Here at TwShiloh, we’re riding high after four consecutive months of increased traffic. Now, I don’t spend too much time sweating over readership stats (After all, how long does it take to count to five? eds.) but this presented an opportunity to think a bit more about the dissemination of intelligence products.
Intelligence products overwhelmingly continue to function on the ‘push’ model, usually via getting your name on a mailing list. Mailing lists seem to be used as a proxy for how important you are – or at least how ‘in the loop’ you are. I’m always wary of people who brag about how many such lists they subscribe to. Such lists are usually general in nature (‘Sign up to receive all of our products!’) which means people are likely to get a whole host of products that aren’t particularly relevant to them (even assuming -and this is a BIG assumption- that the products are quality ones).
Agencies, for their part, remain stuck in the numbers game. Success is determined to a great extent by how many products are published and how many recipients receive (not necessarily read or use) those products. Attempts to gather feedback from those that receive intelligence products usually involve inefficient (and ineffective) surveys that have abysmally low response rates. Among those that are returned, they’re marked by low information quantity and quality and their only real use is to bolster the position of one side or another in petty office squabbles.
And that brings me back to where I started. If you consider this blog an intelligence product you could do what most other agencies do and count the number of posts written (20% more than last year! We’re winning the war on terror!) or I suppose I could record how many outlets I push these posts out to (WordPress! Twitter! Facebook!) but so what? In the blogosphere, there might be better (but certainly not perfect, let’s not kid ourselves here) ways to measure ‘success’* (Go ahead…don’t skip this footnote…it’s kind of important).
For example. Blogs really thrive of a ‘pull’ system of readership. In most cases, people have to look for the information you write. Granted I’m at the mercy of the Google algorithm but still, people are looking for information and ending up here. I can see which posts are getting consistent views (as well as some rough indication of how long each page was viewed), what links were clicked through, what attachments were downloaded, etc. Those can begin to address issues of relevance and timeliness.
Blogs also allow for comments and discussion. This is a way for readers to directly interact with the author. There are few opportunities to do with in many current intelligence shops both because of cultural issues and technological ones. It does, however, allow for some insight to readers who are particularly motivated about a particular subject.
You can subscribe to a blog. Normally it is (kinda sorta) like getting on the mailing lists I bemoaned above. There may be some difference depending on how specific a particular blog is but generally, it’s still a scattershot approach.
Where they differ is in the ability to drill down on on content through subscriptions or tags. If you love when I write about Afghanistan (and let’s face it, how could you not?) but could do without all the zombie talk (Philistine!) you could just subscribe to posts with that tag. Or, if that’s too difficult, you could just click ‘Afghanistan’ in that tag cloud over to the side and see everything that I’ve marked in that way.** This allows you to get close (depending on how conscientious I am about tagging my posts) to building a personal information stream tailored for your needs. Tracking the specifics of that would not be particularly difficult.
Just like in academia, the number of times an article of blog post are referenced can be an indicator for the influence it has within the broader community. There’s no way to track that now since most shops exist in the digital realm only reluctantly and insist on using PowerPoint or .pdfs for their production which precludes things like linkbacks or just good, old-fashioned hat tips.
These tools can let an intelligence shop know (without bothering the customer) what readers are interested in and provide insight into how influential existing products are.
And this should be much easier than in the blogosphere. After all, I’m doing this part time. I don’t have anyone conducting outreach, working groups, or professional organizations touting my ‘product’. Yet, somehow, the word gets out there and that happens all the time with social media that has a more serious bent than just pictures of cute cats that spell badly.
So…what can’t those benefits migrate to intelligence shops?
Right now, agencies feel compelled to ‘rebroadcast’ or rehash intelligence product that are sent from other agencies. Let’s say the FBI puts our a product on subject X to everyone it its distribution list. Agency ABC gets it and will almost always do one of two things with it:
- Send it out to everyone on their distribution list (with no concern if people on that list may have also been on the original distribution list***).
