Now that “the tumblelog” has morphed (# of google results being symptomatic) into “the tumblr”, a different beast altogether, I think it’s appropriate to analyze it. Call it a post mortem. When n+1 can do What was the hipster?, I’ll do What was the tumblelog? Potential accusations of navel-gazing acknowledged before moving on undeterred.
The first tumblelog, not by virtue of being completely unique (although it sort of was) but by virtue of being the thing the word was invented to describe, was Christian Neukirchen’s Anarchaia. Anarchaia was my favorite blog for years, until Neukirchen closed it down and started up Trivium, which was supposed to be different, but ended up being Anarchaia without the Anarchaia brand. The term was coined by _why the lucky stiff, the artist-cum-programmer who committed digital seppuku last year by pulling all of his numerous blogs and programming projects from the web (but the web will not forget). Here’s the essence of the tumblelog as an historical phenomenon:
I don’t think I’ve seen a blog like Chris Neukirchen’s Anarchaia, which fudges together a bunch of disparate forms of citation (links, quotes, flickrings) into a very long and narrow and distracted tumblelog.
The important criteria here, I think, are 1) disparate forms, 2) citation and 3) distraction. I remember in the early days of Tumblr, there used to be heated debates about what a tumblelog is and what it’s supposed to be, but in looking at it from a historical perspective, I think going back to the original sources makes the most sense. Criterion 1) says that a tumblelog consists of disparate forms of content. You don’t need me to go academic on you to tell you what this means: just look at Anarchaia. It mixes links, quotes, images, irc chat transcripts and occasional “thoughts” in a free-flowing fashion. Neukirchen summarized Anarchaia thus: “Anarchaia consists of 1163 posts in a span of over 3.5 years. There were 14358 links, 2437 pictures, 1186 IRC quotes, 2180 lyrics, 410 quotes, and 232 thoughts… totalling in 20803 items, 96553 lines, 504428 words and 5.5 megabytes of text.” Criterion 3) is related to criterion 1), basically stating that the tumblelog is distracted, i.e., it falls short of the laser-beam-like focus of niche blogs, preferring to veer all over the place in terms of both content, form and style.
The possibly contentious criterion is number 2), which states that the essence of the tumblelog is citation. You see this clearly in Anarchaia: at the nucleus of each little post is some form of outside citation: a quote, a link, a picture, a chat transcript. The tumblelog generates few original trains of thought (although it occasionally does, in the posts labeled “thought”). It expands upon, contextualizes, or simply cites the work of others. There’s little editorializing beyond the choice of which sources to cite and which particular portions of the sources to highlight. Blogging always was a two-headed beast, one tradition springing from the early diaristic blogs, the other springing from the early link blogs like Robot Wisdom, and both sides always accusing the other side of not being true or authentic bloggers. Tumblelogs clearly descend from the linkblogging family, but expand to also cover other media than plain links (pictures, videos, quotes, chat transcripts, &c).
Anarchaia’s audience rapidly grew after _why linked to it (“thanks to _why for making me popular”, Neukirchen wrote in Anarchaia’s last post). The concept, being a good one, was copied all over the place. One of the first, if not the first tumblelog explicitly modeled on Anarchaia was Projectionist. (When Tumblr launched, its default themes looked very similar to and borrowed tropes from Projectionist. Since I wasn’t involved in creating Tumblr, I can’t say how much of that was conscious choice.) Established bloggers wanted in on the fun, too: Kottke declared that he’d “been slowly moving kottke.org in a similar direction for awhile”; although I don’t recall John Gruber ever making the claim himself, there were plenty of people around (including me) that thought Daring Fireball was another example. But while these blogs had criteria 2) and 3) down, citation and distraction, for the most part they lacked 1), disparate forms of content. Daring Fireball’s Linked List is and was always a plain linkblog; until recently, Kottke.org was similar, although there are now some videos there as well. The tumblelog not only centered on outbound citation, it not only contained different kinds of content, it also presented that content in different forms: a quote didn’t look like a text which didn’t look like an image or a chat transcript.
