YouTube versus the National Film and Sound Archive: Which Is the More Useful Resource for Historians of Australian Television?

This article compares YouTube and the National Film and Sound Archive (NFSA) as resources for television historians interested in viewing old Australian television programs. The author searched for seventeen important television programs, identified in a previous research project, to compare what was available in the two archives and how easy it was to find. The analysis focused on differences in curatorial practices of accessioning and cataloguing. NFSA is stronger in current affairs and older programs, while YouTube is stronger in game shows and lifestyle programs. YouTube is stronger than the NFSA on “human interest” material—births, marriages, and deaths. YouTube accessioning more strongly accords with popular histories of Australian television. Both NFSA and YouTube offer complete episodes of programs, while YouTube also offers many short clips of “moments.” YouTube has more surprising pieces of rare ephemera. YouTube cataloguing is more reliable than that of the NFSA, with fewer broken links. The YouTube metadata can be searched more intuitively. The NFSA generally provides more useful reference information about production and broadcast dates.


In 2001 I published a book called Australian Television: A Genealogy of Great Moments.
It was an attempt to write a history of Australian television from the point of view of the programs rather than the industry, legislation, or audience. At the time I wrote about the difficulty of establishing an archive of television's programming. I noted the lack of governmental institutional support for such an archive-in the sense of those institutions whose job it is to remember and preserve culture (museums, archives, etc.;McKee 2001, 3). The lack of support in these governmental institutions had a flow-on effect for individual researchers. Without access to the material it was difficult for researchers to discuss television programs in detail. We could talk about their production, the legislation that governed television, the staff involved in making it-but the programs themselves remained an elusive part of Australian television history.
In writing that book, I visited the National Film and Sound Archive (NFSA) of Australia. This institution is the national audiovisual archive, playing a key role in documenting and interpreting the Australian experience. . . . We collect, store, preserve and make available screen and sound material relevant to Australia's culture. . . . We are leaders in the preservation, presentation and development of screen and sound culture in Australia (NFSA 2009) Staff at the NFSA were incredibly helpful with my project. But the institution was not strongly geared up to support television researchers. Many television programsparticularly those in "ordinary" genres such as lifestyle programming and game shows (Bonner 2003)-were not held at all. Of those which were held, many existed only as preservation copies, and it was impossible for a researcher to view them. Even when material was available, the access provisions were not outstanding. One either had to fly to Canberra to view programs or have them shipped to a local state library for viewing-a process that takes several days.
The situation with popular institutions was somewhat different. Indeed, "the Australian media . . . are engaged in a daily process of remembering some television programs while forgetting others" (McKee 2001, 6). But even those institutions did not archive the programs themselves. They remembered elements of the programs in other forms: clips in retrospective shows, written descriptions in newspapers and magazines of well-remembered moments. Again, the texts of the programs themselves were difficult to access. How things have changed. A number of technological innovations have since revolutionized the archiving of audiovisual material. These include DVD box sets, bittorrenting, streaming video web sites, and YouTube. It is this last technology that I address in this article. For there now exists a popular institution that both remembers television programs and makes the texts available to researchers. This article compares YouTube to the NFSA as sources of material for researchers of television history. To study this, I returned to my research project of 2001. At that time I identified seventeen programs that represented "great moments in Australian television." All were Australian produced. In Melbourne Tonight was an early variety show . The serious current affairs program Four Corners began in 1961 and is still running. Homicide, a cop show, was Australia's first successful prime-time drama series in 1964. The Mavis Bramston Show was a satirical comedy program . The children's program Playschool, based on a British model, started in 1966 and continues in production at the time of writing. Number 96 was an outrageous soap opera stripped five nights a week . Paul Hogan's cigarette adverts for Winfield ran on television from 1972 to 1976. Countdown was a youth music program with an antiauthoritarian air . The Sullivans was a historical soap opera set during the Second World War . Prisoner (named Prisoner: Cell Block H in some territories) was a soapie drama set in a women's prison . Sale of the Century was the Australian version of an international game show format . A Country Practice was a medical series with a soapie edge . Return to Eden was a melodramatic series that took the aesthetic of American prime-time soaps to the Australian outback (1983). Neighbours is a nightly suburban soap opera, one of Australia's biggest export successes, and has been running since 1985. Burke's Backyard was a gardening-based lifestyle program, one of the innovators in the genre (1987)(1988)(1989)(1990)(1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004). Live and Sweaty was a sports entertainment program that heavily featured female talent . And Frontline was a workplace sitcom set in the production office of a commercial current affairs program .
For each program I identified a single "moment" that stood as a synecdoche for what was interesting about the program and how it was recalled in popular memory. For example, the "moment" I discussed from Prisoner came from the first episode when Bea crushes Lynn's hand in the steam press (McKee 2001, 166). This moment is well remembered in popular accounts as emblematic of the program and allowed me to talk about the way in which Prisoner was based around an aggressive, physical form of femininity that had rarely been seen on mainstream television (and has rarely been seen since). Popular memories of television tend to be organized around such "moments" (Goldberg and Rabkin 2003, 23-24). But by using these moments as examples of synecdoche, I did not want to leave my analysis simply at the level of anecdote (see Hartley, Green, and Burgess 2007, 28). Rather, I wanted to use them as a way to link popular and academic histories of television.
For the current project I visited both the NFSA web site and YouTube to see what program material from my list of seventeen "great moments" I could view through each. I also took note of what other content from the programs was available to view. During the comparison I made notes on what material was available, how easy it was to find, and other issues that arose in the course of searching for these examples of programs from the history of Australian television.

