vog blog::vlog 2.0

New Gig

My friend Darren.

I scored a gig at Interact 2003, a large annual IT sort of conference. Lab.3000 here at RMIT is running a digital design showcase, I had a 10 minute pitch of my networked interactive video work (aka vogs) to a group at Lab.3000 who then made a selection of 6 people. Made the cut, so now I get to do more or less the same pitch to a crowd of 300. The event is free, and is on Monday September 8 so if you want to come, get in touch with the people at Lab.3000 to get a seat.

Nelson Holiday (nontiled)

Click on the image to load the vog.

This is the same work as just posted but without the video being tiled.

Nelson Holiday (vog readymade)

Click the images to load the vog.

This weekend just gone I spent quite a few hours working on a new vog, which is now done. This continues the same work that I put up last week, for more content from the same holiday to western Victoria, all of it derived from still images. There are two works, though each works identically. One has the video tiled, the other has the video untiled, partly to see how different they are, and also because Anna took most of the photos and she is dead set against the tiling. She is a painter after all and they weren't snaps so to tile them in such arbitrary ways she finds (not unreasonably) galling.

Technically I dropped the images into iMovie, applied the film age filter, the odd dissolve, and that was it. I tiled them in the usual way (batch in Cleaner) and wrote the text in BBedit. The text got dropped into LiveStage Pro with the video, and then I scripted the sprites so that when you mouse into the left vertical video pane the text track increased in font size by 2 points (up to 22 points). Mousing into the middle vertical pane paused the movie (mousing out begins it again), and then mousing into the right vertical pane decreases the text track font size (down to a minimum of 2 points). Since the rate of playback of the text track directly relates to the point size of the text as the text increases in size playback accelerates. I tried turning the text track into a child movie that ran for a much longer duration than the parent movie, so that even at 22 points the text was not too fast, but that proved difficult, so I introduced the pause sprite so that the user can pause at any point size and time if they wish to read the text.

The point of the film age filter was to make it look like some old super 8 footage, and also to artificially add movement to still images, continuing my recent interest in where cinematic movement might reside, if it isn't in the literal representation of movement. It is also using free software, and this in conjunction with simply choosing the images at hand produces a sort of televisual readymade aesthetic, a vog readymade. I played around with layering the text over the entire movie, which I quite like but I think I'll save that for a later work, and I was particularly interested in having the variable font size because I'm very interested in the relation of text to image and here the video sort of drives the text, the user's mouse varies font size, and through this the relation of image to text to user gets rather interesting. One (text or image) appears to me here to be neither primary. When the font is very small the text works primarily as visual, much like the video, at some point it becomes primarily text, but once it gets above a particular size it returns to being primarily visual. So this is sort of like noise in the text much like the noise I've added to the images. The text narrative is analogous to the use of the photos, a sort of text to hand writing style.

The Invisibility and Perniciousness of our Televisual Prejudices

While working around some aesthetic and theoretical questions in some recent vogs of mine a useful analogy arrived. Well, it isn't an analogy, since I think there is an isomorphic relation between the two terms, so I'll leave the correct terminology to the rhetoricians.

Anyway, I was thinking about presenting Discovery Bay and Glenelg Observations play in the QuickTime Player as full screen movies. This is contrary to what vogging is supposed to be about, though yes it does look sexy when you do it. My rationalisation for this was a) I wanted my work to look sexy, b) the blue pixellated background against the grey of the vog site looks awful, and c) I'm going to be chasing some funding shortly for work around vogs from a screen arts body and I realised that if work online using video didn't go full screen with black around the edges then they didn't really understand it.

What I mean by this is quite simple. Those who come to networked media (I think for a while that is my term of choice for what I do, it isn't new media but networked media, but it does look to and from net.art so perhaps it should be networked.media? No, I know, net.media, that's what it is) from a film or television background assume that visual content is supposed to fill the entire frame, that the frame is not a window but extends to include full extent of the screen, and that this is the centre of all attention. When net.media does not do this then it is treated with suspicion, even disdain. It isn't serious about itself as screen based media. This is one of the reasons I suspect that a great deal of the work that is funded and presented by such bodies tends towards a full screen aesthetic, and work that doesn't have this ambition, like much of net.art, falls under the prejudice of their radars.

