For the past three years I’ve been doing a sort of annual letter to shareholders/call for sponsorships a la Warren Buffet’s Sage of Omaha act, roughly around March-April. I am about two months late this year. I am just going to start calling this my annual letter from now on. I plan to make it approximately 5% more magisterially smarmy every year until people start calling me the Sage of Ribbonfarm (the name of a short-lived gag panel that I experimented with in 2008. I had to give it up because Yurij, my off-oDesk Russian artist, suddenly dropped out of sight. I sincerely hope Putin didn’t do something to him).
So if you consider yourself even a minor shareholder in ribbonfarm (through comments, guest posts, sharing, recommendations, playing couchsurfing host to me on my travels, sponsorships or whatever you’ve been doing to help keep this show going), this letter is for you.
Each year, I also add one line to my evolving business philosophy. In 2011, the line was go where the wild thoughts are. In 2012, it was go deep, young man. In 2013, the line is grow branches and roots. Continuing with the arboreal theme from previous years, this year, my line is practice artistic forestry. That’s the first topic on the agenda. Here’s the rest of the agenda.
- Practicing artistic forestry
- The state of the forest, in numbers
- The Web, it is a-changing
- Bitcoin and online publishing
- The future of longform
Let’s get started.
Practicing artistic forestry
I like the arboreal metaphor for online publishing because it really is very precise, and a fine source of decision-making heuristics any time there is tension between the commons-like aspect and the financial aspect of this show.
Artistic forestry is my term for environment stewardship in such situations. When it works, you can avoid two failure modes.
The first is the authoritarian high-modernist yield-management trap of “scientific” forestry, which is incapable of seeing forests as anything other than timber or unexplored oil fields. That’s the drill baby, drill mindset of SEO-focused bloggers. The second most popular post on this site is probably A Big Little Idea Called Legibility, which explores this failure mode.
The second is what one might call anti-authoritarian low-naturalist environmental preservation ideology which is opposed to all change on principle. Forestry based on high-minded but empty pieties and economic delusions (what one might call religious or mystical forestry). The sort of view that treats humanity and its technology as a purely destructive force turning a Garden of Eden into Mordor desolation. Bloggers who take this view typically quit after an extend stint of masochistic suffering. Usually with resentful, suffering-saint final posts declaring lost faith in humanity for failure to make their self-indulgent pursuit economically viable.
How does the arboreal metaphor help?
Forests are interesting things. They are an existence proof that biological systems can be really long-lived, perhaps even immortal. They are a good model for human social, as opposed to economic life (which is why people talk of having roots in communities… and then proceed inexplicably to infer that communities must also be entitled to roots in geographies like literal forests).
With every passing year, Ribbonfarm looks more and more like a forest, with a complicated root system that is largely invisible even to me (tools for completely mapping the online geography around a site simply don’t exist). This main site is sort of the gaudy big tree that’s easily accessible from the highway; the one that tourists short on time slow down to look at before speeding (or on the Web, bouncing) on. But there’s now a good deal more: two loosely affiliated Facebook groups, three tracks of meetups in the Bay Area, New York and London, a Hangout track on Google+ and a small annual barcamp.
But there are also aspects to Ribbonfarm that don’t map well to the forest metaphor. There’s the fact that I get most of my consulting leads through the site, three published books that bring me a nice steady trickle of revenue, affiliate revenues from Amazon via reading recommendations and sponsorships. There’s the fact that it does cost some money to run this site well (I recently upgraded to a more scalable and reliable, but more expensive hosting provider (WP Engine) that costs me $99/month plus overages during traffic spikes).
My first principle of artistic forestry is this:
Things that are primarily economic mechanisms should neither be over-analyzed before being experimented with, nor allowed to outlive their usefulness.
Which of course means the second principle is this:
Things that are primarily environmental stewardship mechanisms should never be compromised or endangered for economic gain, because they are not yours to endanger.