Different? Or Worse?
There exists an emerging tradition of writing on "democratic digital archives" such as Wikipedia and YouTube (Gracy 2007, 194), much of it in the disciplines of librarianship, museum studies, and information science. This academic writing compares these digital democratic archives to traditional sources of information such as encyclopedias. Within the tradition there are two key approaches. Both approaches find differences between the new archives and traditional sources of information. But they evaluate those differences in different ways.
The first-more common-approach takes traditional sources of information as the standard against which others must be judged. Any differences between the digital archives and traditional sources of information are taken to demonstrate a lack on the part of the new sources. The archives are evaluated using criteria such as "information quality" (Stvilia et al. 2008), "reliability" (Rahman 2008), "credibility" (Korfiatis, Poulos, and Bokos 2006), "completeness" (Royal and Kapila 2009, 138), "authority," and indexing (Wallace and Van Fleet 2005, 101). Sometimes these studies conclude that the democratic digital archives are "nowhere near as bad as might be expected" (Stvilia et al. 2008, 984). More often they reach conclusions along the lines that a digital democratic archive "can be shown to be an unethical resource unworthy of our respect" (Gorman 2007, 274). But always these new forms are judged on the extent to which they fail to be traditional reference sources (Hilderbrand 2007, 54;Juhasz 2009, 149). Features that are unique to the new archives-such as the potential for hyperlinks-are judged in negative terms ("links can be distracting and can result in the reader being drawn away from the main purpose of an interaction with the encyclopaedia"; Wallace and Van Fleet 2005, 102).
However, a second tradition also exists-one that I personally find more interesting. This compares democratic digital archives to traditional sources of information, maps out the differences, and asks what the strengths and weaknesses of each are (see, e.g., Hilderbrand 2007, 49;Burgess and Green 2009, 89). For example, Don Fallis (2008Fallis ( , 1668 goes beyond the normal critique of Wikipedia in terms of reliability and notes that there are many other epistemic virtues beyond reliability . . . [including] the epistemic values of power ["how much knowledge can be acquired from an information source"], speed ["how fast that knowledge can be acquired"] and fecundity ["how many people can acquire that knowledge"].
He argues that "Wikipedia has a number of other epistemic virtues (eg, power, speed and fecundity) that arguably outweigh any deficiency in terms of reliability" (Fallis 2008(Fallis , 1662.
I have taken this second approach in this research project-to map the differences between the resources on their own terms rather than to assume that the traditional archive automatically represents best practice and any variation from that practice on the part of YouTube must represent a lack.

Curatorial Practices
It is worth noting up front that in terms of access, YouTube massively outstrips the NFSA. YouTube offered at least some content from fifteen of the seventeen television programs for which I was searching (only Homicide and Paul Hogan's cigarette ads were missing). In some cases, dozens of complete episodes were available. A television researcher wishing to view these programs would find YouTube an invaluable resource-within seconds she or he could be watching the actual programs. By contrast, the NFSA offers no material for online delivery. This is not news. Many researchers have noted-often with some rancor-that digital democratic archives offer an accessibility that traditional archives cannot hope to match (Hilderbrand 2007;Prelinger 2007, 115).
There are other issues raised by commentators that I do not have the space to explore here-the picture quality of material on YouTube compared to that of the traditional archives (Hilderbrand 2007, 54), the different relationships of traditional and digital archives to the idea of permanence (Rosenzweig 2006, 119), and the ways in which each of them facilitates the building of community (Juhasz 2009, 148). However, for me the most interesting findings of this research arose in relation to curatorial practices. Several academic writers have condemned the curatorial practices of YouTube. They note that it is users who function of the curators of YouTube-and they are extremely worried by this, noting that "curatorial direction is often minimal or nonexistent" (Gracy 2007, 194) and describing YouTube as "a mess" (Juhasz 2009, 148). Such dismissal of amateur curating is not unique to YouTube: it has long been the case that in museology the practice of popular collecting is usually distinguished from the professional practice of curating. The latter is seen as reasoned custodianships, selection, arrangement. . . . Mere collecting is often viewed pejoratively. (Hartley, Green, and Burgess 2007, 31-32) Rather than engaging in abstract discussions about whether expert or amateur curating is more worthwhile, this research project allows for empirical evidence to be supplied in relation to a particular project. For the television historian interested in accessing Australian television programs, whose curatorial practices are more useful-the professionals of the NFSA or the amateurs of YouTube?
Two of the key elements of curatorial work are accessioning (choosing which objects to archive and adding them to the archive) and cataloguing (managing metadata that will allow users to find items; Gracy 2007, 187). A third key element-organizing exhibitions-is not discussed here. This is done by the end user, and in this project I am taking responsibility for this element of curatorial practice. Indeed, this article represents, among other things, a curatorial exercise in exhibition.
Searching the catalogues of YouTube and the NFSA for seventeen key programs in Australian television history then allows me to present data on the accessioning and cataloguing practices of these archives.

