Scienceblogs, the science network that was my old (where “old” = “a few days ago”) haunt, is in revolt. Okay, well maybe the network is not in revolt, but there is at least a minor insurgency. Yesterday, the amazing force of blogging known as @Boraz, left the network (be sure you have more than a few minutes if you are going to read Bora’s goodbye letter.) Today, the biggest fish of them all, PZ Myers has gone on strike (along with other Sciencebloggers.) Numerous other bloggers have also jumped ship (a list is being kept by Carl Zimmer here.) This is both sad, as I personally think the Scienceblogs network does contribute significantly to spreading the joys and tribulations of science, but also a bit exciting for, as Dave Munger points out, this also represents the prospect of new networks arising and hopefully pushing the entity that is known as the science-blogosphere forward.
I myself am not much of a blogger. What I write here is for my own personal amusement (so if you don’t like it, well I don’t give a damn, thankyouverymuch) and, frankly, to distract my fellow quantum computing researchers from getting any work done (ha!) I do enjoy writing (literature major, you know) and also enjoy trying to write coherently about science, and sometimes, as a consequence, I get read by people who aren’t here just to hear about the latest and greatest in quantum channel capacities. That’s great, but I don’t really consider science blogger as my defining characteristic (my self image, such as it is, is more in the line of a hack who has somehow managed to remain in science—despite being almost a decade out of graduate school without a tenure track position due in large part to being stubborn as hell. But that’s another story.)
But, even though I don’t consider myself very bloggerrific, having had a seat at the Scienceblogs table gave me an up front look at, to use a silly term, new media, and in particular at the notion of a science network. So to me, following Munger’s post, the interesting question is not what will become of Scienceblogs in its current form, but how will the entities we call science networks evolve going forward. Since there are a large number of Sciencebloggers jumping ship, it seems that now would be a good time for a new media science mogul to jump into the fray and scoop up some genuinely awesome bloggers. So the question is, what should a science network look like?
To begin, I can start with Pieter’s comment a few days ago:
…I never fully understood the need for successful bloggers to join an umbrella organization. Did you get more readers when you moved to Pepsiblogs (good one!)?
That is exactly what I thought when I was asked, clearly by some clerical error, to join Scienceblogs! Having joined, I can say that yes, it did increase my blog traffic. But I think a science network also adds something else.
First of all, there is the fact that there is a front page which contains significant “edited” content. It is edited in the sense that the powers that be have a large say in selecting what appears there in a highlighted mode. This great because even the best bloggers, I’m afraid, generate a fair amount of posts which aren’t too exciting. A discerning eye, however can grab the good stuff, and I regularly go to the front page to see what exciting is being blogged about. I’m not sure that the front page of Scienceblog is the best way of providing an edited version of a blog network, but I do think that it is heading in the right direction. So in thinking about moving forward, I wonder how one could change this editing and give it more value. For instance, is the fairly static setup of the front page the right way to go, or should there be a more dynamic front page?
Another important property of a science network is in building discussion, and by discussion I don’t just mean a bunch of people agreeing with each other. For example, Scienceblogs has a “buzz” where articles on a featured topic are posted on the front page. Sometimes this content presents a unified view of a topic, but mostly you get a terrific variety of opinions about a subject. Now I won’t argue that this diversity of opinion is huge: for instance you aren’t likely to find the Christian view on topological insulators, but you are likely to get the opinion of a large number of scientists or science journalists from a variety of areas. This solves, for me, one of the worst problems with my blog reading: only following blogs for which I am predetermined to agree with the blogger. Further this content gives rise to a genuine discussion among the bloggers in that they actually will read what others have written as opposed to just sitting on an isolated island (okay well I rarely read what even I’ve written, hence the horrible typos and grammatical gaffs that liter my writing.)
Third a science network like Scienceblogs serves as a proxy for a certain amount of quality. Despite me trying to bring this quality down, I would say that some of the best science bloggers around have or have had a blog at Scienceblogs and this lets the network serve as a proxy for having to read a bunch of blog posts to see if the person has something interesting to say, or whether they are not worth your time.
So those are at least some of the reasons that a science network is good. I must say, in thinking about these reasons, however, that I can’t completely convince myself that these amount to enough to justify the science networking idea. Many high quality bloggers get along just fine without such a network.
Which brings me to the real subject of this post: how would I redesign Scienceblogs?
Well the first thing that comes to mind is better tech support. Okay, just kidding. Kind of.