It’s like the editorial/business distinction in old media, except that it’s not.
To bio-morphize business mechanisms is to get sucked into dumb, reactionary views of economic life. While there are many things wrong with the mechanistic metaphor of business as a “wealth generation machine” it does have the advantage of encouraging a pragmatic and dispassionate attitude towards what you are doing. If it isn’t working, shut it down and try something else. It is far less traumatic to disassemble, cannibalize or scrap an obsolete machine than it is to chop down a forest that has stopped yielding fruit or quality timber. Equally, to not bio-morphize everything else is to restrict yourself to building very mechanical sum-of-parts things, rather than organic sum-greater-than-parts things.
That’s really all there is to artistic forestry: freewheeling trial-and-error with economic mechanisms, hard-line primum non nocere around things that you’re not sure belong to you (if in doubt, it doesn’t). Keep the forest antifragile, have fun with everything else. Or in Carse terms, go nuts with finite games; don’t endanger the infinite game.
Possibly the most intellectually interesting part of running ribbonfarm is managing the tension between these two principles. It’s a tension some people believe can be structured away using vehicles like a B-corp (benefit corporations that combine commercial and social missions; a model for the Clueless in my opinion). Others don’t recognize the nature of the beast until it’s too late. They realize there is a tension to manage right after things break. The link is to a rather sad post by Andy Baio about learning the hard way what was his to sell and what was not (he seems to have overcompensated and found the forestry religion though).
To the extent that an online venue attracts the sort of goodwill and unsolicited investment only a commons can attract, it needs to be managed like a forest with a view to sustainability. To the extent it represents value created by an individual that can be easily untangled and extracted from the commons, it can be managed like a business. I think B-corps are a bad idea because a fundamentally business-like vision is unnecessarily burdened with a social mission that might as well be separately pursued. Conversely, online properties often fail because a fundamentally hybrid thing is managed like it is an easily partitioned thing, or worse, like it is either purely a business or purely a commons.
In the first few years of this site, 2007-2009 or so, I could claim to being the sole creator of value around here. That’s no longer true.
That’s why artistic forestry is a good metaphor to keep in mind.
The state of the forest, in numbers
Artistic forestry is not just a pleasing high-level metaphor. It gets right to the heart of how you manage real estate online. Avoiding both scientific and religious forestry models is a matter of a hundred day-to-day decisions, many of them informed by numbers.
Since the beginning, I’ve avoided “optimizing” the site around any assumed model of how stuff gets discovered, read and shared. The scare quotes are important: 9 out of 10 people who pay too much attention to models, and metrics that measure performance relative to that model, have no idea what they’re doing. They’re like overconfident individual technical investors who think they can beat the stock market of online publishing by hacking distribution. Yes, it’s possible if you are one of a tiny minority, like the guys behind Buzzfeed or Upworthy. For the rest, optimizing is a recipe for getting killed the next time online distribution technologies or cultural fashions change.
For the typical small site owner, it makes far more sense to pursue a strategy similar to investing in index funds: forget optimization and just invest in the long-term future of online content. This simply means putting out stuff with predictable regularity, and constantly improving quality at the gate so the hit-rate keeps creeping up. If it is good, it’ll work on any distribution channel and weather any short-term trend. If it’s bad, it will die with with whatever technology or fad it is optimized for.
Optimizing for a particular assumed model of content discovery, sharing and circulation is basically a case of setting yourself up for failure through authoritarian high-modernist legibility seeking, which typically vastly underestimates the complexity of the online world. The result is exactly the sort of fragile yield management strategy that destroyed naively managed forests back in the day.
But eschewing active yield management does not mean you shouldn’t keep an eye on what’s going on in the sense of radio-tagging bears, sampling air and water quality, and so on. So let’s take a look.
First Google Analytics (which recently changed its language from unique visits to sessions and unique visitors to users for some reason). Here’s a view of the last three years. Click for full-size.