YouTube versus NFSA-Accessioning
In a fascinating article, Roy Rosenzweig argues that the kind of history presented in the democratic digital archive Wikipedia is quite different from that presented by professional history. Wikipedia presents a popular form of history. Such a form of history is not created by Wikipedia-it already exists. Wikipedia collates and distributes it.
Participation in Wikipedia generally maps popular rather than academic interests in history. US Cultural History, recently one of the liveliest areas of professional history writing, is what Wikipedia calls a "stub" consisting of one banal sentence. . . . By contrast, Wikipedia offers a detailed 3,100 word article titled "Postage Stamps and Postal History of the United States," a topic with a devoted popular following that attracts little scholarly interest. (Rosenzweig 2006, 126) Rosenzweig notes three points where professional history diverges from popular history.
The first point I would like to consider is that professional history has overwhelmingly concerned itself with issues traditionally considered "public." Even the details of private lives are considered of interest only insofar as they can be written into larger stories of a nation, or of cultural history. By contrast, popular history is more concerned with "human interest": Wikipedia's history is . . . more anecdotal and colorful than professional history. . . . An account of [Abraham] Lincoln's life that focuses on debates about his sexuality and dwells on his birth date, nicknames and assassination is not "wrong" but it is not the kind of brief account that a professional historian . . . would write. (Rosenzweig 2006, 143) The second important point is that where professional history prides itself on providing abstract synthesis over the sweep of long-term historical narratives, popular history tends to be characterized by an interest in the particular and a concern with individual facts (Rosenzweig 2006, 131). Indeed, he notes in popular history an "obsession with list making" (Rosenzweig 2006, 142).
The third of Rosenzweig's points I would like to discuss is that although professional history aims for a sense of detached objectivity, popular history demonstrates an interest in the anecdotal and an "affection for surprising, amusing, or curious details" (Rosenzweig 2006, 141).

Accessioning: Public Affairs versus Human Interest
Curatorial theorists have acknowledged that the question of what material should be archived in audiovisual archives is "problematic" and "subjective" (Lloyd 2007, 55). "One may summarize the central tenet of preservation most simply in the following statement; we preserve what is of value. Yet, who determines the value of cultural objects?" (Gracy 2007, 188). As Rosenzweig notes, professional history is more interested in "the public" arena-even private concerns are traditionally seen through this lens. By contrast, popular history pays more attention to "human interest." As Hartley, Green, and Burgess (2007, 32) The popular memorialisation of television constructs a very different picture of "what matters" in TV history than do official, institutional or published histories. For instance . . . the underplaying of soapies by the industry is in tension with the high level of fan activity around them.
It is no surprise that in the genre of serious current affairs, the NFSA far outstrips YouTube in this research exercise. The NFSA has 120 access items for Four Corners. By contrast, YouTube has only a small number of complete episodes and segments from others with links to the complete episodes on other sites. It is also worth noting that the Four Corners clips on YouTube are almost entirely recent (with a single exception from 1992 covering Victorian Premier Jeff Kennett's election loss; http:// www.youtube.com/watch?v=nRFhVoCbSEQ [by the time this article reached the copyediting stage, this link was inactive-"due to terms of use violation". I discuss the issue of YouTube's impertinence in a separate article] ).
The NFSA is also strong for the oldest programs in the sample, regardless of genre. The two archives are largely comparable for In Melbourne Tonight: the NFSA has one full episode and some segments available for viewing, while YouTube has dozens of segments. In the case of Mavis Bramston, the NFSA has strong holdings, with many full episodes and clips available to view; YouTube has only four clips. And the NFSA is much stronger in its holdings of the cop show Homicide, where it has many episodes available to view (because of broken links it was impossible to work out exactly how many; I discuss the problem with broken links in detail below, under Cataloguing), while YouTube appears to have nothing. The NFSA has a strong collection of Paul Hogan's cigarette ads, but these are absent from YouTube.
Dramas seem to be represented reasonably well in both archives. The NFSA has access copies of nineteen episodes of The Sullivans, and YouTube has many clips. Prisoner is represented by eighty-seven access episodes at the NFSA, and YouTube has many clips. NFSA has access copies of several episodes of Return to Eden, episodes 1 and 2 from season 1 and episodes 1, 2, and 22 of season 2. YouTube has the complete season 1 and a small number of clips from season 2. However, as far as my searches could tell (with broken links), no episodes of A Country Practice are available as access copies at the NFSA, but YouTube has over seventy complete episodes available.
And in the most lightweight genres-lifestyle, soap operas, and game shows-the NFSA is poor while YouTube is stronger. The NFSA has five access episodes of Sale of the Century. YouTube has several complete episodes and hundreds of clips. For Neighbours, broken links make searching difficult, but it appears that the NFSA has no episodes available as access copies. By contrast, YouTube has many complete episodes available to view. In the case of Burke's Backyard, the NFSA has nothing available for access while YouTube has several segments. For Number 96, the NFSA catalogue is impossible to navigate because of broken links; YouTube has many clips.
The case of ABC programs is slightly different. The ABC, as a government-funded public service broadcaster, takes responsibility for archiving its own productions. And despite the presence of Four Corners, it appears that generally the NFSA does not replicate this archiving. And so there are no access holdings for the ABC productions Playschool, Countdown, or Live and Sweaty (or the Working Dog production Frontline, broadcast on ABC). All of these are represented on YouTube. The ABC archives are not accessible to the public-it is possible only to buy copies for research. And given that this research seeks to compare like with like, the fact that programs are archived elsewhere does not change the fact that they are not archived at the NFSA. This project specifically seeks to compare the NFSA to YouTube-not to compare all traditional archives with all digital democratic archives.
In terms of the specific "great moments" in Australian television that I discussed in my 2001 book, each of these archives holds six as access copies-but they are largely different lists. Both archives make available Bea crushing Lynn's hand in a steam press from Prisoner and the crocodile attack from Return to Eden. However, only YouTube has Don's declaration of his homosexuality from Number 96, Iggy Pop appearing on Countdown while he was high, Scott and Charlene's wedding from Neighbours, and the episode of Frontline in which a woman gets lost in the desert. And only the NFSA has the first episode of Homicide, the episode of Mavis Bramston from July 21, 1965, Paul Hogan's cigarette ads, and the episode of The Sullivans where Grace Sullivan is blown up. Generally, the NFSA skews toward older material, YouTube toward more recent.
In addition to a difference across genres, I found that YouTube was stronger than the NFSA in particular kinds of human interest-in particular, episodes of soap operas dealing with births, deaths, and marriages. In the accessioning choices made by the amateur curators of YouTube, the typical concerns of everyday human gossip-births, marriages, and deaths-loomed large. A Country Practice was strongly represented in this way (see http://www.youtube.com/watch?v=FlZ0Y0ltoO8; http://www.youtube. com/watch?v=ytNeq9Pg50c; http://www.youtube.com/watch?v=JalkqE2HvDk&feat ure=PlayList&p=310AEABEF081D634&playnext=1&playnext_from=PL&index=10), as well as by a series of "memorable moments," which are mostly moments of high human drama-"Molly Loses a Baby" (http://www.youtube .com/watch?v=wyEZyOQxgm8), "Terrence Has a Heart Attack" (http://www.youtube .com/watch?v=rfBuShmgofE), and "Frank Is Forced to Shoot" (http://www.youtube .com/watch?v=1LJmlNVMEEE). Another "memorable moment" is the appearance on the program by Bob Hawke (http:// www.youtube.com/watch?v=rej6o5Wh_cw).