Actually I do think there is a valid point in this dig at tech support. One of the hardest things for me while I was at Scienceblogs was not being able to dig around and modify my blog in the sort of way I can do on my own hosted server. Why is this important? Well, for example, Scienceblogs does not currently have a mobile version of the website. (Mea culpa: at one point, back when I was writing iPhone apps, I emailed the powers that be at Scienceblogs asking if they wanted me to design an iPhone app for them. I got crickets back in response. Later this came up in discussion among Sciencebloggers and the powers that be emailed me asking for more details. This was in the middle of the impending arrival of baby Pontiff, so I never followed back up on this. I feel bad for not doing this, but it seems that if the management was really serious about this they could have pursued numerous other, um, really qualified people. Note that it took me about 30 minutes to get a mobile read version of my blog setup when I moved back here, and yes this is different than an iPhone app.) But more importantly, technology has that important property that it is constantly changing. Anyone who wants to build a network of science blogs should probably seriously consider that the infrastructure they are building will be out of date every few years or so and need major upgrades at a fairly high rate.
For instance, Scienceblogs should have been among the first to offer an iPhone app, an Android app, an iPad app. Scienceblogs should think of ways to incorporate its tweeting members: as it is, as far as I know, Scienceblogs doesn’t even keep a list of its members who tweet. Scienceblogs, a network about science, doesn’t even have LaTeX support for heaven’s sake, let alone, as far as I can tell, plans for how things like html5 will change what one can do on a website. What will happen to Scienceblogs when technology adapts? Will it adapt too?
So I think, if I were going to start a new science network I would start with an incredibly dedicated hacker. A quickly adaptable platform is a prereq and if you don’t start with a good base, well then you are just going to be out of date pretty quickly.
But of course there is more to a platform than just the tech behind the scenes. There is also the content. I have a lot of admiration for the people who have been the behind the scenes editors at Scienceblogs and I think this is part of the network that worked the best. I do wonder, however, if they have enough editorial control: that is it would seem to me that they should have an even more expansive roll in the network. And it’s not clear to me that there should be as large of a separation between their magazine Seed and the Sciencebloggers. I would wager that many people don’t even know that Scienceblogs is related to Seed or that Seed exists at all. And here is where I think one needs to get a little radical. Seed should (as roughly suggested by Bora), I think, give up it’s print magazine and fold Seed into Scienceblogs. High quality traditional media pieces like those Seed produces are great. So why can’t they be part of the network in an integrated way?
Well these are just my silly initial thoughts about re-imagining science networks, when I should be busy changing diapers. And certainly I don’t know what I’m talking about. But read the disclaimer in the upper right of this blog. So don’t say I didn’t warn you!
Knock, knock, Chad. Please set up a blog network for me 🙂 Just kidding.
“I’d also formally encourage cross-blog discussions in some way– not requiring posts, but asking bloggers joining the network to try to carry on that sort of conversation– and I would avoid back-channel discussions as much as possible.”
I wonder how to achieve this. Is it just a function of size and the bloggers or is there some technological change that could facilitate this (besides eliminating the back-channel :))
Things I’ve thought would be interesting: threaded blog posts, show “similar posts” during the writing of a blog post, the ability to tag other blogs for certain posts (channels at SB effectively did this.)
What about different levels of platform control: for instance the MT system on SB drove some of us nuts, and I understand the need for a centralized template from the ad/revenue perspective, but I wonder if there isn’t a better way to achieve this.
Since I’ve already thrown most of the day away on this business, I might as well throw in a comment here: I think that if you really want to have a network of blogs that is more than the sum of the component blogs, you need to keep it small.
When ScienceBlogs launched with 14 blogs, it was easy to read everything posted, and keep some connection to the rest of the site. By the time it had expanded to the pre-meltdown 70+ blogs, it was impossible to follow everything, and the network pretty much fragmented into a bunch of sub-networks around different disciplines or topics of common interest– at the moment, I pretty much only read the other physics and astronomy blogs on SB, plus 3-4 others that regularly write on stuff that interests me.
If you keep the network small, and encourage discussion between blogs on the network, you’ve got a chance of doing something interesting. Knowing that other people from other sciences will be reading helps keep people writing for a slightly more general audience, which makes it easier to talk across disciplinary lines. Once you have enough chemists, say, to sustain highly technical back-and-forth about details of chemistry, it becomes hard for others to follow that discussion, and they tend to tune out.
If I were trying to build a blog network from scratch, I would pick a set of bloggers who blog about their specialties for a fairly general audience, collect them together, and provide some internal variant of TrackBack to automatically serve up links to (potentially) related posts elsewhere on the network. I’d also formally encourage cross-blog discussions in some way– not requiring posts, but asking bloggers joining the network to try to carry on that sort of conversation– and I would avoid back-channel discussions as much as possible. A number of the problems with ScienceBlogs are directly traceable to the back channel.
