Video citation and blogging frustration

A few months ago, I wrote about the tragic inaccessibility of audio and video content on the web. Today, while trying to summarize one of the key insights that came out of the BloggerCon conference I attended this past weekend, I was again reminded of this problem. The issue I want to highlight is the gap between what technically proficient users of the blogging medium (or indeed any kind of web authoring) can achieve, and what average users can achieve. Ironically the best way for me to make that point -- by citing portions of the webcast -- is yet another illustration of the problem. In several different sessions, people made compelling pleas for simplicity. It ought to be easy for me to send you to those places in the webcast. It isn't.

In this two-minute clip, industry veteran Amy Wohl talks about her frustrations with blogging tools. It's a powerful statement. Writing for the web is something I've done for so long that I forget how much tacit knowledge I've accumulated over the years, and how it empowers me. But for Amy (and for most people) effective use of hypertext, images, and tabular data in a fast-paced communication environment is unreasonably hard. The analogous problem for me is communicating with video clips. Here the knowledge isn't tacit, and the communication doesn't flow easily.

Let's look at the steps required for me to bring you Amy's clip. On the BloggerCon webcast page there is a link to the day 2 technology session. Let's look at that link:

http://cyber.law.harvard.edu/ml/output.pl/35514/stream/temp.ram

In order to refer to a location within that stream, I need to inspect and modify the contents of that file. Here's what it contains:

Of course, I can't just load the .ram file into a browser, because it's hardwired to launch a player to play the stream described in the file. Some alternate HTTP client is needed. Already we've fallen off the continental shelf in terms of any normal person's knowledge. Here's what I did:

curl http://cyber.law.harvard.edu/ml/output.pl/35514/stream/temp.ram

In other words I used curl, a command-line HTTP client, to fetch the contents of the .ram file. Next, I embedded the rtsp URL that it contained into a new .ram file, and added start/stop parameters like so:

Of course it took some fiddling to get the start and end times I wanted. Finally, I uploaded the .ram file to the address I cited at the beginning of this posting. Even this procedure, which I take for granted, is far outside the mainstream. To this day, uploading an arbitrary file to a public URL is something that most people can't do. We (as a species) know how to attach files to email. We don't know how to post files and communicate their addresses.

So anyway, the end result of this process is very powerful. I can point you to something compelling that Amy said, and you can watch her say it. Equally noteworthy is the fact that I have now made Amy's video clip visible to Google. Unfortunately the process that yields these powerful results is absurd.

Hundreds of such moments during the conference (and indeed in every conference that's ever been webcast) ought to have been similarly exposed to Google, and to the linking and citation mechanisms that drive Google. Nobody has time to watch complete webcasts. We have the technology to quote from them, and thereby index them. But it never happens, because we haven't packaged the technology for effective use. This sucks.

Clearly writing and production tools don't even scratch the surface of what's possible and, if we're to realize more than a fraction of the web's potential as a communication medium, what's necessary. For consumers, though, I'd have said that things were pretty well in hand. But the discussion in my own session on aggregators showed me that here too, major ease-of-use barriers remain. Here's the five-minute clip that captures that segment of the discussion. It was the most animated five minutes of the whole hour-and-a-half session. I know that the impassioned plea made by the guy shown in this screenshot annoyed some of the more technical attendees. But let's think about what he's telling us. And let's also think about why these kinds of powerful moments, even when recorded for the web, don't participate fully in the web.

Update: AAARRGGHH!!! It gets worse! The .rm addresses encapsulated in those .ram addresses have changed since this afternoon. Dynamic, maybe? Phooey.

Later update: Fixed. This from Hal Roberts:

Sorry you're having trouble with the bloggercon video links. We store all of our video clips in a media archive (http://cyber.law.harvard.edu/ml) that includes (among other things) permissions on some of the files. The system only gives out temporary rtsp links because it can't password protect them. In any case, I've mirrored the files to our static helix server. You can get to the file you need at:

rtsp://cyber.law.harvard.edu/BloggerCon 2003/BloggerCon Day 2 - Technology.rm

The other files are at analagous locations (eg. 'BloggcerCon Day 2 - Aggregators.rm').

Former URL: http://weblog.infoworld.com/udell/2003/10/08.html#a823