Accessioning: Popular History
More generally, the kinds of material accessioned on YouTube match with popular histories of Australian television. This is not surprising-it is a popularly curated archive. In my original research project, I wrote a chapter on each of seventeen great Australian television programs. As mentioned above, for each program I highlighted one particular aspect that was important and that plays a role in the way it is remembered in Australia. In the Graham Kennedy chapter, "Advertising Transgression," I discussed the irreverent advertorials in his programs as a key part of the memory of In Melbourne Tonight and its importance in Australian television history. Kennedy would commonly mock the products, even as he was being paid to spruik them.
YouTube is clearly aligned with this version of Australian television history. When I searched for In Melbourne Tonight, I was viewing clips of Kennedy making fun of his paid advertorial products within seconds of visiting the web page (http://www.youtube. com/watch?v=YuJsrDzh3_s). An amateur television historian has edited several of them together under the heading "Graham Kennedy-Product Adverts, At His Sarcastic Best!" (11,791 views). This includes advertorials such as "Sitmar Cruises" (which shows a clip from a disaster movie of a ship sinking), a cat food called "Pussy" ("Yes, now you can buy pussy in a can. Pussy in a can. What will they think of next?"), a Raoul Merton shoe-throwing contest, Kennedy gagging over Peter's real fruit yoghurt, and Kennedy sabotaging Marchants "sparkarklarkling" lemonade by filling the cans with still water (http://www.youtube.com/watch?v=0YsNGv9aPcw&feature=related). The famous example I discuss in my book of Kennedy's dog refusing to eat Pal dog food and then running around the studio to piss on a camera is also available on YouTube (http://www.youtube.com/watch?v=aLIg69WRLTw&feature=related).
My history of Playschool in the book looked at how adults remember the program, particularly taking a transgressive pleasure in seeing the presenters as adult people with real lives. YouTube supports a similar history with its clip "Brooke Satchwell's Playschool Bloopers" (http://www.youtube.com/watch?v=PzLIXrVTNVo). My account of Number 96 focused on the program's breakthrough representations of homosexuality-using the "great moment" when Don Finlayson tells Bev that "I'm a homosexual." This is the second item returned in a search on YouTube for "Number 96"-"Number 96 (Don & Bev) World's 1st Gay Man Portrayed on TV": Don Finlayson has gone down in history as the first man portrayed on television as homosexual in 1972, twenty years before Melrose Place. Don is also the most centered character in Number 96. Here he breaks Bev's heart in telling her that he is a "homosexual" (http://www.youtube.com/watch?v=oe1ffkRG0Ao) Prisoner is remembered on YouTube, as in my book, for its representations of female violence. A whole series is catalogued, like wrestling matches, by the names of the competitors: "Bea Smith Vs Geri Doogan" (http://www.youtube.com/ watch?v=900AGxj5CJM), "Bea Smith vs Kay White part 1" (http://www.youtube.com/ watch?v=UdqsytXVbZ0&feature=related). In writing about A Country Practice, I had found that many viewers remembered Molly's death: "I still get a lump in my throat as I remember Molly slipping in and out of consciousness (and us slipping in and out of the 'blackness')" (McKee 2001, 156). As I already had a chapter on Grace Sullivan's death, I did not want to replicate this aspect of memorable TV, so I focused on another key issue about A Country Practice-its "social issues." But I was aware of the popular importance of this great moment. And I found that this was archived on YouTube (http://www.youtube.com/watch?v=_SOJjkNFqqU).

