Stories from SXSWi 2008 – Creating Findable Rich Media Content

Here are my notes for the Creating findable rich media content session at SXSW Interactive. Listen to the podcast for yourself if my haphazard notes are difficult to follow.

• Navigation typically not followable for Flash, etc. Text is embedded, not retrievable by spiders, key text is not prominent or differentiated (even XML).
• Lack of a unique URL hurts your linkage and Google ranking subsequently.
• If content is not coded or tagged correctly you’re not as findable.
• Disney example – their entire site is Flash. You can make Flash search-friendly, navigation is key – just make sure spiders can get through.
• Javascript function detects non-Flash capable browsers, so viewers get primary content (text, anything you can add to an HTML page).

Samsung example – Flex and AJAX for 20,000 SKUs of different tv models, used XML site maps to get all the deep links (which were previously unfindable).
Economist has a video site – 1 page for each video linked from master.
Tubemogul lets you upload videos in bulk with good tags, good titles.Not always rich media that’s the problem, but the execution, making sure you think about search and findability early on in the project, and tag early.

Sometimes content goes up only for a month and then comes back down, so search is irrelevant. Plus, if you want a rich experience, then you don’t worry about search – you actually want fewer people to have that rich experience.

Creating a findable strategy – or make your content find your users. (Now that is an interesting concept to ponder for technical writing.)

Fiat website – Flash-based
Layered approach – CMS backend with XML that transforms either to HTML or to have Flash consume the content. This approach could be mistaken for a form of cloaking, make sure intent is legit and alternative is a faithful representative of Flash content.

Other SEO suggestions – break up container, create deep links from blogs to specific content allowing inbound links.

Other findable strategies
Never ending friending report 2007
Asked people ages 14-29, if you had 15 minutes of spare time, what are your top two choices for using that time? Social networking or talking on cell phone were the top answers.

Target example (Adweek article) -Back to College campaign on Facebook
2-3 months lifespan, so this is an example of not worrying about findability, but rather ensuring that your content finds your users. How does Target create a dialogue with college students; one that would inspire and support their transition into college life?
Give freedom to kids to discuss produts, within their own community.
Personalized checklists sent to mobile.

Funny side note – I think this Target campaign was a nominee of one of the “Suxors” as one of the worst social media campaigns in 2007.

Consider everyone’s accessibility – mobile phones, text to speech, and so on.

Google webmaster tools – google.com/webmaster – these are relatively new.

Q: What is the biggest challenge coming up?
A: Something should be invented to work in the authoring stage to give info to the search engines.
Q: What about exclusionary methods? They don’t understand the ping pong effect something that’s cool will come around everywhere? His clients don’t want to pay for the bandwidth and so on.
A: I don’t think they actually answered this other than to say viral is always good.

Q: What about microformats?
A: The Google panelist said it needs to get more standard and have more attached to the content. He did point to http://www.google.com/experimental/.

Serendipitous trips through the river of information

Anne Zelenka has once again nailed down the idea of how to work in a new world where there are far too many sources of information and no way to use technology to help you sift through that information. So you get on your raft and float along the current, and along the way you find the information you need to get a job done.

In The Future of Ignoring Things on Internet Evolution, Cory Doctorow says “I’ve come to grips with this — with acquiring information on a probabilistic basis, instead of the old, deterministic, cover-to-cover approach I learned in the offline world.”

One of the commenters on Cory’s article talks about navigation, saying “We need a sensible, extraordinary navigational system which helps the things we care about at the moment rise to the surface.” How amazing would your help system be if you could apply this technology to your product’s help?

My own journey – searching information

As for me, I am coming to terms with the idea that I’ll stumble upon useful information. But I tend to think I’m a pretty darn good researcher and that my Google Fu is strong, enabling me to find the best relevancy and authority in my information seeking. The more I travel on the river of information, though, the more I’ve come to appreciate the serendipitous discoveries.

My occupation – supplying information

Since as a technical writer, I write the loads of information that others need to sift through (or travel upon, choose your metaphor), this mind shift seems very important to follow. While Cory and Anne are miles ahead of most of the audience of the product I work on for my day job at ASI, my work on the OLPC project has a very different audience and that audience might be closer to Cory and Anne in their ability and desire to find information probabilistically. So for the OLPC project, documentation in a wiki seems appropriate. For ASI, only subsets of the overall audience might benefit from a wiki, such as the technical implementers using the Software Development Kit.

So, back to the basics it is – consider your audience while creating that flow of information. And keep an eye out for that extraordinary navigational system.

An example of serendipity in information finding

And now for the storytelling portion of this blog post. I often write blog entries and leave them in draft form for a while. This very post happened to be in draft form while I wrote and published my recent Wikislices entry. Sheece Baghdadi, who is a technical writer for Webaroo, saw my link to Webaroo, came to my blog, and posted a comment on my entry. I read Sheece’s comment, and I followed the link to Sheece’s blog, where the current post is about searchradar.webaroo.com, which attempts to address the very problem that “searching is easy but finding is hard.” With the Search Radar FireFox plug-in, additional search queries are suggested to you in a column to the right while using Yahoo or Google. I immediately saw a connection to this post, which was still in draft form. So I added this story to the post. What a journey.

This type of suggested searches might be helpful to our end-users since the vocabulary varies widely for non-profits, organizations, or churches using our technology. Other searches might suggest a serendipitous path for our users to take to obtain the information they need. Take a look at the tag cloud that offers other search possibilities when you search for “agile” at searchradar.webaroo.com.

Blogs are a great way to find information probabilistically. I think there are other applications of technology that can help our end-users find information probabilistically as well, if they are ready for it.

The anti-social folksonomist

I’m finding that in del.icio.us I’m using keywords that no one else uses. Does this mean I’m an anti-social folksonomist?

This discovery reminds me that it’s important for findability to have tags that others will use too. On sites where popularity counts, it might be better to match others’ tags depending on what your end goal is – is your goal to categorize for your own easy retrieval later, or to tag so that others can see what you’ve tagged as being similar to their own findings?

For example, I use a tag called “blogthis” for items that have caught and held my interest but I don’t have time to immediately write a blog post about the item. Apparently no one else uses this tag! Or if they are using a “blogthis” tag, it’s those items aren’t shared.

Tagging is so easy, especially the way it’s implemented on sites like flickr and del.icio.us. But what conscious decisions are you making when you select a tag? This article has a good cognitive analysis of the act of tagging. To me, the lovely and freeing part about folksonomies for taxonomy is the decoupling of concerns about “matching” others tags and the ability to have multiple categories with similar meanings.

What does easy tagging mean for indexing professionals? Is it the crowdsourcing of indexing? Or is the printed book still a strong enough meme that a professional index is a requirement for certain media? Or is it a third, hybrid being, with identity and authority tied up in the tag selection?