But then, nobody is knocking on my door asking me to build a blog network for them.
Not to continue on with the SB bashing theme, but… As to the technical problems on SB, I guess I just find the response of “it’s because of all this customization” unlikely, or at best just an indication that the tech people running the website were in over their heads. I mean I’ve looked at the code that I could get access to, the source of the blogs, the customization code people have used, and installed MT on my own system to play around with it, and it’s my conclusion that the tech support was, and yeah this will sound harsh, abysmal. I would argue that this is probably because they had too few resources dedicated to tech support, but I can’t argue this because I don’t know the day to day of how things were run. But because I believe thinking about the future is more important than the past, what I wonder about is whether you can imagine Scienceblogs implementing a new tech feature that goes beyond what others are doing? As a stable platform, SB is fine, but, you know me, I’m all about the future (too infinity and beyond!)
The APS should totally be involved in blogging! My experience, however, with the topical group on quantum’s website doesn’t make me completely optimistic about the tech support however. But I do think this is largely a function of getting the right person. Strangely it seems that technical proficiency in computing varies by many orders of magnitudes!
I do think that discussion is strongly a function of individual ownership of blogs. Group blogs, for instance, don’t seem to generate as much interesting discussion, but perhaps this is just because a group blog, by definition, consists of bloggers who agree with each other. Actually this is a fine idea: I should get someone for whom I agree strongly as a co-blogger so that we will have something to yell and scream about 😉
A lot of the technical headaches are rooted in the way MT was customized, and the fact that they were trying to have both extensively customizable blogs and the ability to fit ads into things. It would be a lot easier to maintain with a less customizable template– a lot of the fixes we’d like to see are rendered inordinately difficult by the requirement that they not break any of the customized gewgaws people have on their blogs.
As far as encouraging cross-blog discussion, I think the key would be getting the bloggers to buy into the idea up front. Once the idea is set, as long as the site doesn’t get too big, new additions will tend to abide by the established local norms. When things get too large, though, you can have sub-communities with lots of nominally cross-blog discussion that doesn’t actually cross disciplinary lines. I do a lot of linking to Rhett at Dot Physics and Matt at Built on Facts, but that’s not the same as having interchanges between people who do very different things.
At points in this process, I have toyed with the idea of seeing in the APS would be interested in acquiring the physics component of SB for Physics Central. They wouldn’t be able to pay us, but they could probably take care of the tech support side of things, and it would potentially be a big boost for the APS outreach program on-line (I am on the Committee on Informing the Public, which oversees the outreach program, so this would be something of a money-where-mouth-is step for me). I doubt they’d be willing to do it without dramatic changes to the character of the blogs, though, and I’m not willing to do that yet. If things go totally pear-shaped, though, I may look into it.
I don’t disagree that the tech support has been pretty bad. The kludgey code that’s used to customize the MT installation is the chief manifestation of the bad tech support. It worked all right when the site was small, but it wasn’t set up in a way that worked well with big expansion.
I’m not sure what sort of new feature they ought to add, other than LaTeX, so I can’t really guess at the likelihood of that happening. I’m pretty low-key when it comes to tech features– as long as the site stays up and the spam gets filtered, I don’t need much more than that.
For the moment, I’m hopeful that the SB problems are sorting themselves out, though this is, of course, subject to change on a moment’s notice. Which is good, because the APS outreach office is at Comic-Con this week, and wouldn’t be available anyway.
They should add a mobile version pronto, as I’ll bet they are losing a lot of traffic with out.
Actually, just prior to this mess, traffic was better than it’s ever been. They might be losing out on some mobile traffic, but it’s not obviously crippling the site.
Raw traffic isn’t the problem. Converting that traffic to ad dollars is the problem, and that’s not as easy as it used to be.
– Great to find you in your new digs. (Not that I’m tremendously active as a commenter, but us Pynchon fans gotta stick together.:)
– I think the APS idea is very good. Another thought: I hear Ira Flatow has invited former SciBloggers to the Science Friday site. That or even NOVA’s site might be promising places for producers and consumers of (more or less) popular science writing to gather. I don’t know what potential problems there would be with controversial non-science content and the public funding of these sites, though some might consider a more exclusive focus on science to be an improvement.
– I think cross-blog discussions depend on discussion topics as much as anything, and there are some sure-fire ones out there, e.g., “What’s the best book you know of to interest non-[physicists/chemists/biologists, etc.] in your area of expertise (or other area of science that fascinates you?”
– If you do manage to link up with a network, would you please tell Scott Aaronson to get his shtetl-optimized behind onto it, then force him to post more rather than actually, y’know, helping students?