Accessioning: Moments versus Episodes
Rosenzweig also notes that professional history favors the grand sweep and the synthesizing account, whereas popular history focuses more on discrete facts. And so popular histories of television are commonly written in terms of a series of great moments rather than a synthetic historical narrative (Hartley, Green, and Burgess 2007, 28). I think I can see a link here with a difference between the NFSA and YouTube holdings. As Hartley, Green, and Burgess (2007, 30) note, Legions of amateurs, fans, retired technicians and announcers from the heyday of broadcasting have stepped in to fill the void [of television history] left by cultural institutions and by television itself. . . . They collect . . . video clips, theme music . . . idents, intros, and test patterns.
The NFSA favors complete episodes of programs. Some segments from magazine or variety shows are indexed separately-but they are usually held in the collection as part of the complete episode. By contrast, there is a presence on YouTube of television "moments" (Hilderbrand 2007, 50). Sometimes this is explicit. Several episodes of A Country Practice are catalogued under the heading "A Country Practice Memorable Moments" (see, e.g., http://www.youtube.com/watch?v=HF7waBlQAVE&feature=re lated). I found clips such as "Number 96-Moments, The Bomb Blast" (http://www .youtube.com/watch?v=cOqd4-zu0ic&feature=related). And the synecdochic moment from Prisoner that I used in my book (http://www.youtube.com/watch?v=eViG3HCo wnc&feature=related) is annotated by a poster, "Acclaimed as one of the most iconic moments of Prisoner's history." Other clips are not named as "moments" but are archived as clips that reflect the popular memory of great moments from the program. As well as the moment when Don Finlayson comes out as a "homosexual," the archive for Number 96 includes the murders committed by the pantyhose strangler (http://www.youtube.com/ watch?v=bChTCkmysyY) and the bomb blast (http://www.youtube.com/watch?v=k4x -eJvn0H8). Countdown is represented by the "moments" when Iggy Pop appeared on the program high (again, the representative "moment" in Australian Television; http:// www.youtube.com/watch?v=m1TUgmFsm3A&feature=PlayList&p=355E54EAC153 392D&playnext=1&playnext_from=PL&index=32) and when Prince Charles was interviewed by Molly Meldrum (http://www.youtube.com/watch?v=5-72muL9zrI& feature=fvw).
But it is also worth noting that this is not the only way in which television programs are remembered on YouTube. Just because some programs are archived as a series of great moments does not mean we can dismiss YouTube as being unable to provide complete episodes of television programs. As I mentioned in the previous section, several of the programs I was looking for in this research exercise are in fact archived on YouTube as complete episodes-in some cases, up to several dozen of them. They are broken down into the requisite ten-minute clips and linked as part 1, part 2, and so on.
It is notable that it tends to be more recent programs-or at least programs that have been recently broadcast (e.g., A Country Practice)-that are more likely to have complete episodes available. In the case of Neighbours, recently broadcast episodes are complete-whereas older episodes of the program are represented by clips such as "Charlene Has a Go at Mrs Mangel" (http://www.youtube.com/watch?v=QtDySEfGR JI&feature=related). Rosenzweig (2006, 141) finds in popular history an "affection for surprising, amusing, or curious details," and Hilderbrand (2007, 54) claims that YouTube "realizes much of the internet's potential to circulate rare, ephemeral and elusive texts." In this, the You-Tube presence of The Sullivans is perhaps archetypal-it is represented on YouTube by clips of its credits (http://www.youtube.com/watch?v=sJBNO10MspM), bloopers (http://www.youtube.com/watch?v=hgY_UHuvJgg&feature=PlayList&p=6797EAC BD7D2B210&index=0&playnext=1), and Kylie Minogue's appearance on the program (http://www.youtube.com/watch?v=m1fc_zBbhvE). The bloopers clip is an example of the surprising discoveries that I like to call the "YouTube treat." As annotated by the poster, this clip shows "some bloopers from the Channel 9 TV show, The