This is, I suspect, exactly the same bias as happens in the academy in relation to electronic writing, where there remains for many a pervasive suspicion of any work that is written for and presented exclusively online. In Australia in the humanities everyone except the near dead will accept that online peer reviewed journals are as legitimate as print peer reviewed journals, but work that moves away from the print model as a writing practice is, in general, regarded with deep scepticism and mistrust. It isn't really academic work, or writing, or research. Each of these prejudices constrain much of what can be done, and ought to be done, in these environments.

A way of showing this is what I think might be one of the histories of Macromedia's Flash. Flash came along when graphic designers had been grappling with the web for a while. We'd moved away from tags as structural entities and were reveling in tables, font tags, vspaces and invisible gif tricks. Problematic appropriations to be sure, but the point was that designers were working and many got the web, and had success there. Now, for many graphic designers pages are static things. They don't move in time. And when they first moved online to try and get a decent sized image with the compression algorithms and bandwidth available was difficult and constrained. So along comes Flash. Vector graphics that move. All of a sudden graphic designers could now author graphic time based narratives, rediscovering traditional animation techniques along the way, and before long add interactivity. They didn't see vector graphics, smaller than full res images, and 6 frames a second as a reduction but as a liberation. Movement, variability, interaction. And they hit the ground running. Television people can't even conceive of 6 frames a second video, let alone figure out what they could author in it.

So one group came to web with a set of prejudices that allowed Flash to fly. The other group (who are still coming) have a set of prejudices that prevents them from thinking the net. So to help me apply for funding, my work needs to be presented full screen. I can keep the video at its current 320 x 240 pixels, on a 640 x 480 static background, but it needs to appear monumental, black borders and all. Just like the hypertext thesis that must be printed and bound for examination.

Discovery Bay

Click the image to go to the Vog.

I used the template from the Glenelg Observations blog to do a bit more experimenting. In this work there are no child movie video tracks (which is why it weighs in at 1MB), but there are 5 child movie video tracks. To play the vog (it will autostart) mouse into the video panes, this loads and plays the text movies, so mousing in each time will load a new text movie and play it (a total of 5 available, all brief).

Technically I just wanted to see if I could get better playback by not using child movies for video, but it hasn't made that much difference as far as I can see. What I should do is make them play in the QuickTime Player (the blue with the grey at that size is disgusting) and also full screen. Completely contrary to vogging but would work well for these works. (We'll see.) I was also interested in having a series of text works related to the video content appear, so that the video is now singular (in that it is a discrete linear sequence) but the text narratives that accompany it become plural. Now since they're text tracks they are very small files, but the text is keyed over the video which does require some CPU horsepower (ah, when the world runs on G5's, then desktop cinema will appear obvious and trivial), but I could run them backwards. But they're text, and each is written as a sequence so that wouldn't make much sense.

Instead, we have 5 text movies of 5 lines each and these are loaded by the user mousing into the active video panes. This sets up really interesting moments and stories since which text movie is being viewed is extremely variable (and can be easily replaced by another courtest of a mouse entry before it is completed) and of course different textual fragments appear juxtaposed with each of the five images used in the video.

This produces a simple combinatory engine that has, well, no fixed number of outcomes and the work, if the work is thought of as the relation of the video to the text, is only ever realised in each reading. I like that. Got a thing about process (as my beleaguered students can testify). Another thing I might do is up the data rate and make it seriously broadband, see if that makes these works flow in the way that they need.

Glenelg Observations

Click the image to go to the Vog.

I've been working for nearly two weeks now on a new vog. Since disk space has become critical I've had to work in a rather piecemeal sort of way, which has really slowed me down. Plus my scripting sucks, so there's been way too much trial and error going on. The work, Glenelg Observations, is as much a technical as aesthetic experiment.

Technically, the work is a one second parent movie that has two child movie tracks. One child movie track has a list of six quicktime videos (each made up of 9 individual video tracks), while the other has a single text movie. Mousing into the region where the text movie appears is counted by a sprite track and controls loading of the video child movies, while mousing into the moving video panes causes the one text track child movie to play. Both of the child movie tracks are set to loop and are not slaved to the parent movie. The parent movie is just a one second container consisting of a picture track (the blue graphic), two sprite tracks (one over the video, one over the text movie) and two child tracks.

