Pingbacks: hiding in plain sight

Having set thesis drafting aside pending feedback from my supervisors, I’ve returned to my data … and each time I write that phrase ‘my data,’ it bothers me. It’s really not my data at all; I don’t have any particular rights over them, other than, with the help of a bunch of other folk. having assembled them together. Anyway, I’ve returned to my flânografie and am casting my eyes back over the notes I made during the seven months of participant observation. These were the episodes which appeared on Twitter, sometimes in my timeline, sometimes through using search terms on Tweetdeck and often as a result of someone pointing me towards a tweet or post they thought I might find interesting.

“Bloom’s Taxonomy and Twitter” by Aditi Rao, @TeachBytes

In this instance I have Andrea Stringer to thank for pointing me towards the blog post which prompted me to write this. “22 Ways To Use Twitter With Bloom’s Taxonomy” was written by Aditi Rao, @TeachBytes on Twitter. Usually when an item like this came into view, I’d make some notes describing what I saw and adding a few reflective comments. Back in January of 2017 when I read Aditi’s post, I remarked neither on Aditi’s brief introduction to the graphic, nor on its contents. What struck me more was the effect it was having on other people and how they might be learning from it. My attention was therefore drawn to the ways in which other people had interacted with the post and their reactions to it.

As I scrolled down to the comments section of Aditi’s post, I first passed the social media metrics which revealed the post had been shared 179 times on Facebook, twice on Pinterest and, given that I arrived at the post from a tweet, somewhat surprisingly, not at all on Twitter. I’m more than sceptical about figures like this as indicators of the influence of social media. The tweet which led me to Aditi’s post and which contained the graphic had 42 RTs and 43 Likes. If we also take into account that the post had also been Liked by 12 bloggers, the best we can say is a good number of people have seen the graphic and the resharing of it suggest that it has been largely well received. We can’t really say if they have learned much from it or whether it had an effect on their practice, but that’s where the Comments come in.

WordPress blogs often open the comments section with “n thoughts on blog title.’ This particular post had 19 thoughts or comments (as of 15th Jan 2017), so I thought I’d perhaps learn something of how people used or reacted to the graphic. In fact, only two of the comments were actual comments: one which was slightly critical of Bloom’s used in this way, and a reply from Aditi. The other comments were all Pingbacks, which is to say automatically generated links to this post from other blog posts, which WordPress detects and presents.

For example, Alice writes an interesting article on her Web log. Bob then reads this article and comments about it, linking back to Alice’s original post. Using pingback, Bob’s software can automatically notify Alice that her post has been linked to, and Alice’s software can then include this information on her site.

Langridge & Hickson, 2002

If you click on the pingback link, you’re immediately taken back to the blog post which linked to Aditi’s post. When I made my research notes at the time, I checked a couple of the pingbacked posts and briefly remarked on their contents. At the time, I left it there, but as I returned to my notes earlier today, I started to sense that pingbacks might be more significant than I had originally considered.

The Pingback as actor

I’ve never really thought about Pingbacks on blog posts; they just appear. On my own blogs, most of the pingbacks are in fact internal referencing as I link from one post to another. But maybe they’re not as mundane as they might at first appear and in fact they work much harder than I first thought? When someone reads a blog post and is subsequently minded to write their own post, either referencing or extending the ideas in the original, they are extending knowledge. Were it not for the pingback, the link between the two posts would be one way only, from the body of the new post back to the original. The pingback is initiated automatically from within the original post platform and consequently makes this a two-way exchange by providing that link to the new post.

This extending of the knowledge web offers opportunities, but I wonder to what extent people use it? I know that if I write a post which attracts a pingback, I usually follow it up to check out the post and the author. The outcome might be that I learn something new about what I originally thought, or that I find a new blog to follow, or a new person with whom to connect. The interesting part is that it’s an algorithm or script that’s doing that. A nonhuman. My learning is once more being affected and enabled by a nonhuman actor.

With those thoughts, I decided to ‘follow the (pingback) actors’ on TeachByte’s post and see where they led. Here are each of the destinations:

Each of these posts may of course spawn their own pingbacks; Larry Ferlazzo’s post itself had 53 comments, of which 11 were pingbacks. It’s easy to see how a knowledge repository can soon build up, but interesting to speculate on whether someone with Larry’s clout (influence? popularity?) has to be part of that.

Next I followed each of the above pingbacks to their source to see how far information was diffusing, but here I encountered an issue. Many of the pingbacks came from what appeared to be trainee teacher blogs who seemed to be on the same course (EDC3100) together and had perhaps been given an assignment on Bloom’s. Few of their blogs seemed currently maintained, and there were no pingbacks leading out. A couple of the above listed pingbacks gave 404 errors, the blogs presumably having been dismantled.

I then went on from there to Larry’s blog to see where the second level of pingbacks came from. After searching fruitlessly for a tool that would crawl these automatically, I thought I’d attempt to collect them manually, with the intention of producing a visualisation. I chose NodeXL for this, entered the blog names/authors and from where the pingbacks had come

“Pingbacks” flickr photo by IaninSheffield shared under a Creative Commons (BY-NC-SA) license

I’d much have preferred the image to be interactive so you could jump out to the sites ‘live’ so to speak. Nevertheless, at least you get a sense of the extent of the network which is assembled. We can see that after Larry’s post, the pingback trails pretty much ran cold; some were again 404 errors and one was spam. The small circles attached to two of the nodes indicates an internal reference between posts within the blog itself. I was rather surprised though and expected more of a cascade effect. I guess this might be an indicator that teachers’ blogs in general simply don’t attract that much traffic. Mine doesn’t. Or perhaps it was the subject matter? Bloom’s taxonomy has a strong following in some circles, but is heavily criticised in others; a marmite topic. Of course in focusing on the pingback, I’ve neglected the other more textual contributions within the comments sections; these can also be followed back to the source, though that often isn’t a blog post. The other aspect which isn’t particularly visible in this case is what prompted the post from which I chose to start. Aditi claims it was colleagues asking questions about Twitter, but how Bloom’s came to form part of her solution isn’t clear. Perhaps a post she had read had helped prompt her in this direction, but without the backwards link, we can’t know. Any starting point we choose, it could have been Larry’s post, will likely have precursors.

