Why do bloggers kill kittens?

A couple of days ago I posted a rather aggressive link through to 2lmc the other day complaining about their post Most read blogs least original which cited an article from Wired News called Warning: Blogs Can Be Infectious (itself quoting HP’s Blog Epidemic Analyzer – I could go on). My link read: “I’ve noticed that people are much less intellectually rigorous when they read articles that they agree with. Case in point: Most read blogs least original, says blech 2lmc”.

Since posting that, Paul and I have been having an extremely civilised discussion behind the scenes about some of the issues surrounding that article and our individual responses to it. I won’t post his e-mail to me for obvious reasons, but I wrote such a lot in response (and I think there’s enough stuff in it worthy or argument if not agreement) that I thought I’d post it up here to see how far out on a limb I was. I’ve added in some hyperlinks, edited a little for clarity and extended a couple of sentences here and there to make it readable in a larger context. Comments welcome as ever.

Dear Paul,

Firstly I should apologise for the link-text – I was halfway through writing a longer post addressing your post when I got distracted by work. When I came back to it, with the initial enthusiasm gone, I couldn’t motivate myself to finish it. I had another seventy tabs open in Safari that were squealing for attention, so I thought I’d linklog my response instead. It came out a little more snipey than I’d intended, and for that I apologise.

You say in your response that with regards to my site, you enjoy some of my larger posts but find my linklog slow on the uptake. I can’t say that I’m surprised at the latter – my linklog is exceptionally slow on the uptake. It’s that slow because I stockpile things that I’d like to write about in greater detail and then – when I realise that I’m not going to have time – I end up posting them on the linklog. I have some links stashed away in a bookmarks folder called “Backlog” that are at least four or five months old that will probably end up on the linklog at some point once I’ve finally accepted they’re not going to have the fuller treatment they deserve.

First things first, let me admit that I’d like to be the first one to post all of those links (or at least up in the first few). Certainly as an aspiration, to be first in the flow of sexy fat links would be glorious. But I want to make it clear that even if I can’t get them out that quickly, I still see considerable value in getting them out way after the fact.

My reasons are manifold. Firstly, we get straightaway down to the distinction between weblog as written for an audience versus weblog as written for me. Now clearly it’s not just for me – I’ve have to change what I write on occasion because there are people out there reading (my mother, my brother, some potential employers) who I have to be aware of. I don’t write in the same way as I did when I started. Then I could bitch about people I didn’t like and talk about my life without feeling particularly exposed. I was talking to strangers with no impact on my life. Now I’m not. But while my weblog isn’t any longer ‘just’ for me, it’s definitely not just for an audience either. I use my weblog as a searchable archive – a repository of things that I’ve seen and read and that I thought were interesting, I use it to record thoughts that I think might be useful and that otherwise I’ll forget. I use it as a notepad, as a chronicle, as a place to store my photographs. There’s an interplay between trying to be fresh for other people and not really giving a damn about other people. I think this comes back to my understanding of a weblog as a representation of a person online – an avatar with a voice. A self-representation is about being both true to yourself and knowing how to self-edit in different circumstances. That’s what a weblog is to me.

Secondly I operate with an understanding of my links as a kind of microcontent vote (also here and here). It’s the idea that by linking to something I say, “Yes – this deserves some of your attention – this is a good thing”, and that the more sites that do that the more attention something will get. So by voting for something I like (alongside dozens of other people) that thing becomes incrementally more visible in Google, Blogdex, Technorati, daypop. Also, in turn, those people who don’t read a lot of other weblogs but read mine also get exposed to it. And those other people who have weblogs may choose to pick up that link and post it to, thus exposing more people to it in turn – putting their votes behind it too. Massive link propogation (as far as I’m concerned) is not a bad thing at all – it’s how the web determines what’s worth reading.

Now the other question posed in your piece is why people would come to a site like plasticbag.org or kottke.org or BoingBoing when most of the links on those sites come from other places. Well firstly, I’d like to state up front that I can’t comment on where Jason or the BoingBoing crew get their links from. In my case, yes, it’s certainly true that the links I publish have often been posted somewhere else first (either because I’m slow to publish or because I found those links elsewhere), but I want to make it clear that even in these circumstances there’s you can characterise that behaviour in very different ways: is it mining for links as content on the one hand or is it collating the good things you’ve read on the other? I’d say it was more of the latter.