I had wanted mouse movement into the text movie to also make the child movie video to play backwards, since the work is loosely interested in the idea of the palindrome, but playback was just appalling on my tiBook so I gave up that particular dream. I might try to make one where the child movies are not each made up of nine video tracks, since that is just asking an awful lot of QuickTime and your computer's CPU.

Aesthetically, theoretically, creatively, or whatever hat I'm supposed to be wearing at the moment, the work consists of a series of photographs which have been dropped into iMovie and 'animated' via the so called "Ken Burns affect". I am interested in how this might work, what it might look like, and how to use it, largely because I am attracted to the idea of being able to use still images in work like this but not keep them as still images. I'm not happy with how it looks, though I might try the same thing using Photo to Movie or reading the 750 page manual that came with Final Cut Express to see if it can do it.

Apart from appropriating still images into movies (continuing an interest of mine which harks back, as ever, to Marker's La Jetée and Deleuze's argument that cinematic movement is not represented movement) as a lo-budget lo-tech sort of vogging, the work explores what might happen when you have a text track which is a vague sort of narrative that can run independently of the video track/s. So the textual commentary, which is user controlled, loops and has a different duration to each of the six video clips that can play. The video clips are each of round things found at one location in the Lower Glenelg National Park, and the idea was to develop a sense of place only through these minor objects. It doesn't quite work because the pans and zooms are just too lumpy and stutter, but one thing I keep returning to in my vogging is the building up of narratives, themes, ideas, worlds and movies from fragmentary fragments. The background of the movie sort of repeats this since it is one of the photos used in the work.

So this work continues my recent interest in vogs that have multiple durations, so each vog contains at least one child movie track that is not slaved to the parent time line and so is able to play as a separate movie to the parent movie. This has the effect of making a work that has two times (the time of the child movie or movies and the time of the parent movie), though that's wrong, since the more interesting and significant time is the time that happens between these two.

I've added a title come proper noun to the text track following Tim Hall's recent examples.

Photo to Movie

This is shareware that does a much better job of turning still images into QuickTime video files than the tricky little effect that comes with iMovie 3. The site has some demo's and the like, and you can use the software without paying the registration cost to test it out, but you will get a watermarked movie.

Troy

Troy Boulton, who I know from a former life as webmaster here at RMIT, has posted a vog. (Got the link via Tim Hall aka vogner, a one person vog factory.) This particular work seems to pick up Tim's ambient work which as far as I can tell is, well, ambient. Non narrative, combines what I expect to be self composed electronic music with VJ sort of graphics, though indexical to the extent that the images are derived from the world rather than generated. Troy's work is much more grounded in the world, which isn't surprising given his indymedia credentials, though for a first effort I'd recommend upping the bit rate and resolution, or using text tracks for the text and not trying to do it in an editing program.

This is the difference between what I've characterised as hard and softvideo. (I have two essays on this forthcoming, but if you send me an email I can point you to them preprint, it's also the substance of the talk - click on the globe - I gave at AOIR02). Hardvideo is when you use the computer but the paradigm of publication remains a physical substrate. Softvideo is when the medium of publication is the computer screen and where you author video on the computer.

A simple difference is in the text issue in Troy's vog. Text in hardvideo (iMovie, Final Cut Pro, Premiere, doesn't matter which one) is rendered into the movie at the end of editing and so is not an independent object in the movie. It becomes part of the picture track. This means that when you compress the movie for networked delivery, the text gets compressed too, and if you don't use a decent codec and understand a bit about key frames and the like, it will degrade very very badly. If you used a text track in QuickTime to do this, then it is literally an ascii text track that is stored in the QuickTime architecture as an ascii text track. This means it is tiny, it doesn't need or receive compression, and so at run time it simply reappears with no loss of resolution. Problem solved. How to do this? Check out the tutorial (and check out my intro tutorial on simple track editing and building in QuickTime) and spend the few dollars to licence QuickTime pro, it is an authoring environment more than a publication environment.

