we realized that MIDI files, they don’t work in, um—
DE: In Linux. So if you got to a webpage with MIDI or you just want to replay a MIDI file, it’s not easy. Once I managed to make it work on Olia’s computer, but then there was a system update, and then it didn’t work anymore (laughs)—a little too much work, so we use a virtual machine with an old XP.
OL: So you can forget about it in normal everyday Web life, but if you’re going through the GeoCities archives and it’s all so silent, you understand that you’re missing half of the performance, so we now surf through the archive with a virtual machine …
CA: Wow, so like, so in your daily life, you’re on Linux, but when you’re doing your research you have to fire up a kind of virtual machine?
DE: Yeah.
OL: Yeah.
CA: Oh wow.
DE: Well it’s more authentic to the environment—the people who created these webpages, what they were using …
CA: Right, right. Does that feel like going to work, when you fire up a virtual machine?
(Laughter)
DE: Sometimes. It’s definitely another thing than just typing in the Web address, you understand, but it’s also maybe like putting the needle on the record instead of just double-clicking an mp3. So, it’s a kind of ritual that sets you in a certain mood. I don’t really think it’s work, but it adds to the quality of what you feel. It prepares you, and you’re definitely going to an environment that has less trouble because you’re not on your usual computer, so you’re shut off from your usual email, from the usual chat, and it’s a special environment to look at the old stuff.
CA: To look at the archive. Yeah, and so, let’s talk about the archive. I guess the one thing that I’m curious about is—let’s say you’ve just fired up your virtual machine. Where do you start? Or what is your thought process?
OL: We have two different strategies.
CA: And first I should explain—or do you want to explain, just real quick? Because these interviews have no intro, they just start. So do you want to explain, like just give a little background for people reading, just about what we’re talking about.
DE: Okay, cool. We’re doing cultural research on the GeoCities archive. And GeoCities was a Yahoo! hosting service that, originally, Yahoo! bought it from another company. And it was very popular among Web amateurs to just host there. They created webpages there. And in 2009, Yahoo! announced they will delete all of this, because apparently it didn’t make money for them and they didn’t see the value of it. So they set a deletion date and—GeoCities was already some kind of laughingstock on the Net because it just contains amateur stuff and nothing professional, and nothing 2.0. It had MIDI background music and all this GIF animation. But some people realized that a lot of culture was actually lost there. And the Archive Team, which is a coordinated, grassroots … kind of underground, activist-archivist group, copied as much as they could from GeoCities before it was closed down and released it as a bit-torrent download. We are still downloading it because it’s a lot of stuff, but what’s already there, which is almost everything—we’re using it because we’re interested in the amateur webpage user tools.
OL: We’re going through it and we are blogging about it.
CA: In the virtual machine.
OL: We go in the virtual machine, but we blog about it outside.
(Laughter)
DE: Yeah, sometimes I blog in the virtual machine and copy and paste from the virtual machine, and then the host system doesn’t work again, which happens a lot. (Laughs)
CA: So, Olia, when a virtual machine has just been opened, what happens next?
OL: Yeah, what happens. In Dragan’s computer, what happens is that he hits “random” and goes to random pages, which is very smart, but it’s only one way to go through it. I also have this script installed, which can show to me random pages, but I really like to go to a particular neighborhood and then see the lists of the hundreds of suburbs in the neighborhood, and to go through them one after another. It creates a picture about a particular neighborhood. Do you understand what I mean, what was making it very special?
CA: And when people were creating these pages, how did they determine the neighborhood their pages were placed in? We should also mention to the home audience who is un-familiar with Geocities, that neighborhoods showed up in the domains, http://www.geocities.com/SunsetStrip, etc, etc.
OL: The neighborhood was determined by—
DE: Beverly Hills Internet.
OL: Yeah, it was, um—
DE: That was the company, that originally created GeoCities.
CA: What a great name.
DE: Yeah, it’s awesome.
(Laughter)
DE: Yeah, they had this idea that there should be a map of the city, and you choose your neighborhood, you choose your street.
CA: Oh, so when you signed up, there were pull-down menus?