With regards to why people read our sites, I imagine to an extent it’s just because of the time we’ve been around and carrying readers with us since we started. (At this stage I should probably state out loud that I’m nowhere near in the same league as Jason and BoingBoing by the way – they’re a full order of magnitude ‘larger’ than plasticbag.org). I would also imagine that they read us because on occasion our commentary or original work is valuable enough that people are prepared to read all the other guff we write or link to as well.

But to be honest, I think the question is a bit of a red herring. First and foremost weblogs started off as carefully-selected links (maybe with commentary) to a whole range of other sites. There is an element to weblogs which is nothing but aggregatory and filtery in nature. Pretty much all weblogs (including 2lmc) find their links either from mainstream news sources, or their daily searches, or from some mailing list somewhere or from another weblog that they read regularly. That’s absolutely normal behaviour. I suspect, then, that the main reason that people read our sites is because we’re relatively consistent in the selection of links that the people who read our sites find interesting, wherever those links originally emerged from (and – in fact – however old they might be). To an extent (and this may seem tautologous), popularity has got to be to some degree correlating with how interested people are in what you produce.

Which brings me to the question of originality. My weblog grew up online with ‘via’ links on weblogs. That was the way we did things. In fact I vaguely remember us starting to use them because at the time the commodity of a good link seemed the most valuable that any weblog had. As time has passed, I think the culture has changed a bit – and there are all kinds of reasons for that. Personally, I no longer often know where I found a link. I’ll open up dozens of tabs from a cursory reading of my NetNewsWire subscriptions and then close down the windows that aren’t particularly interesting to me. Then I’ll store the pages that are interesting until such a point that I’m able to give them the attention I think they deserve. In that process I often simply lose track of where I found them. Moreover, the most practical structure of a linklog in my opinion is – well – a link. Nothing more. That seems to be the simplest and clearest way of referencing something, and as a result there’s not even always a space for adding in a via link or a reference to the originating source.

In fact as time has gone by, I’ve increasingly come to the opinion that links are everywhere and that referencing where you found the link alone is no longer quite as necessary or as useful as it once was. When I know the source and when I have the ability to link to it, I’ll still make that reference, but nowadays I’m more interested in useful links to useful material rather than just a reference to the person who told me about it. I look for commentary – something I can actually cite that had a useful contribution to make on those links concerned (however short a line that might be). A lot of other people may disagree with this as a strategy – and god knows it’s not something that I’m entirely unconflicted about myself. But I think if you consider that (as is most often the case) the originating sites are linked to anyway – “blog-rolled” in my sidebar – then I’m not sure it’s as big as anxiety-producer as it might be.

[Quick sideline: Andrew Orlowski gives webloggers hell for what he considers ‘circle-jerk’ or ‘meaningless self-promoting cliquey’ inter-linking or whatever, while other people who are not fans of weblogs and weblog-culture take us to task for not doing those things because then it’s an elitist culture within webloggery.]

With regard to the article at Wired and my slightly barbed comment that “I’ve noticed that people are much less intellectually rigorous when they read articles that they agree with”, what I think is interesting about the Wired article (and quite a lot of the stuff that’s written on the HP site) is that it’s far from immediately obvious how many ‘stolen’ links exist, how accurate their linguistic analysis is (versus what the proportion of links use headlines or site-names and reference material that comes into the public arena at specific times, like – for example – a news story), or whether this kind of behaviour is limited to specific groups or cultures online or is rife without weblogs as a whole. Nor is it clear whether or not the originating sites are linked elsewhere on the weblogger’s page. This is not to say that I dispute the results, but that I wait to be convinced that these questions have answers that could support the kind of headline, “Most read blogs least original”, which really does appear to be almost Daily Mail-like in its bluntness and lack of any qualification. I have a feeling that the originating study will eventually demonstrate a rather more nuanced and qualified version of the picture – one with fewer value judgments imposed up on it. Because at the moment (Why do bloggers kill kittens), I suspect that they’ve been taken rather out of context…

Yours, Tom Coates

Addendum: Danah Boyd on Blog Attribution