Accessioning: Objectivity versus Curious Details
Sullivans. These bloopers have never been shown on TV anywhere. I am sure the stars will be wondering how I got hold of them." Some of these "treats" are extremely useful for a historian of Australian television. Searching for "In Melbourne Tonight Graham Kennedy," I came across two items that provide a fascinating context for Kennedy's work on that program.
As noted above, Kennedy was famous for his transgressive approach to advertorials, making cheeky fun of the products he was paid to spruik. His cheekiness was always the kind that could be broadcast on a commercial channel-double entendres rather than obscenity. But nevertheless, the image one has of Kennedy from this archive is of a somewhat anarchic figure, refusing to follow the rules of decorum.
Two clips on YouTube put this image into perspective. The first is an advert of Kennedy doing an advert-not an advertorial-for Pal dog food. In the famous In Melbourne Tonight advertorial, Kennedy mocks the product and then his dog urinates on a camera. But the advert I am describing here is completely generic. Kennedy does not mock the product. He sells it exactly as he is meant to and performs a complete commitment to the dog food (http://www.youtube.com/watch? v=2Q6Pzk8-YV8&feature=related). For a researcher used to the anarchic image perpetuated by In Melbourne Tonight's advertorials, to see Kennedy as a purely generic celebrity is unsettling-and helps to define a quite different image for him. This is also true of a clip called only "Graham Kennedy" (http://www.youtube.com/watch? v=I9hhVtc0Bck&NR=1). This comes from an internal Christmas tape, and it is completely unbroadcastable. Inset against a background of ringing bells, Kennedy says, "Hello everyone! Here's wishing you a whole load of shit, falling from a great height, all over you, poop everywhere, and horse manure and vomit and stuff, fighting your way through it, and you're masturbating up there." This is another different image of Kennedy-not cheeky but completely obscene. The range of material and different versions of Kennedy available on YouTube leads to quite a different account of his performed image from that which I have developed from my previous work with the NFSA (see McKee 2009).
Other pieces of serendipity are less directly useful for my research but still delightful. Coming across Paul Hogan's 1984 ad for Australian tourism, "I'll Slip Another Shrimp on the Barbie," is lovely (http://www.youtube.com/watch?v=i2kzjxq0uuQ&feature=P layList&p=7326B8DD7FEC0CB3&playnext=1&playnext_from=PL&index=2). And a Canadian Fosters ad from 1986 emphasizes his overseas appeal (http://www .youtube.com/watch?v=r_AmlG11cWg&feature=PlayList&p=7326B8DD7FEC0CB 3&playnext=1&playnext_from=PL&index=8). The "lost" ending of Return to Eden series 2 was shot and broadcast when the program was not renewed but not included on the DVD release (http://www.youtube.com/watch?v=mzZsI79jghI&feature=rela ted). An episode of Burke's Backyard that was guest hosted by the D-Generation is extremely funny-particularly the guest gardener John Clarke, who makes it clear he knows nothing about gardening-"What's this plant here, John?"; "That's the sprinkler." "It must take a hell of a long time"; "Yes, it must" (http://www.youtube.com/ watch?v=6qc8gIKxcEg). I should note that there are also a few interesting pieces of related material in the NFSA catalogue-almost exclusively within the genre of news items about programs. News is a strong genre for the archive. For example, searching for The Sullivans, I found the news item "Paul Cronin attends and promotes the inaugural Innesvale Octoberfest and discusses his work on The Sullivans."

YouTube versus NFSA: Cataloguing
The second key element of curating I wish to compare is cataloguing-the generation and organization of metadata that allow users to search the collection.

Cataloguing: Mechanical Issues
On the most mechanical level, the cataloguing of YouTube is superior to the NFSA. During the period in which this research was done-the August 13 to September 10, 2009-many links were broken on the NFSA online database. The entries for Homicide, Number 96, The Sullivans, A Country Practice, and Neighbours all suffered from extensive broken links (in some cases, several hundred items were inaccessible because of broken links). By contrast, no missing links were found on YouTube in the course of this research.
In terms of searching and finding material, the NFSA catalogue is often difficult to navigate. Take the example of In Melbourne Tonight. This is a variety program starring Graham Kennedy. Versions of the program ran from 1951 to 1975, under a variety of titles-in addition to In Melbourne Tonight, it took the titles The Graham Kennedy Show and The Graham Kennedy Channel Nine Show. In popular memory, it is always remembered as In Melbourne Tonight.
Searching for "In Melbourne Tonight" in the NFSA catalogue produces several pages of results. The first page contains ten items, including news segments whose metadata includes the words "Tonight," "Melbourne," and "In"-"Basketball: Boomers 2nd game against Russia tonight in Hobart following Victory in first game in Melbourne," for example. There also appear a variety of apparently unrelated titles-"Amazing cricket discovery." It is only when these items are clicked for their full details that the researcher finds that this is "Sketch comedy featuring a young kennedy [sic], looks to be filmed during the 'In Melbourne tonight' series." There does exist in the catalogue an entry for the program itself-"In Melbourne Tonight." However, it is difficult to find. On the third page of entries we find "Graham Kennedy's In Melbourne Tonight Comedy Sketches" (1960). Within the page for this entry is a link: "Contained within." When you click on this, it gives you an entry for the program In Melbourne Tonight. However, this entry was not returned in the results when "In Melbourne Tonight" was typed into the search box. As noted above, unrelated news items appear higher up the list of returns than the program of the exact title that was entered. This is also true when one searches for Homicide, Number 96, and Neighbours. Bizarrely, when one searches the NFSA catalogue for "The Mavis Bramston show," the top item is Anything Goes, described in the synopsis as "varietycomedy show, replacement show for The Mavis Bramston Show"-the replacement is listed before the program itself. Mavis Bramston appears as item number 5.
By contrast, a search for "In Melbourne Tonight" on YouTube produces on the first page a number of hits from the program itself, clearly labeled as such. The first item is "In Melbourne Tonight (Graham Kennedy) segment": "A humourous [sic] segment from Graham Kennedy's In Melbourne Tonight programme from 1964." The segment is an advertorial for Kool Mints with "Dr Sandy McGriffith." Kennedy tries to eat a Kool Mint, while an actor doing a bad Scottish accent, dressed in a kilt, interrupts him with information about the Kool Mints he is eating. They speak over each other, get confused, laugh at each other, and finally join each other in a song.
But there are also irrelevant entries on this page-from the ill-fated revival of In Melbourne Tonight hosted by Frank Holden. As I did not want the new version of In Melbourne Tonight, I searched again for "In Melbourne Tonight Graham Kennedy." This returned a clean page of relevant entries. The searching is intuitive.
The biggest problem for YouTube is programs with titles that are made up of common words. Simply searching for the Australian current affairs program "Four Corners" produces no relevant results. However, the search "Four Corners ABC" worked and produced a page of relevant listings. Searching YouTube for the Australian cop show "Homicide" proved particularly difficult. I could not find any clips from the show itself, even including the names of particular actors that starred in it. A search for "Homicide Crawford" returned a couple of clips about Hector Crawford. "Prisoner" had to be searched as either "Prisoner Grundy" (the production company) or "Prisoner Cell Block H" (the British title of the program). On the other hand, the search algorithm used on YouTube coped very well with the single word "Neighbours"-that search immediately returned a full page of clips from the series.
It is worth noting that this can also be a problem for the NFSA catalogue-searching for "Countdown" produces lots of countdowns. Searching for "Countdown Meldrum" produces the entry for the program. Similarly, YouTube requires "Countdown ABC" or "Countdown Meldrum" to produce relevant clips.
Another problem with YouTube comes from the ten-minute limit on clips. Where complete episodes are available they are broken into multiple short segments. For some programs these are clearly ordered. For example, Return to Eden series 1 is presented as a playlist of twenty-five clips, in order. But for other programs the clips do not appear to be linked so that the viewer is automatically taken to the next in the sequence. Although one would expect that when one is watching, say, "A Country Practice memorable moments Terrence has a heart attack episode 1 of 2 part 1" (http://www.youtube .com/watch?v=rfBuShmgofE&feature=related) part 2 would be listed under "Related Videos" or given as an option to click when the clip has finished playing, this is not in fact the case. Often the rest of an episode's segments are not even displayed under the same search. It can take several minutes of dedicated searching to find "A Country Practice memorable moments Terrence has a heart attack episode 1 of 2 part 2" (http:// www.youtube.com/watch?v=lycKnzgxszc).
Beyond simply searching for a program, YouTube is more intuitive when searching for particular items-one types in what one would ask another human being, and this often produces the correct item. When searching for particular "moments" from a television program, you can type in exactly how you would describe that moment, and if it is in YouTube it is likely to be returned (see below).
By contrast, several programs in the NFSA have only a short synopsis, which may not include the relevant descriptive data. Some of the NFSA programs do have useful descriptions-for example, many episodes of Homicide (particularly the early ones) have short synopses that are also searchable, allowing for easy identification of particular episodes.