For a while, tumblelogs were hot. Sites like Tumblr, Soup and Posterous turned up and made it easy and free to set up a tumblelog in seconds. Tumblr pioneered automated reblogging, which made it a lot simpler and more convenient to repost things from other blogs, and also made it a lot easier to discover new tumblelogs, since back in the day the Tumblr sphere was small and you could connect to most of the good ones by following a few reblog trails. (This is harder to do today, when popular posts are reposted thousands of times, and Tumblr has millions of users.) But in the process of popularization, the tumblelog got diluted and bit by bit lost its soul (with both good and not so good effects). For a time, Tumblr officially discouraged using the term “a tumblr” to describe a blog hosted on Tumblr, but it’s long since become standard, just like Twitter and Facebook and MySpace have become nouns not referring to web services or hosted content management systems but to individual accounts on those websites. Now, we have both “the tumblr” as a generic term for any website whose underlying system is hosted on Tumblr.com, the website, and also “the tumblr” as a subgenre of blogging that has its own conventions and poster boys and haters and lovers. The pure tumblelog is almost dead.
I call this “In Praise of the Tumblelog” because I have some good things to say about tumblelogs; this, however, doesn’t mean that I intend to piss on all the things that the historical tumblelog — it feels weird to talk about things that happened five years ago in a historical perspective, but that’s the internet for you; it evolves in dog years, if not faster — has developed into. There are many negative things that could and should be said (and frequently are) about “the tumblr” as a genre. I choose to focus on the positive aspects of its predecessor, “the tumblelog”; you don’t need me to make the contrast explicit.
The good thing about historical tumblelogs like Anarchaia was that they embodied the good qualities of the much-vaunted digital curator. Since there weren’t many tumblelogs, tumbleloggers couldn’t simply be meta-curators whose digital curatorship consisted in curating the already-curated collections of citations from other tumbleloggers. (I’m noticing the word “curation” and its derivatives are howevering close to semantic satiation here.) A tumblelog was a collection of citations, but it engaged directly with its source material; its citations were to primary sources, rather than to secondary sources who in term cited primary sources. Tumblelogs like Anarchaia also lacked ego or personal branding. They weren’t about their authors, they were about their authors’ interests. If your field of interests overlapped sufficiently with a tumblelogger’s, you could expect to find a collection of interesting sources to check out on a regular basis. You did, after some time, feel like you got to know the person behind the blog, but this was a matter of osmosis: their personality seeped through in their choice of sources and the very occasional, pithy comments that were sometimes appended to links. I don’t think there’s a single picture on Anarchaia of Christian Neukirchen or any of his friends, family, cats, dogs, llamas or whatever animals he may or may not have possessed. Early tumbleloggers had no interest in building up themselves and their brand.
A tumblelog was long, distracted and frequently updated, but never frantic or nauseating. Anarchaia updated once a day, more or less every day, and each daily post contained a wealth of microposts with links, pictures, quotes and so on. Many of the blogs I read to this day are ones that I simply had to begin following because they turned up so many times on Anarchaia. The thing that Neukirchen and his copycats (I use the word lovingly) were so good at was staying on top of a large number of sources about a only slightly smaller field of interests and making it look easy. I rarely went a day without finding at least one link that interested me, and I read Anarchaia religiously for years. More than that, I never got the feeling that I’d read it before. I was equally surprised each time I read Anarchaia and found I’d seen a picture or a link before, because it happened so rarely. I never got that “meh, that’s been making the rounds already” feeling.
The authors of the historical tumblelogs seemed to understand and take the consequence of the following realization: most of the good and interesting thoughts in the world are being had by other people. On the one hand, they took the consequence and so tirelessly worked to find and spread those thoughts without shifting focus onto themselves; on the other, they deprived themselves (and us) of the potential good thoughts they may have generated themselves.
Daily Meh started out as a pure tumblelog modeled on Anarchaia: long, narrow, distracted, impersonal, frequently updated, centered on various forms of citation. I’ve gradually moved to a more personal and essayistic style, with fewer reblogs and pieces centered around outside citations. I’m wondering if there’s a way to bring back the good qualities of the historical tumblelog without sacrificing the good qualities of the essayistic style.