Not much to report there. Roughly comparing June 2013 – May 2014 to June 2012 – May 2013, we see a gain of 333,707-273,731 = 59,976, or a mean gain of just under 5000 additional unique visitors per month. Which is a completely meaningless figure because much of the difference can be attributed to the large transient spikes in a given year.
And that’s why kids, you don’t use means for stuff like this. You leave out big spikes and look at the typical rather than average behavior. In other words, you look at medians.
I haven’t computed medians, but eyeballing the data, I’d guess median monthly uniques for typical months have drifted up by much less: something like 100-200 or so per month (from the median ~20,000 uniques visitors two years ago to the median ~24,000 range now). In other words, the torrid initial growth years are over around here (which, if this were a pure business, would be very worrisome). Absent a major shift in what we do around here, we’re settling into a more sedate typical growth rate occasionally bumped up slightly by spikes. But really, there is no meaningful way to systematically interpret this stuff. The main reason to look at it on occasion is to make sure weird things aren’t happening.
On the other major front, feed statistics have flattened out far more sharply, most likely due to the Great Killing of Google Reader, which I think made a lot of people give up on RSS entirely. Modest gains from something like 5800 to 6400 over the past year in subscriptions, I’d guess, and sharp fall of about 30% in reach as far as I can tell.
As far as I am concerned, this probably means RIP RSS. It had a good run, but I think the center of gravity of content distribution has shifted pretty much irreversibly from deterministic syndication to stochastic social diffusion. Though it is an imperfect measure of social sharing, the (largely neglected, I am afraid) ribbonfarm Facebook page provides a rough indication of the growing importance of social discovery and sharing for blogs (as The New York Times appears to have finally officially discovered). Here’s the growth in Likes for the page since I put it up 3 years ago:
And here are the demographics of people doing the Liking (I haven’t yet gotten around to turning on demographics tracking on the site itself; that would be an interesting comparison).
This doesn’t even begin to cover the complexity of how people find and share stuff these days. There’s a constantly shifting, spiking ocean of aggregators, communities, streams, Twitter link-passing etc. out there.
Final little factoid. As with the rest of the Web, mobile devices are slowly becoming more important for blogs. Today, roughly 30% of traffic to this site comes from mobile devices. I had a mobile-friendly theme running for a while, but I’ve concluded it isn’t worth the trouble now. I suspect most long-form readers use mechanisms like Pocket and Instapaper.
The Web, it is a-changing
There’s a bunch of other things I want to sage about briefly, that aren’t yet showing up clearly in the numbers. But they signal an upcoming era of intense innovation and changes.
In case it wasn’t already obvious, for most blogs, the bulk of quality discussion and commenting activity has shifted to social media. There are two dumb responses to this you see across the Internet. One is to neglect the on-site comments section entirely, letting it degenerate into a cesspool of Youtube level trolling, idiocy and spam comments. The other is to eliminate it altogether, forcing readers to go into various controlled social media venues (such as dedicated Facebook groups or authoritarian hashtag use on Twitter) in order to strike up a conversation.
One result (I suspect) of the prevalence of the former dumb strategy is an overall increase in sophistication and intensity of comment spam. My theory is that public comment quality has fallen to the point that spam filters have trouble telling the legitimate stuff apart. From what I see here, I think Akismet (the main spam-blocking service) has been falling behind in the arms race a bit. There’s more weeding to do around here these days.
I’ve concluded that the only smart move is to retain an open comments section, keep it clean and lively, but not attempt to either drive that activity elsewhere, or try to pull activity elsewhere on-site. This means, unlike in the early years, the sheer number of comments is no longer as critical a health indicator. But quality matters more than it did before.
The biggest looming development in commenting technology is probably RapGenius style annotation. I suspect somebody will figure out how to do that natively for blogs. I like RapGenius, but can’t stand its garish look. So I am waiting for somebody to bring that stuff to WordPress.