DE: Yes, and all the different neighborhoods were used for different topics, so there was an Area 51 for Science Fiction fans, and there was Pentagon for military people, and …
OL: Or HeartLand, for Genealogy, and children and … But it was early history because after Yahoo! bought GeoCities in 1999 they changed the system to vanity profiles.
DE: Also, Yahoo! usernames.
CA: That’s what I remember, yeah.
DE: So, maybe—
OL: Neighborhoods, they were —
DE: — they were more focused on the person, on the single person than on the community.
CA: Yeah, totally. And Olia, when you’re going through each suburb of each neighborhood, are you just clicking on each one, or are you randomly going through?
OL: No, I try to go to from the list (Editor’s note: points to a director listing of the archive on screen).
CA: Whoa.
(Laughter)
OL: Because then you can see that it’s linked, you see that it’s purple, so it’s done. Of course you go away, you go somewhere, then you follow the link … It’s a special pleasure that you can go back and then go to the next one … to see this list, it’s great.
CA: And where is the list, is the list generated by Yahoo!? In each suburb?
OL: It’s in the archive. You go—I don’t know how it was originally, but it was not like this when it was online, but now on the hard disk, you can—
DE: Yeah, the website generates an index, which can take five minutes. So, we also saved the generated index page as an html file so it doesn’t take forever to generate it.
CA: So what are you looking for, when you go through each? Tell me what the experience is like, just clicking on page after page of like forgotten … This is like a ghost or something.
OL: I sometimes really have a purpose, I know what I want to find, like I look for Felix.
DE: Or what you want to verify … you have an idea—
OL: Not necessarily what I want to verify, but I realized it was just good luck that I started to go to the Pentagon Pages and then I see this, oop! there that Felix is there, the image, and then two pages later, again Felix is there. And then I stopped looking for him, but of course I found a lot of other stuff on the way, but what if I’m looking for him?
CA: Yeah, so you’re surfing with a purpose, kind of. You have a hunch now that maybe people in the Pentagon neighborhood are using Felix, and you’re kind of also cataloging other information, but you’re mainly looking. Maybe it’s like these things are being presented to you, the information is being presented to you and you are just trying to identify what’s important.
DE: Yeah, of course it’s a lot always, but you have to really make notes on what you encounter and to not forget about it because there’s a lot of interesting things.
OL: Yeah, that’s why we started to blog about it immediately. And wanted to—because—yeah, it involved even the smallest finding. We wanted to write about it immediately and to have a smart, helpful tag system so that we can really find all this later, and on the way we are notifying our text so that we really make something become narrative.
DE: Because we can’t use Delicious for that anymore, because it will be shut down soon, we had to actually write our own tagging plug-in for the blog software. Because there was never any tagging as good as Delicious, and so we made our own plug-in for WordPress that—
CA: Really?
DE: —is supposedly the same combination and same effect and …
OL: Yeah, we’re so happy about it.
CA: Really? Did you release the plug-in?
DE: Not yet
CA: It sounds amazing! Because WordPress works in categories, right? Which is really dumb.
DE: It has tags, but the combination is just very simple, and almost useless. But that’s most tagging environments. Only Delicious had gotten it really right.
OL: But you know what was the funny thing, we thought that we were not so ambitious. We thought, “Okay, each new animated GIF is something small, something curious,” but then you know, each find was kind of sensational—
DE: — it’s just touching, it’s such an epiphany that you finally understand something, something that we only had anecdotal evidence before of, it just proves itself again and again and it can be a—
OL: We also need this blog to archive.
DE: Because it goes into a lot of directions at the same time and we hope to be able to come back.
CA: It’s like a scratch pad.
OL: Yeah, otherwise I don’t know how would we organize all this—otherwise, it would be just, I mean, to collect the links and just be postponing, all of this would die.
DE: Yeah.
OL: But with tags it’s good.
CA: Yeah, and how does it work, how often do you do this?
DE: As often as we can.
OL: It’s actually, right now there is a routine—
DE: The daily routine.
OL: The daily routine.
CA: Yeah, tell me the daily routine.
OL: After we put the kids to bed—
CA: 7 o’clock, 8 o’clock?
DE: At 5 (laughs). No, At 9.
OL: We start at 9, we have two hours—or one hour—to go through there.