Desktop Vogging

A self serving entry so that my online tutorial on how to desktop vog can be placed in my vogging tools category. So, if you have QuickTime pro, or want to see why it is useful, check out the tutorial. Available fresh at:
http://hypertext.rmit.edu.au/vog/desktopvogging/desktopvogging.html

John's QTVR Project

What We Will is a broadband QTVR work published in 2001. John Cayley, and others, made this and it is a significant new media piece. It is a work that moves QTVR out of the 'lets be immersed' genre, and out of what really is just a sort of value added infotainment category, towards narrative. This is profound. What is point of view in QTVR? What is point of view in an QTVR narrative? How do you narrate in QTVR? One of the primary issues for all multillinear narration is time. The time of the story, the time of its telling, user time (many have written on this, including Jill Walker, Markku Eskelinen, and Raine Koskimaa), but also the role of time in reading pleasure. This is the task of deadlines, where it is virtually impossible to find a popular narrative that doesn't embed constant deadlines against which the narrative strives. Deadlines are teleological, multilinearity tends to work against teleological narrative, and so to find and develop a narrative form that has an understanding of diegetic duration, multilinearity, and is a graphically rich environment, in a medium which is not timebased, is very significant. Have a look to see what the answers are.

Must be Wednesday

By Wednesday afternoon I'm pretty slow. So now instead of doing the 1001 one administrivial things calling from my mess that is otherwise known as my desk, it is a fine time to consult my blogosphere. Where I find myself listed at educational.blogs.com/movable_type which is a, as the url suggests, a portal come cluster for education blogs using movable type. Thank you Albert Delgado.

Two Birds

With one stone. I've just installed ImageMagick on my OS X server and so now movable type lets me upload stills, nominate thumbnail resolutions, and the rest of it. I've very much missed having images in my blog, and now they're back.

This is a pier post photographed by Anna on the Glenelg River, and will be turning up in a new vog shortly.

Alphabet Soup

Matt has an interesting post about alphabetisation and how the sequence is arbitrary, but so pervasive as to appear natural. He also points out that in California the ballot of candidates is not listed alphabetically but randomnly. The same happens here in Australia where a ballot is held to determine the order of names on the ballot.

The alphabet as ideologically naturalised is something I often turn to when teaching hypertext, and a favourite personal example when I'm talking to academics and teachings about new media or hypertext literacies. I use the example of a bibliography, and point out that in general there are two ways a bibliography is used, to look for a specific author, or to simply browse. Now, when browsing order, certainly order created by an alphabetised hierarchy based on author's family name, makes not a sod of difference. On the other hand if you're looking for a specific author then this helps (and is the only reason we have alphabetisation in things like indices and bibliographies anyway), but if it is an electronic environment it is completely irrelevant. It is just as easy and useful to type in a search string than it is to scroll through a bibliography, and so in electronic environments things like alphabetised bibliographies are only a hangover from the late age of print. Or, if you want to be really provocative, an illustration of the ways in which a lot of print literacy is in fact an ideology and in some manner an expression of power as a consequence.

If the designers of software, including major word processing packages, understood this then we would get much better search tools, for example search systems that could reveal all instances of a text string in context, and so on. However, since we rely on legacy infrastructure like alphabetisation to achieve 'simple' tasks like finding an author or an index entry, there is no real development within software for these tools. This is unfortunate since this simple instance of random access is one of the major defining and empowering features of these environments.

QuickTime Tutorials

Using QuickTime pro and a text editor there is a lot that you can do in terms of vogs. Text tracks (layering text over image), HREFTracks, simple collages and montages. This is the Apple QuickTime page that lists the tutorials available, including all the embed tags and so forth. Necessary resource for those interested in desktop vogging.

HREF Tracks

From an essay I'm currently writing: QuickTime’s HREF track is a track type which allows a movie to load Web based content as it plays. The HREF track is a specific type of text track within QuickTime that contains a combination of timecode and URLs. The URLs contained in the HREF track can be either manual or automatic. Automatic URLs means the movie will send a request for the URL as the movie plays, while manual URLs means the user must click on the movie during the nominated interval (defined by the timecode) to load the URL. The Apple tutorial on HREF tracks is pretty straight forward. Textation, the tool I mentioned yesterday, makes this much simpler.

Translation

Here is the Google translation of the Liberation piece of Stiegler. It is in what I like to characterise as Googlish, the sort of English we'd all speak if we learnt our language via the Web (as many learn it from television). Still hard to make out, but seems the bullet in the gun was intended for himself...