Cataloguing: Intuitive Searching
Searching on YouTube is more intuitive than searching in the NFSA catalogue. The metadata supplied for episodes tend to describe story content and, in the language used by nonexperts, to describe the kinds of great moments that popular history remembers from Australian television.
A Country Practice provides the perfect example of popular indexing. As noted above, this program is well represented on YouTube-austvclassic has uploaded over seventy full episodes of the program. The kind of material accessioned is that which popular histories of television remembers-births, marriages, deaths, and a series of "great moments" largely based around emotional events. Importantly, this logic of popular history informs not only the accessioning process but also cataloguing. On YouTube, these episodes of A Country Practice are organized into a series of categories that would be unfamiliar to Dewey but that make perfect sense within popular history of television. One group is titled "A Country Practice the deaths"-"Molly," "Daisy" (http://www.youtube.com/watch?v=FlZ0Y0ltoO8) and "Donna" (http://www.youtube.com/watch?v=-n_GmgzFTlE), among others. Another category is "A Country Practice the births," including "Tom and Charlotte" (http://www.youtube.com/watch?v=ytNeq9Pg50c) and "James" (http://www .youtube.com/watch?v=fSv31BCcM7s). Then we have a whole host of "A Country Practice the weddings"-"Vicky and Simon" (http://www.youtube.com/watch? v=JalkqE2HvDk&feature=PlayList&p=310AEABEF081D634&playnext=1& playnext_from=PL&index=10), "Tom and Anna" (http://www.youtube.com/ watch?v=KnAkLBPGUvc), and "Frank and Shirley" (http://www.youtube.com/ watch?v=okNvdwG5avs) among others. And as noted above, there is also a category of emotional "Memorable Moments" (which is actually a series of complete episodes rather than "moments").
This popular cataloguing of YouTube makes it easy to access. To find the episode of A Country Practice where Molly dies, you can type in "A Country Practice Molly dies." This returns the complete episodes of her death, with good picture quality. By contrast, searching for this search string in the NFSA catalogue returns "Sorry, no results were found." To find this in the NFSA it is necessary first to find the episode number where Molly dies from another source and then search for that in the catalogue. This is true for many of the programs in the sample. Searching for "Neighbours Scott and Charlene's wedding" takes you immediately to that clip. "Countdown Iggy Pop" returns Iggy's interview and performance. "Number 96 Don comes out" takes you to his confessional speech to Bev.

