Blogtalk Downunder
   May 19-22, Sydney

Blogtalk Downunder

Granular Video

Text is granular. Blogs, as perhaps the first indigenous medium to have developed on the web, are granular. Video on the web generally is not.

This is an important, and possibly fundamental distinction, for as I have argued video is highly granular (certainly as granular as text) during the production process yet once published it becomes quite closed to all of those activities that granularity affords. These activities have two basic aspects, one is granularity in terms of the network, and the other is granularity in terms of users.

In current practice videobloggers compress and embed their content into their blogs and this content becomes a closed object. It is assumed and expected that users will watch or listen to this material in its entirety, and is presented and constructed around these assumptions. This is, for example, why it is common for videobloggers to have opening and closing credits to their work - they assume that anyone will view the entire piece, and hence credit sequences are a part of this work. However, if we could quote just parts of a videoblog, just as we do with text, then obviously credit sequences are redundant. Furthermore, once we quote parts, not wholes, the assumptions that credit sequences rely upon become visible, and we will need to develop alternative methods for nominating such information within video. Just as we have for text.

Furthermore, this time based media, once published, is generally published in a manner where it has little awareness of its networked contexts. Such video does not automatically contain or embed, for example, its URL, time or date of publication, and other basic metadata. Current video architectures, existing and proposed (for example Apple’s QuickTime, and potentially MPEG21), can contain this information. Alternatively it could be embedded textually in a post’s metadata allowing it to be collected by existing blog systems. That this is possible, but not being done, is perhaps symptomatic of the manner in which video and audio is still conceived of as a ‘closed’ system, of finished rather than partial or fragmentary works. (Similar issues also arise with the duration of much work presented in this manner, their length makes them the equivalent of blogs posts that run to several screens - in blogging this is probably the exception rather than the rule, in podcasting this is the rule, rather than the exception, currenty videoblogging shows all the signs of following podcasting.)

If we use blogging as our exemplar, and if it is videoblogging then presumably the intention is for blogging to be the exemplar, then video in videoblogs should be granular in relation to the network. Furthermore it is reasonable that videoblogs should also exhibit the general qualities of what makes a blog a blog. Hence, blog based video would be made up of small parts, reflecting or expressing the life world of its makers, and an individual video blog would (much like television) become a serial form where the continuities and discontinuities between parts become important.

More significantly, however, video itself and not just its finished artefacts would become granular. For example, in my web browser when I am reading your blog I can click and drag my cursor over your text and copy this text for insertion into my blog post. However, in my web browser when I view your video I cannot nominate a passage of video to copy for insertion into my video post. Why not? The technology certainly supports this.

Similarly, if we recognise that a blog post is not just the text of the post, but includes its title, date and time of publication, trackbacks and possibly even comments, then a blog post is constructed of many parts and blog CMS’s have tools that recognise and can extract these parts in meaningful ways. In video and more specifically in video blogs, these parts also exist and can (or could) be extracted. For example a QuickTime movie can read an XML file and include within itself all of the above information. Futhermore movies could read this information from or about each other, and so exhibit the sorts of network awareness that characterise blog posts.

In addition, just as text is granular after its point of publication in a blog, so too can video. This refers to how we might use other video within our video posts, which is what this essay has concentrated upon. However it can also describe a method of working in video where we no longer conceive of video as being the production of something with a single image and sound track. This process, which has been elsewhere described as softvideography (Miles, 2003) lets us author video in ways that make it more comparable to text. Video in this model is always, even after publication, something that is constituted from parts that may or may not appear or be realised in the final work. It is, if you prefer, thinking of the video object as more like a blog so that just as a reader may only view part of a blog (indeed only part of a blog post) so too they may only view or listen to parts of an individual video entry. This change is a paradigmatic shift in what we think we are doing when we make a video blog entry, and a similar shift in what we think the role of the viewer or user of the video will be. It is a move towards a more active user, though I’d argue certainly no more active than what we expect the average blog reader to be. It does bear repeating that the change is simple, but deep, and is no more complex than recognising that our video can now be made of variable parts, just as our blogs are.

It is possible, though currently nontrivial, to treat blog published video as granular. The prototypes that have been authored to accompany this paper (which have all been published in a video blog) are early demonstrations of such a process. The first prototype shows a video file which contains partial time based links so that we might be able to imagine a video blog practice that lets users link to other networked items, just as we do with text.

The second prototype is a (rather dull) commentary that mentions two other video blog entries. When clicked upon during their mention, the commentary pauses and the prototype retrieves the mentioned video blogs and plays them within this movie. Such a video blog entry shows that it is technically possible to include other networked videos inside a similarly networked video, and helps to illustrate the questions that this raises.

The third prototype takes this a step further so that parts of an individual video blog (published elsewhere by someone else) is selectively quoted within another videoblog. In this example there are multiple selective quotations so that here commentary is woven around the originating video blog entry.

What each of these prototypes does not achieve is as significant as what they demonstrate. However, what I wish to emphasise at this nascent point in videoblogging is not what generic conventions or even practices ought to be pursued, but to observe that applications could be developed that allow us to work within video so that it retains its granularity after publication. Just as blogs have with text. This would be a hypertextual video, and much like blogs and their emergence, we do not know what such a practice will become. A blog, if printed, is no longer a blog, it cannot be a blog without its permeation by and within the network. If video in a blog can be removed and played, and is qualitatively no different, then it is not yet blog video. That difference requires invention. The architectures and tools exist, the hindrance is simply our prejudice, that is our horizons of understanding and expectation. This essay is an invitation to reimagine that horizon.

note: This page forms a part of a hypertext essay by Adrian Miles. The homepage for this essay is located at:
A long version of this paper (containing some but not all of the text contained in the hypertext version) is available at:

No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

generated in 0.072 seconds. | Powered by WordPress

part of an ever expanding network including edublogs - free education blogs, blogsavvy - professional blog consultant, wikisavvy - professional wiki consultant, incsub - social software and open source for education, in australia: melbourne blogs, sydney blogs & brisbane blogs, in the uk: london blogs, in the usa: new york blogs and incorporated subversion - social software, online education and james farmer