In 2013, we witnessed probably the most dramatic evolution in visual design in the years I’ve been blogging. I don’t know if it has a name, but it’s the practice of putting huge images at the top of every piece of content, requiring you to scroll to get to the start. With many non-text sites, the rest of the page is also structured that way: stacked graphic panels that give a scrolling-coffee-table-book feel to the page. For text sites, the text begins below the large image. The rest of the layout tends to be minimalist. Often there are no sidebars or minimal sidebar content. While there is a rise in annoying excesses like parallax scrolling and unclickable pop-ups, the overall trend seems to be towards less clutter, a lot more whitespace and a clear aesthetics-over-functionality shift.
These large images are typically a notch or two above old-school stock photography in terms of quality and relevance to the content. The trend is big enough that the designer of the theme for this site, Chris Pearson, released a skin optimized for it recently, and I could probably switch to it with a weekend’s worth of work. I suspect three things are going on.
First, so many sharing mechanisms now pull an image from the original source, it has become an important way to grab more screen real estate in sharing streams. When I raised the question on Twitter, somebody (can’t recall who, sorry) suggested a great name for this aspect: sharebait lipstick. I don’t think sharebait optimization (SBO!) is the whole story though.
Second, I think many publishers are operating by the theory that such images have a special appeal for tablet-users (probably the fastest growing segment). The hypothesis is apparently that tablet-users want the magazine-like glossy packaging for their content offered by apps like Flipboard. Either that, or content publishers are betting that the rapidly growing image-first demographic (Pinterest, Instagram, imgur, memes) will soon overwhelm the text-first crowd.
Third, the much-hyped revolution in blogging called Medium, brought to you by the creators of Twitter and Blogger, has adopted this visual strategy across the board. Much of the content is mindbogglingly tedious, but I do think Medium represents an important evolution in blogging technology, and a wake-up call for WordPress. I also think Medium itself is ultimately going to be a dead end for various reasons (its uninspiring early adopter crowd, closed and commercial nature, and the odd, cultish management philosophy of its founders among them).
The best innovations from the Medium experiment are going to be exploited better by other platforms that figure out an open, distributed implementation.
I don’t know whether I am going to jump on this bandwagon or sit this one out yet. It depends if I can think of a good way to create or source good images. I wish I had the talent to draw quick little gag panels or cartoons for each post.
The ‘Web is dead’ meme
The idea that the Web is dead (in the sense of an open network of hyperlinked destinations explored with a browser) is one of those endemic sentiments that rises and falls with each new development. In 2014, native apps and the massive rise of video (and video-centric experiences like Netflix streamed directly to televisions, which barely seems like part of the Internet) are triggering yet another round of doom-saying. A few months back, a reader and young designer even emailed me, offering to help design some sort of magazine-like packaged reader app for ribbonfarm. Other readers periodically suggest turning each post into an audio podcast or narrated PowerPoint.
I frankly don’t know what to make of these ideas. There’s definitely a lot happening to online media, but the safest bet still seems to me the open, browser-accessed Web, whatever the content form. It is important to note that for all the power they offer, app ecosystems today resemble CompuServe/AOL style walled gardens, with no open alternative. That’s definitely not a model you want to bet on too much.
Overall, this sudden spurt of new developments suggests there is a big, unprocessed pile of innovation ideas around blogging that might be really hard to fold into an aging CMS like WordPress. This suggests that some smart from-scratch platform (not Medium) will win the next generation of blogging.
But important though all these developments are for online publishing, I think they all pale in comparison to Bitcoin.
Bitcoin and online publishing
As an experiment this year, if you want to support this site, your only option will be Bitcoin. Right now, there’s just a link to my Coinbase page and to resident bloggers’ individual accounts on the sidebar. I’ll be experimenting and figuring out other useful places to insert suitably tasteless cues and prompts. If you want to send bitcoins my way, go for it (I’d appreciate a note de-anonymizing yourself and including your email, via the message box). As more tools become available, I’ll be experimenting with them.