Cataloguing: Usefulness for Research Practice
The kinds of information about a clip that are presented in YouTube do present a problem to researchers of television history. Take the example of one of the Graham Kennedy clips. The YouTube entry tells me that "This one dates from September 1957, a mere four months into the program, with a still raw GK doing a live ad for Guest's Tru-Bake biscuits." But there are no more details of it. I can give the reader the URL-http://www.youtube.com/watch?v=YuJsrDzh3_s-but if that has gone by the time she or he wants to check on it, then there is no easy way for the reader to track down this item. And suppose that eventually a researcher does find a 1957 advertorial for Tru-Bake biscuits-how can she or he be sure that this is the right one? There could easily be many adverts for this product, over many weeks.
And even if the item still remains at that URL, there are problems. Yes, this is useful for the reader to view the footage and check any claims I may make about it (the basis of archival research-and empirical research more generally). But it is less useful for, for example, historical accounts of production practice or a discussion of the relationship between this advertorial and contemporary events, for which more detailed knowledge of the production date would be necessary.
Having said that, I then return to the NFSA and find that such problems are not the exclusive preserve of the popular archive. In addition to fully indexed items with production or transmission dates, the NFSA contains segments from In Melbourne Tonight. The segment "Amazing cricket discovery" contains the following information in its entry: Sketch comedy featuring a young kennedy [sic], looks to be filmed during the "In Melbourne tonight" series. Kennedy walks down a street and looks up to a sign that points to famous citys and a cricket game, fielders on the cricket pitch include Joff Ellen of In Melbourne Tonight fame. At one point in proceedings Kennedy tests the pitch with a rolling pin. This is the entirety of the information available about this segment. This is not sufficient information to allow a researcher to track down this item from any other source should there be a problem with the NFSA item. There are items both in YouTube and the NFSA that pose such problems to researchers.
And in the same way, there are items on YouTube that are provided with production information as complete as that in the NFSA, as in the case of "GTV-9 Xmas Staff Show 1968": GTV-9 Xmas Staff Show from 15th December 1968. This is ultra rare Xmas tape of the GTV-9 staff having a good piss-arse around and having fun in the studio way back in 1968. News reader Eric Pearce and the king of television Graham Kennedy appear in this tape. (http://www.youtube.com/watch?v=MhvwBiTFE88 [this is the second YouTube video which had vanished by the time of proofreadingagain for "terms of use violation"]) I was particularly surprised that Sale of the Century on YouTube had instances of the most fully annotated entries, with two complete episodes provided with transmission details: Episode 2841, recorded June 9, 1993, and aired on June 29, 1993, comedy episode of the Australian Sale of the Century, with Glenn Ridge, clip 1 of 3. (http://www.youtube.com/watch?v=jM8MooDKqsA&feature=PlayList&p=79 08D5FF4D58875B&index=5) The indexing of Neighbours is particularly impressive, presenting episodes with both popular indexing-"Bridget dies"-and indexing by production information-"ep. 5733" (http://www.youtube.com/watch?v=nzZaF9t7u84&feature=related).
It should be emphasized that the NFSA, in the course of this research, tended to have more items that included production or transmission information and dates.

Conclusion
The NFSA and YouTube each has strengths and weaknesses for a historian of Australian television. In terms of online accessibility, YouTube is massively more accessible than the NFSA. The NFSA collection is stronger on current affairs and older programs. YouTube is stronger for the popular history of Australian television-game shows, lifestyle genres, moments of human interest, "great moments" of television programs, and content that matches with the way that programs are recollected in popular memory. Both archives can provide full episodes of programs, but YouTube also has a strong presence of clips of "moments." YouTube is far more user-friendly. The catalogue is more reliable, and the metadata are more intuitive. However, the entries in the NFSA provide more production and broadcast data that will allow researchers to track down their own copies of material should it not be available from that particular archive.
My account of the current state of traditional archives describes the situation for the majority of the sector. While important work is being conducted by research archivists on the ways in which audiovisual collections can be opened up to the public (Gracy 2007;Lloyd 2007), as Rick Prelinger (2007, 114, 115) has noted, "access to most moving image collections is still minimal." He argues that "institutions still tend to define access in reductive terms." But it is important to note that this is not to say that it would be impossible for traditional publicly funded audiovisual archives to develop in ways that would increase their accessibility and map more closely with popular history. This is particularly important considering the relative permanence of digital democratic archives and traditional public institutions as a whole-rather than simply of individual items in the collections. The articulation of traditional archives with states-one of the most stable of human social institutions-gives them a longevity that commercial objects such as digital democratic archives rarely match. There is clearly still a role for traditional archives in the digital world.
Which brings me to my final question-would it be possible for traditional public archives to learn from digital democratic archives? To do so, it may be necessary to abandon the tendency of some archivists, as noted by Prelinger (2007, 115), to "debate the ethics of making collections available to 'just anyone'" and accept a (literally) "vulgar" approach to archiving. Take the example of serendipity-the fact that I found more surprising instances of unexpected material on YouTube than in the NFSA catalogue. Researching this article on YouTube allowed me to view for myself an important moment in Australian television history that I had not previously seen-"Kylie Minogue's First Scene in Neighbours" (http://www.youtube.com/watch?v=OVcRVS UhQVA&feature=related). I had heard industry accounts of this classic example of the "meet-cute"-Scott sees Charlene climbing in through a window, thinks her a (male) burglar, and tackles her. This is the start of a story that would lead to their marriage, viewed around the world, and set Kylie on the road to international superstardom.
To finally see it, on my computer, at my desk, was very exciting. At the moment, it is possible to access this moment only through a commercially run digital democratic archive. I want to make sure that such an experience is open to researchers for generations to come. Traditional archives have a role in this. To make this possible it would be necessary for archives first to decide that such a moment was important enough to archive (following popular accessioning). It would then be necessary, either through legislation or through individual negotiation with production companies, to arrange for access to the material to be made available through the traditional archive. The online catalogue would then have to be redesigned from the user's point of view to be as intuitive as possible, and the cataloguing system would have to be reworked to include popular vocabulary and systems of organizing televisual moments. It is a lot of work, all relying on a basic shift in mind-set to a consumer-driven view of television history. Will it, then, ever be possible for traditional archives to match the excitement of making a discovery such as this?

Declaration of Conflicting Interests
The author declared no potential conflicts of interests with respect to the authorship and/or publication of this article.

Funding
This research was supported under the Australian Research Council's Discovery Projects scheme (Project DP0879596).