CA: So, both of you do it each on your own computers?
OL: Yes.
CA: So, Olia is systematically going through neighborhoods, and Dragan, you’re using your random script, which—what does it do, just look for index.htmls, or how does that work?
DE: Yeah, I used UNIX find to create a list of index.htm or .html with the regex-i option of find and it’s just a don’t-know-how-many MB text file and this random script that just picks one of them and shows it.
OL: We now have a competition of our methods, not only in this, but also in—how do you say—our proof-checking?
DE: Yeah, fact-checking.
OL: Yes, fact-checking. I make my conclusion, like once I wrote that this is the most used border in the Pentagon neighborhood because this is my impression.
DE: And then I want to check if it’s true. And then we decided we needed a way to prove it.
OL: You decided.
DE: I decided? Because it’s—and so then I—
OL: He just wanted to program something.
DE: I love programming.
CA: When I put this on the web, I’ll make that phrase blink.
(Laughter)
DE: Or Marqueeing.
OL: Instead of blogging, Dragan would write the program.
CA: So, Olia, you’re blogging?
DE: No, we are both doing it. I picked it up again because the program is finished now.
CA: So tell me what this program is.
DE: It’s a database of all the HTML pages we have and all the images that are on these HTML pages and the way that you can visually compare the images.
OL: I don’t like it. (laughs)
DE: What do you mean you don’t like it, you use it!
CA: So, tell me more, this is fascinating. This is crazy, you’re writing custom software to check theories about GeoCities neighborhoods?!?!
DE: Yeah, so there’s, for example—
OL: You ruin my theory! (laughing)
DE: It’s not ruining it, it’s progressing it for humankind. So, there’s um—For example, we have Felix. So you can’t just look for the icon “Felix.gif,” because people give different names to them. And you also can’t search for the size because the size changes and all of these things are problematic, so there are three ways you can look for them, for pictures that are familiar. One is that we extract bit by bit the same file, because all the images are indexed in that way, so it’s a checksum. And the other is to have it visually compared, by making a smaller size of the image, really very small and then you have a small pattern and you can match this pattern with the others, but then still, if there’s a small modification it doesn’t work. So there is the same problem with reduced accuracy—
OL: Nobody needed it. You see? It ruins it— (laughing)
DE: Yeah, in ten years. I’m not interested about tomorrow, I’m interested in ten years.
OL: Definitely you can find something—
DE: Now I can do a checksum reduction error rate. And error distribution, like this—and then Olia can take the URL of a picture she’s found and paste it in this interface and then it will show all the pictures that look the same.
CA: Okay.
OL: I can’t imagine doing this
DE: Yes, I’ll do it for you and give you the summaries.
CA: So Dragan will respond to your earlier theory about borders in the Pentagon neighborhood and say, “Actually, these are not the most used borders in the Pentagon neighborhood, these are the most used borders.”
OL: Yes this is what he wanted to do.
DE (to Olia): I didn’t do this, instead I presented all the instances that this dancing girl appears, and all the instances that Felix appears, and not to annoy you. Have you noticed I didn’t make a note of all the most used borders, just because of you. Because I respect you.
(Laughter)
CA: No literally, this is totally fascinating. We were saying the other day, you ask a painter, “Where do you get your paints?” This is literally the same kind of discussion. How do you determine—
OL: I get my paints by surfing.
CA: You get your paints, by, like, your intuitive knowledge and history, and, you know this stuff internally, right, and Dragan comes and he checksums it for you.
(Laughter)
DE: I think it leads to a different kind of findings, and they are both valuable and I think the combination is the interesting thing. It would be not as rich if we only had one approach. For example, I find ten combinations of Felix with a different other element and then Olia can say “Aha, so it probably means that”—or, I’ve seen it completely in a different way—or the other way around.
CA: Can I ask, what—this process that we’re talking about … What do you consider yourself when you’re doing this? Is it terrible to say—is it research, is it fine art, or is it in between, or … I hate, no, this is not a very—
DE: No, but it’s interesting.
CA: I’m wondering if you’re studying humanity, really. I guess, right?
OL: True.
DE: In some ways.
OL: Anthropological research.