Feelorium

Feelorium has a small suite of QuickTime programs, including Textation, the premier shareware text track creator for QuickTime. Very useful program if you want to create text tracks and don't have something like LiveStage Pro, nor need something as big as LiveStage Pro to get a text track or HREF track done. Trés gewd.

Post Structural Dinner Table Gossip

French philosopher, Bernard Stiegler (who's work on technics and time is of major significance to new media theory) in answer to a recent question about how he became a philosopher replied that it was during his 5 years in jail for armed robbery! The story is here, and though it is quite a few years ago apparently this has been a well kept secret.

First Vog Off the Rank

David, over at dabra.com, has just released a video blog template built using Flash MX. He's provided the source. As far as I'm aware this is the first available video blogging tool out there, so hats off to David and the gong for first cab off the rank.

Early days yet to see what David is going to do with his video blog, and I'm not convinced that Flash MX is the way to go (just check out what Tim is up to to see what it means to use QuickTime as your authoring environment). Things are getting exciting!

Ethics

There is often a lot said in the blogosphere about ethics. Something I wrote in the DAC blog some time ago got a, probably justifiable, serve, and it is something that is important for such a public form of writing that also offers itself as commentary. Personally, given my film theory background, I've always thought documentary has a lot to offer blogging. Not only in that blogging and documentary have an affinity, but that many of the ethical issues confronting blogging have already confronted the documentarist and so there is a rich literature and practice here. Which goes to various extremes, I might add.

But that's the subject of another writing, another time. In reading Jonathan Delacour's post about ethics I see he points out, contra Rebecca Blood, that facts and truth aren't equivalent. 'Bout time someone with a readership said that. They aren't. It is a hackneyed but useful example. It is true that unicorns have a horn, but unicorns aren't real. Now we could say that it is a fact that unicorns have horns (or a horn), but that is a pragmatic use of 'fact' where the sense being described by Delacour is more like facts as information and truth as what the information is taken to mean. Truth is then subject to interpretation, and unlike the old days, can no longer to taken as something that is always so. What was 'true' today may not be 'true' tomorrow. Facts, on the other hand, may be true, but truthfulness isn't their primary quality, they are more like objects that we attempt to make truths, and vice versa. In other words a fact can more or or less true, but a truth can't be more or less factual, nor more or less true.

Or have I got all that arse about? Home time.

Assessment Matrix for Journals

In Advanced Seminar this semester students have nominated a research question or problem in a field of their own choosing. They are also required to maintain an ongoing reflective journal. Some have chosen to blog, but most are using pen, page, scissors and glue. I have written an assessment matrix for their journals. This describes the qualities and requirements that a journal entry or entries would need to meet within each of the assessment values we offer (high distinction down to fail).

With matrix in hand each student then chose an existing journal entry, wrote an entry assessing it and grading it using the matrix, and then had their entries read by two peers. If the peer reviewers didn't agree with the self assessment according to the matrix I would arbitrate. The whole thing worked really well, I sent the bloggers off to do their version in their blogs, everyone else stayed in class. No arbitration was necessay, they could all work out where their work lay against the scale, and could easily identify how it would need to change to improve.

Pragmatically this is good since it frees up some assessment load from me. However for the students it is excellent as their assessment no longer happens in a vacuum. A regular high distinction student doesn't know what credit work looks like, and vice versa, a task like this really helps erase those differences.

Journal Assessment Matrix (pdf).

tBox 2

Catching up on updates and older news, and I note that Tinderbox 2.0 is out with blog support. I used tinderbox to write and publish vlog1.0, which it did admirably, and have since been using it for brainstorming, note taking, and trying to structure my research a bit better. Updated manual, more abilities. I think it is nearly time that I wrote a new multilinear essay and I ought to do that in Tinderbox.

Tim Hall, aka vogner

This is a repost of something I wrote a couple of weeks back, which somehow disappeared from my blog. Gremlins or voglins?

Michel Serres

A new web site has been launched dedicated and/or devoted to the work of Michel Serres. Don't know if it is the first, best, or biggest, so YMMV, but he is a major figure in the philosophy of science who I've rather enjoyed reading in the past. Some of his work is useful, in round about ways, in relation to networks and new media.

