The Internet of Garbage

As an old white guy, I keep having my worldview shifted by people who tell me things that I hadn't noticed because my world was primarily engineered by old white guys... this is yet another example of some wisdom from a non-old, non-white, non-guy

Here's Sarah Jeong on The Internet of Garbage. She's smart, and has several good points to make.

A Critique of modern discourse

A Metafilter story got me to writing about the problems with modern discourse as seen on the Internet. I thought I would spend some time refining and extending those thoughts here. By no means is this the definitive word on what's wrong with the internet, but I hope that a cogent summary might help fix things.

Time Pressure

1. First post phenomenon - On a site such as /., comments are sorted by default in the order they are received  This means that the first post gets the most feedback, and sets the tone for the rest of discussion. This results in a bunch of hasty, emotional, not well considered crap that then has to be moderated around in order to extract value.

2. 15 Minutes of fame. - Any post gets attention, and then gets pushed under by the oncoming store of other stories in 15 minutes, maybe a bit more if you're evil like Carmen Ortiz... but only a bit more. This means that any effort you do put into something isn't going to pay off much.

Text is Messy

3. Comments on a message board are usually undifferentiated text, sometimes with a bit of formatting. There is no easy way to tell visually what the meaning is without forcing oneself to read it all, and then start to draw out conclusions about structure, agreement/disagreement, etc.

Moderation is a kludge

4. Most moderation systems are a layer of code designed to filter out crap, and help extract some value from an otherwise overwhelming amount of undiluted text. Like spam filters, there are behaviors and techniques that get used to route around them.

  The most common form of moderation system is to have a ranking system (like/dislike), which turns commenting into a popularity contest. The funniest or most vocal viewpoints drown out everything else.

5. Popularity contests are 1 dimensional - There are lots of reasons someone might want to flag a post (which isn't quite granular enough for me, but you have to start somewhere)... agreeing with a post as "having value" is the standard here in the blue, but elsewhere it's a direct measure of the groupthink agreement. (/.  for example)

Wouldn't it be better to have multiple dimensions of ratings? Factual/wrong, Conservative/Liberal, Cheap/Expensive, True/Lie, etc.?  They wouldn't have to be the same set of things either, but it would be easier to code for lets say Funny, True, Insightful as separate (orthogonal) dimensions.

6. The missing half of Facebook - Facebook doesn't allow the inclusion of negative votes, so it's actually only 1/2 dimensional.

Anonymous people are assholes
7. Anonymity allows people to say things they'd not say in person. Remember the Id in Forbidden Planet?

So, how can we help fix it?
There are a number of strategies to be used to help fix this.

Use your blog more
The first is to blog more, and comment less. If you find yourself writing something in a comment that seems to be really insightful, make a more refined blog post out of it (linking back to the discussion for context). Blog posts are better than comments on someone else's site for a number of reasons:

• You own the post, and can tweak it later if necessary. This times into and builds your reputation internet-wide, instead of in on little corner of it, subject to whim.
• The post can be found later, as a stand-alone piece to be referenced, and can be more self-contained.
• Time pressure is less because blog posts can stay "popular" for years. A few hits per month done a few dozen times means you're always getting feedback and links.
• If you've already written about a particular view, link back to it, and perhaps tweak it a bit with improvements.

Build a better commenting system

Someone (me?) needs to come up with a better commenting/moderation system, that allows the multiple dimensional rating that I discussed above. It would be nice to discover visually how the arguments are inter-related about a given topic, to allow one to focus ones time better, and improve signal/(signal+noise) ratio.

Review your comments periodically

It would be good if we all where to re-read the things we've written with some distance of time, and get a better sense of ourselves. This can help us to all be better writers and readers.

The sad state of US weapons technology

I saw this while I was waiting for the train today.... it was sufficiently surprising to warrant a $2.00 purchase of a newspaper... my first impulse purchase of a newspaper in more than a decade.

I can't believe how stupid the situation is.... go read this and weep....

• 

• DECEMBER 17, 2009

Insurgents Hack U.S. Drones

$26 Software Is Used to Breach Key Weapons in Iraq; Iranian Backing Suspected

What this means is that our $10,000,000 drone can be subverted with a laptop for less than $1000.... a 10,000:1 advantage. It doesn't end there of course.

If the signal can be detected, it can be tracked, and jammed... possibly with something like a $50 microwave oven.

Spread spectrum technology was invented in the 1940's to allow us to securely control torpedoes... I can't believe they haven't come up with a communications channel for this stuff that can't be a bit harder to detect in the last 60 years with military budgets.

UGH!

--Mike--

Indeterminant Intermediaries Imminent

Doc Searls recently wrote a blog post - The Revolution Will Not Be Intermediated in which he states some hopeful things:

We still seem to think that progress on the Net is the work of “brands” creating and disrupting and doing other cool stuff. Those may help, but what matters most is what each of us does better than anybody or anything else. The term “content” insults the nature of that work. And of its sources.

The revolution that matters — the one that will not be intermediated — is the one that puts each of us in the driver’s seat, rather than in the back of the bus. Or on a bus at all.