DE: What I am personally interested in is the narrative potential of all of this, because the people that created these pages had no other means to create representatives of themselves in a quite complicated way, while today, you fill out a form on Facebook that “I like this, and I like that, and this is my picture,” and they have to really put forth how they represent themselves and how they express it and what they express and it’s, uh—
OL: I know what you want to say. Yeah, that all you want to say is … that the elements that they usually considered are now just all together as a sign for early Web or amateur Web. That they are not only scientific, but they have a lot of meaning and meanings can be very different.
DE: It’s like hieroglyphs, I think. I know you don’t like metaphors, but I think that’s okay.
CA: You don’t like metaphors, Olia?
DE: Oh, we both hate them.
(Olia laughs)
CA: Do you mind elaborating? So, when we were talking before about you guys using paints as your scripts, that was a metaphor, so should we cut that out?
DE: No, no, that’s fine. It’s just that in this case it’s relevant because it compares an artistic practice that’s very established, and where people think they understand what’s important about it, and they will look for the kind of points that are representative of this in the other so that’s okay but—-
OL: Do you know this part of our blog called “Car Metaphors”?
CA: Yes, yes.
OL: I started it because it was already unbearable that everything that happens in the digital world of computers is always explained by comparison to something in the analog world. And especially I brought up “Car Metaphors” because they always compare computers to cars in order to explain something which is much more complicated.
DE: Or something completely different.
OL: Yeah. And that’s why—that’s okay, but maybe we are sensitive to the use of it, maybe over-sensitive.
DE: Yes, I think we’ve already developed a pathological condition for this. So if someone says, “Yeah, but if you copy this file, it’s just as if you leave your car doors unlocked, or if you flip through this like you flip through a book, and you want to read your Twitter feed like in the newspaper.” And it’s outrageous.
CA: They have that now.
DE: It’s outrageous, and it’s um—
OL: I always try to find a way to explain something without—
DE: We’re actually looking for a very native, very authentic core somewhere, which is also probably why we are doing it with GeoCities. I’m personally very interested in the native way of how to tell a story online, or how to express something online, which is not uploading a video or just writing, like, a diary or a book. Online there is an actual—there is a native core, and it makes some things very strong if it happens—if this set of expressive behaviors or whatever is used, and it’s probably like this in any medium: if you work very close to the medium, it can be super powerful, and in GeoCities it is very often like this, because at that point it was not possible for many people to transfer a video to—or other foreign things.
OL: Yeah, that’s a great thing that was not possible in the very beginning, when the Internet was slow and computers were not so powerful.
DE: Yeah, imagine what would happen if the Internet would have been so fast, like now, when it was just beginning. We would have a second TV. Everything we understand today about the Web comes from this time, because of the problems, so to speak, that are seen as problems. But this was the power of it.
CA: Okay, one more question. Is there a “done”?
OL: What?
DE: When we’re done?
CA: Yeah. Because you’ve been doing this research for almost ten years, right?
OL: Do you mean when we’re done with the early Web research?
DE: No, how we will know when we’re done.
CA: I feel like I may be falling into the trap of considering this as an artwork.
DE: That’s okay.
OL: We were almost done, but then there was just this archive—
DE: It was also sometimes very painful to work because the material was so difficult to get. And you had to have an epiphany somewhere—or a serendipity, not an epiphany—a serendipity online, and you had to sometimes spend hours surfing just to find something small … But now it’s the other way around, because you have so much material, and then inside there are links to things outside that still exist online, and that are not database-driven template profiles.
CA: So the world came to you, in a weird way. You had no choice.
DE: Yeah.
CA: Maybe that’s a good place to end it.
DE: We had no choice.
CA: Do you feel happy with that, do you want to add anything?
DE: Yeah, but it’s true, I really feel like there’s no choice. Because I can’t choose to say, “Okay, today I will not—I don’t care about it anymore. I’m—” (laughs)
OL: Okay. Good.
CA: Good. Thank you.
OL: You didn’t ask us, “When do you wake up in the morning?” (laughs)
CA: Oh, do you want me to ask that?
OL: No.
CA: I got to, “When do you start working?”
DE: Yeah. It is in the evening. We are evening people.
[Editors Note: This conversation took place in April, 2011 
]