Until early 2013, I had a tipping plugging running on the site ($3 “Buy me a coffee” links at the bottom of every post) that generated a couple of thousand dollars over four years (I think I had it active 2009-13).
I shut down the tipping plugin because it was turning out to be more book-keeping overhead than it was worth. One part of the problem was high transaction fees. Another part was the awkward idea of “tipping” being applied to what is essentially a top-of-the-economic-pyramid cultural activity like writing. Waiters are some of the least privileged bottom-of-the-pyramid workers around. Writers, if they are managing to find the time and creative energy to write things worth paying for, are by definition among the most privileged workers in the economy. Whether or not they make money at it. The tipping metaphor sort of worked, but really, it’s an awful way to culturally normalize micropayments. All the other models had the same problem: tipping was the only workable metaphor and it only went so far. The buy-me-a-coffee plugin worked as well as it did because it tried to get away from the tipping metaphor, but it didn’t entirely succeed.
Something similar is true for online sponsorships. For the last three years, I’ve been doing an annual call for sponsorships that generated a total of $8625 ($2250 in 2011, $3750 in 2012, $2625 in 2013). This year, I am discontinuing that model as well, for a similar reason. It’s simply a very awkward industrial age model bolted uncomfortably onto online publishing. One that tries to force-fit blogs into the PBS/NPR mold (a big thanks to my sponsors though, the sponsor page, which now lists sponsors from all three years, is going to stay up).
The problem with mixing money and online publishing, it now appears with 20-20 hindsight, is that dollars simply represent the wrong economic paradigm: they are the medium of impersonal, low-resolution transaction economies, not highly personal, high-resolution relationship economies.
Any individual piece of online content — at least the well-hyperlinked no-blog-post-is-an-island kind — has a natural market price range that hovers just above free, due to the massive over-supply at every point on the quality spectrum (you could easily spend a lifetime reading out-of-copyright classics). Unlike offline publishing, which was shaped by scarcity (namely, the high capital costs of printing presses and TV/movie/radio production), hyperlinked online publishing is also most naturally a part of the relationship economy rather than the transaction, attention or gift economies. This is because finished blog posts like the ones on this site are an outcome of large, sprawling, ongoing conversations all over the web and offline. To a far greater degree than the work of industrial-age content creators, the work of online publishers is dependent on, and a result of, conversations with others.
As a result, industrial-style branding doesn’t help create a premium perception. What does create a premium perception is the sense that there is a deep-rooted commons around a piece of content. Wikipedia and Tvtropes project that perception. The New York Times doesn’t.
So it is simply incorrect to think of online publishing in terms of individual production and consumption decisions mediated by two-way transactions around properties with clear ownership. Blog posts are simply packaged partial snapshots of ongoing conversations in underlying commons communities. In my case, pretty much every post in the last year owes at least half its ideas to conversations with others.
In other words, “charging” for true online-native content is like charging for all of Linux instead of just for the value added through packaging and support of the sort Red Hat offers. Or like constantly going out for dinner with friends and always free-riding without ever offering to pick up the check. From the reader’s point of view, the fair price is close to zero not just because there are so many excellent free substitutes for any given piece of content. It is also near zero because paying a more-visible-than-others individual for the output of a larger, invisible and uncredited group of participants in a conversation seems unfair (as it should).
This does not mean, however, that online publishing is part of the free economy. It is just part of the too-small-to-meter economy. Content is hyper-abundant but cannot be produced at zero cost. It emerges from illegible, forest-like commons whose members can legitimately claim both partial authorship and a right to part of the rewards. But it does also get consumed by individuals and organizations who had no part in creating it whatsoever and ought to be treated like foreign trading partners. This means there is a need for mechanisms that can mirror the domestic and foreign economics of any kind of creative production activity.
The same goes for tools like WordPress and foundational sources like Wikipedia, Tvtropes and the like.