Hannah's House

One of my project groups in Professional Communication is developing a 'knowledge resource' for new international students. The form of this is going to be a fictional blog, more mockumentary than diary, written by their created character Hannah. Hannah is going to document her experience of arriving in Melbourne (from Singapore), and what it's like to deal with a large institution, in a foreign country, away from home for the first time.

None of these students have any experience with blogs, online writing, or new media. None of them appear to have any experience outside of very strictly defined professional writing genres (journalism and press releases). So what and where they end up is going to be, um, interesting.

It is an exciting project, and one that has a lot of potential, and if done reasonably well could be a great resource for new students. That's the pitch, anyway.

Skiiing (again)

Well, the trip to Lake Mountain was a success. Plenty of snow (by Lake Mountain standards), glorious blue sky, no wind, so almost spring skiing with light shirts and not enough to drink. Anna and I were both poor so I decided to take my skating ski's to save on hire, which I haven't used for years and dreaded trying to get up the hills, and she hired some nifty revolution skis that were really short and surprisingly zippy. I fell I think 3 times, Anna once (ha!) and after nearly 3 hours we were both exhausted.

Now I'm busy planning to try and go again this coming Friday, just for another half day. Oh, some photos should be coming soon.

New Book

Via Matt, this is a book on the text and image that I definitely must read! This is one of the things I'm interested in exploring in the vogs, but also in general through the research and work that I do. Particularly in relation to developing new ways of undertaking cinema studies.

QuickTime on the Web (1)

For the next three weeks I'm teaching a brief module for our Advanced Media Production students on QuickTime online. This is going to cover compression basics (what a codec is, loss compression, variable bit rate compression, and so on). Hopefully some of the tools out there that help with these sorts of things (for instance what tools are out there that help with things like the data rates of files and so on). We'll do some compressing using Cleaner but also to MPEG4 in QuickTime 6, and then embedding this online. Which will include poster movies, reference movies, some of the features that the QuickTime embed tag supports, and so on. The first session will cover:

• getting work out of the edit suite
• uncompressed QuickTime
• transfer codecs
• delivery codecs
• spatial compression
• temporal compression
• variable bit rate compression
• key frames
• frame differencing
• data rate versus file size
• http versus rtsp delivery
• codec installation
• cinepak
• sorenson 3
• sorenson 3 pro
• mpeg4

In the session we will shoot some DV, handheld with flash pans, then locked off with a still subject, so that the issues of compression can be clear. I'll probably use Cleaner to do this since you can see the data rates, key frames, and play around with natural versus manual key frames and the like. All crucial stuff. Then at the end of the class I"ll ask them to answer

• A key frame is?
• A variable bit rate means that . . .?
• Real Time Streaming Protocol (rtsp) is better than the HyperText Transfer Protocol (http) because? (That's a trick question)
Which of the following directly reduces the bandwidth requirements of an online audio or video file?:
• duration
• image size
• colour depth
• frame rate
• sample rate
• computer processor speed
• network speed
• What is the difference between natural and manual key frames?

This is not about specific codecs but about the general introductory things that if you get some handle on then help you understand what is going on. After all most codecs need you to understand something about these things!

Video Conversions

Jeremy Wood has written an OS X and Java program, Segue, that supports drag and drop file conversions of .avi, .dvc, and .mov video formats. I tried an earlier release and ran into some sort of QuickTime bug on my computer, and what I was actually wanting to do was convert a .wmv file to something useful, like a .mov, which I still haven't figured out how to do. In my own practice I only use QuickTime but it is handy to be able to convert .avi's gleaned out there in network land to QuickTime so that I can use them.

DV and QuickTime timelapse

Gary Fielke has written three small programs that do nifty things for DV editing. While these are what you might call QuickTime and/or DV tools, they are the sorts of things that are interesting in vogging because they let you capture things in ways that your hardware (camera) may not support, and because vogging is somewhere about these minor interventions (stutters) within a dominant argot. So, I am hoping to figure out where I can safely set my camera and tiBook up using DVTimeLapse to record a day's worth of the usual (clouds, crowds, traffic). What I'd love to do is to have it record the view from where I used to live, where there was a view over the bay and across the docks. But don't live there now, though I guess I could get an OK view out of the back window of my cruddy little flat. . . .