As much as I like Doc, and wish he were right, I can't help but be cynical for a number of reasons. I'm hoping to be proven wrong, and/or convinced that there is room for hope. Please find some flaws with my logic...

The future of the live web is in doubt, for good reasons.

The long tail - theory vs practice

The Long tail theory holds that everyone should have a blog, and they will have a voice. This is widely interpreted to mean that the tools are sufficient to get your views into public. This theory is valid, but as with any description of reality, it fails at the edges, and a better model of understanding should eventually replace it.

The blogging tools now have succeeded in making it possible for anyone to write an opinion, and have it accessible in an instant from any Internet connected PC in the world. There are limitations when Governments, or ISPS get in the way, but for the most case those limits are exceptions _post _facto__ and do not impose prior constraints.

However, the freedom of speech enshrined in our US Constitution is worthless without the freedom of assembly, which is the freedom to hear someone else speech, and to have conversation with them. It is here that the intermediation sets in.

Historically, the cost and logistics publishing provided a natural damper on the quantity of material, and people generally focused on getting quality of content up to the point where it seemed acceptable to expend the effort to publish and distribute it. The nature of the Internet and the web has done away with this limitation, and made a more basic limitation which has always been there, apparent to all... the limits of human attention.

Now that we all have fast and effectively zero incremental cost to say what we want, the problem now becomes connecting with an audience. Out here in the long tail, where this blog will reside for the foreseeable future, the loyal audience is very small, mostly family and a few friends. In this blogs history, the rare moments in the spotlight are for things unrelated to the normal subject matter which I tend to be interested in, for example... this post "Songs about teamwork" is the source of more than 50% of all hits here... ever.

This means that while the tools make it possible to speak, there is very low probability of being heard when you take the time and effort to set up a blog, especially if your areas of interest are many and varied. An intelligent response is to consider alternate ways to route around the attention problem, and to write where the readers are.... thus putting content on social networking sites, into emails and other non-live-web related channels, and into comments, twitter, and other aggregated sources of attention.

The long tail in practice works out far from theory, we now all have a voice, but we have to find an audience... everyone has a soapbox, but the public square is full of people with ipods, tuned into their own narrow circle of interest. The key is to find a topic already in progress, and to attempt to join into the conversation. We all have our own bus (or car), now we need passengers who are going to our destination.

I'm a commuter, I regularly share my trip with a circle of friends on the South Shore railroad daily to Chicago. We all give up a bit of our privacy to have a shared experience that is more cost effective and efficient for each of is... this is the same logic behind using social media sites, twitter, FaceBook, etc.

If the train fails to arrive, gets delayed, etc... we route around it, share rides, etc... just like when twitter fails, etc. The metaphor can be extended quite a way if you like... it almost writes itself.

It's a silo, sure.. but that is easy to ignore if it stays out of view, and your friends are there. It's only when you have a dispute or disagreement with management that the situation becomes unacceptable.

With this in mind, it's not hard to disagree when Charles Arthur wrote in the Guardian that The long tail of blogging is dying. I'm spending upwards of 4 hours working on this post, if not more... and I might have 30 people read it, or maybe a few thousand if I'm VERY lucky and it goes viral. The residual effects will be very, VERY limited, in my estimation. It's only faith and the need to speak my piece which keep me going.

With all of this rambling in place, let's go back to the quotes from Doc again...

We still seem to think that progress on the Net is the work of “brands” creating and disrupting and doing other cool stuff. Those may help, but what matters most is what each of us does better than anybody or anything else. The term “content” insults the nature of that work. And of its sources.

I think it's better to say that we're used to having reliable labels on our content, enforced by law (Branding, Trade Mark, Copy Right). These labels made it possible to aggregate many different pieces of stuff under one reputation umbrella.

The main challenge in the live web is to figure out how to aggregate enough reputation to be easily quantified by anonymous third parties for purposes of deciding where to spend their limited resources of time, attention, energy, and possibly even money.

Google tries to use PageRank as a proxy for quality, on a per site basis... which isn't a fine enough grain to measure quality on each piece that someone like Demand Media might produce, so the quality is going to be spotty. The idea of tying PageRank to a person is interesting, but that really won't work out in the long run, because we all have different areas of expertise. For example, you might decide to trust my opinions on computers and security, but you probably disagree with my opinions on political matters. These areas are orthogonal to each other, so there need to be many degrees of freedom in ranking and labeling content.

Google is slipping, but all engines slip from time to time. They can only keep up a certain pace, as their infrastructure gets more complex and entrenched in a certain paradigm. They have made excellent use of the resources they have available, especially the 20% time they give to feed (and harvest) what would otherwise be the seeds of their destruction.

I'm a geek, and I tend to think in terms of technical solutions to problems... I see this all as a matter of lack of metadata and the infrastructure to support it. I think over time it'll all pull together, it's just a question of timeframes. It took 500 years for print to get to where it was when the Internet took off... it may take a few lifetimes for the Internet to get figure out to the same extent.

I'd of course like to skip some of the learning curve and get more benefits now, while I can appreciate them, and pass them along to my daughter.