- Usually in addition to #1 above, the agency will put together their ‘own’ version of the product. In most cases, this is essentially a cut and paste job with a couple of throw away lines about the local environment.
The risk, of course, is that some people on this second distribution will go through their own two step process and products start appearing like tribbels.
That system simply doesn’t work for most information. Sure, if there’s a zombie apocalypse or a meteor heading for earth, I suppose you want a redundant system to make sure that everyone gets the message. But when all information is handled this way people start tuning out…usually in an imprecise way that begins to mimic a system where information isn’t accessible at all. If people immediately transfer information from their inbox to their trash because the past 8 bazillion messages have been duplicates or poor quality, it’s like they aren’t even plugged into your network.
What is to be done?****
Here’s where intelligence communities could learn some of the lessons from the late 20th century. The internet provides a shared space which precludes the need to repeat the same information ad naseum. Instead, there’s the ability to link, quote and (most important) comment and riff. No such space really exists now, however, so we’re stuck assuming that no one has the same information.
The Intelligence Community’s ‘A-Space‘ is pretty close but it doesn’t have a suitable audience. Participation is hindered by the regular organizational pathologies, an unfamiliarization of technologies (see note below) and a general disincentive for people to generate their own content for consumption by others (which is why so many people are on this danged thing so much). Few people (like, very few) have it hardwired in their job description to generate content in A-Space. There are few opportunities (or rewards) for A-Space to be an equivalent to the internet where people actually spend their free time generating useful content (uh…go easy on that ‘useful’ talk, cowboy. We’re just hear until the LOLCats page refreshes. eds.).
But…that’s really where we need to go. The marginal efficiencies we’ve gotten after 9/11 in terms of information sharing are 1) not big enough to fix the problem that was identified by just about everyone and 2) are in danger of getting swamped by a glut of information that can’t be navigated or refined using existing tools.
The question really is if the powers that be will loosen their grip enough to allow their minions (that’s us) to begin creating and managing information on their own. Make no mistake…this will be a serious loss of control for them. There will undoubtedly be scary moments where people ‘go off the reservation’ in ways big and small. Right now, we don’t have any way to deal with that other than to discipline people or ask for resignations. That will have to change. We’ll need to distinguish (and have consumer understand the difference) between ‘official’ positions of an agency, individual assessments and opinions of analysts (or whoever), and off the cuff thinking which even the author may not want to associate themselves with after some consideration and reflection.
Those are big issues that are going to take some time to figure out. I think we’re long overdue to begin…
*Before we go any further, let’s define success because I suspect that’s a big part of the problem. From an analyst’s point of view (and ideally from everyone’s) success here should mean getting useful, relevant analysis to people who can use it in time to conduct and implement plans and resource allocation to affect a desired outcome.
It is possible (and I suspect common) that some people see that definition of success as being too difficult and so fall back on a definition of success that is the appearance of getting useful, relevant analysis to people who can use it in time to conduct and implement plans and resource allocation to affect a desired outcome.
In those cases, I’m afraid there’s not much to be done as we’ll just talk past each other. So…the rest of this post will concern those who accept the former definition of success.
**Forgive me if all this seems super basic but I’ve recently been reminded that for many in the law enforcement/homeland security/intelligence field, ground well trod in the social media world is virgin territory. Perhaps it shouldn’t be surprising but there are still large segments of the community that don’t know (or view with the same sort of suspicion that ancient mariners viewed the edges of the map) what blogs, wikis, etc. are or why anyone other than 14 year olds or those obsessed with celebrity gossip would waste their time with them.
In addition to the traditional bias against open source information there’s a different, deeper suspicion against any open source information that doesn’t have an ancestor in the traditional (pre-internet) world. But that’s another blog post…
***No exaggeration. In one recent incident like this an analyst told me she got 7 copies of the same alert from different sources. That does nothing but clog up your inbox and distract you from other, perhaps important work. Still, this way everyone gets to claim credit for ‘information sharing’.
**** ‘Что делать’ Indeed.