So far, there’s never been an economic model capable of capturing the complex nature of how all this works. Content has been free so far because the right market mechanism for it did not exist.
Now it does. Bitcoin, and more generally, the underlying technology known as the blockchain, have the expressivity to accurately reflect the socio-economic structure of creative production under conditions of abundance. Bitcoin can be used to construct an economic model for artistic forestry. As a simple example, it should be possible to set up arbitrarily complex, flexible and automated microfluidic routing mechanisms to automatically move even tiny fractions of bitcoins around to various parts of the forest ecosystem that sustains a site like this (contributors, commenters, open source projects). Security issues, chargebacks and high transaction fees made that effectively impossible with dollars.
Bitcoin is the start of making all that possible, and it seems increasingly clear that it is going to happen. I have some ideas about how it might happen, and specific developments I am watching out for, but I figured I ought to start experimenting today. So stay tuned. The hydrological landscape of this forest will likely change a lot in the coming years.
For now, the Bitcoin presence on this site won’t look very different from old-school sponsorship and tipping. The difference is, this has room to grow in ways that dollar-based models do not.
The Future of Longform
I want to close out this letter with some thoughts on longform.
Today, longform is just one of at least several dozen major forms of online creative expression. The number of forms is growing every year. Here’s an incomplete inventory of the environment of content forms today, with which longform must compete for attention. Besides longform, I have interests in perhaps a dozen of these forms (the starred ones):
- Virtual reality/Augmented reality (coming soon)
- 3D content (via SketchUp)*
- Original video (instructional, viral, TEDdy, movies, music)
- Derivative video (soramimi/mondegreen, parodies)
- Video conferencing (Hangouts in particular)
- Audio music
- Game content (from scripts to spectator videos to machinima)*
- Shorts (<300 words, including poetry)*
- Long form (1000+ words)*
- Curated link blogs
- Serialized online books*
- Listicles (now being disrupted by tweetstorms)
- Glossaries and dictionaries*
- Image memes
- Structured content (such as recipes or lyrics)
- Aggregator discussions
- Bulletin boards (Reddit, HN, Facebook groups)*
- Narrated slide decks*
- Strip cartoons*
- Gag panels*
- Full-page or book-length comics*
- Animation (classical)*
- Animation (novel, like RSA Animate)
- Pinterest boards
- Instagram feeds (selfies being a sub-category)
- Video visualizations
- Interactive visualizations
- Annotations (RapGenius)
- Maps of all sorts*
- Offline meetups catalyzed online (yes, running these well is a form of creative expression)
- Offline conferences catalyzed online, and program content thereof
- Hybrid online+offline conferences
Looking at this exploding universe, it strikes me that there is a close parallel to the explosion in programming technologies. That gives us a simple framework to understand what’s happening.
- At the bottom you have assembler like elements: tweets, memes, poetry.
- Next up, you have the equivalent of shell-scripting: tweetstorms, listicle posts.
- In the middle, you have longform, which is something like C in the picture: a mid-level language for compactly and efficiently building very complex systems programs. I think it has no competition in this range.
- Above, you have languages that involve more complex abstractions such as PowerPoint (which is rather like Object-Oriented Programming, in its design intention and and practical inability to support the creation of truly reusable slides) or wikis (which are very vaguely like functional programming in my inexpert opinion).
- You also have all sorts of complex kinds of glue content corresponding to library objects, includable content, runtime environments, Web APIs and so forth.
When you consider the complexity and try to figure out what the hell is going on, there is only one inescapable conclusion: we’re liquefying all content into one giant Big Hypermedia Book. John Donne was right: “all mankind is of one author, and is one volume; when one man dies, one chapter is not torn out of the book, but translated into a better language; and every chapter must be so translated.”
So that’s it for the 2014 letter. Get on the Bitcoin bandwagon and send me some if you want to experiment with the future of publishing, and help keep Ribbonfarm National Park going. But if you just want to visit and enjoy yourself, that’s fine too.