So, even in this single example, we start to get a sense of how the learning builds, grows, spreads and changes. What began as a fusion of Twitter and Bloom’s may continue as someone discusses the same topic in their post, or perhaps they use this in the context of a discussion purely about Bloom’s and its applications. Alternately they may be prompted to think about Twitter in a different way, or even segue into a different but related topic like educational technology implementation. The knowledge assemblage becomes richer and more diverse with each iteration, but is tied together by the pingbacks. I wonder if a Deleuze and Guattarian view might see these as potential ‘lines of flight’ for a learner?

One other aspect which struck me here was the 404s and what they indicated. The visualisation I produced of the pingbacks fails to reveal any temporal features and is more of a snapshot in time. The web of connections and the knowledge assemblage which forms grows with time as I mentioned earlier, but it also decays too. There are plenty of reasons why what was once on the end of the pingback may no longer be there: people shut down blogs or move them to new locations, or blog platforms fail (RIP Posterous). There’s an almost natural cycle of growth and decay, and even regeneration here. The tweet Andrea used which started the ball rolling is captured in my research notes, but is no longer on the Twitter platform. The tweet to which she pointed and which directed me towards Aditi’s post was fresh (as of Jan 2017), but referred back to a 2013 post, thereby breathing fresh life into it and perhaps starting a further round of pingbacks. Growth and decay.

Wait! Aren’t you researching Twitter?

I am indeed and the preceding discussion has largely centred on pingbacks, a feature of blogs, rather than microblogs. I have two points to make here: firstly that microblogs and Twitter may have features which function in a similar way to pingbacks. The retweet for example provides a similar link to a text or resource that someone else has produced. I’ll admit that it has less permanence than a pingback, patiently ensconced at the foot of a blog and ready to whisk the reader off to the linked blog, but then the structure and function of Twitter is one of flow and change when compared with a blog; it’s a different beast. The second is that my point of entry to the blogs and their interconnected web of enabling pingbacks was a tweet. Two actually. Andrea’s tweet took me to another tweet which referenced Aditi’s blog post; had I not been on Twitter and had Andrea and I not made a connection through that platform, the likelihood of me ever being aware of Aditi’s post and the learning opportunities that it and its wider assemblage brings together would be minimal.

A sociomaterial view

I’ve used the word assemblage a couple of times now and I perhaps need to lay out how learning is perceived when one comes to it with a sociomaterial sensibility. Conventionally, learning is often framed as either cognitive ‘acquisition’ arising within the individual (growth in representational knowledge) or social ‘participation (skilled participation in the practices of a social group assumes that learning is an outcome of social interaction and that the practice of participation is an exclusively human one.) Sociomaterial perspectives on the other hand:

…do not privilege human consciousness or intention, but trace how knowledge, knowers and known (representations, subjects and objects) emerge together with/in activity.

Fenwick, Nerland & Jensen (2012)

It’s not the people alone who are immersed in the enactment of learning, it’s the materiality too in the form of the WordPress platform and its algorithms,  It’s a ‘a performative knowledge practice constituted and enacted by people and tools in complex collectives or assemblages’ (Mulcahy, 2014) and as such, what professional learning is depends entirely on how it is performed by those (humans and nonhumans) involved.

I’m still working at conceiving learning as more than a process which is contained solely within human consciousness, but I’m getting there. Each time I write or talk about it, I’m enacting a knowledge assemblage of keyboards and texts, projectors and visualisations, which together constitute my (our?) learning. Getting there.

Fenwick, T., Nerland, M., & Jensen, K. (2012). Sociomaterial approaches to conceptualising professional learning and practice. Journal of Education and Work, 25(1), 1-13.
Langridge, S., & Hickson, I. Pingback 1.0 (2002). URL http://www. hixie. ch/specs/pingback/pingback.
Mulcahy, D. (2014). Re-thinking teacher professional learning: A more than representational account. In T. J. Fenwick, & M. Nerland (Eds.), Reconceptualising professional learning: Sociomaterial knowledges, practices, and responsibilities (pp. 52-66). New York: Routledge.


3 thoughts on “Pingbacks: hiding in plain sight

  1. I’m finding your short study and thoughts on pingbacks while I was thinking about Webmentions (and a particular issue that Aaron Davis was having with them) after having spent a chunk of the day remotely following the Dodging the Memory Hole 2017 conference at the Internet Archive in San Francisco.

    It’s made me realize that one of the bigger values of the iteration that Webmentions has over its predecessor pingbacks and trackbacks is that at least a snapshot of the content has captured on the receiving site. As you’ve noted that while the receiving site has the scant data from the pingback, there’s not much to look at in general and even less when the sending site has disappeared from the web. In the case of Webmentions, even if the sending site has disappeared from the web, the receiving site can still potentially display more of that missing content if it wishes. Within the WordPress ecosystem simple mentions only show the indication that the article was mentioned, but hiding within the actual database on the back end is a copy of the post itself. With a few quick changes to make the “mention” into a “reply” the content of the original post can be quickly uncovered/recovered. (I do wonder a bit if you cross-referenced the Internet Archive or other sources in your search to attempt to recover those lost links.)

    I will admit that I recall the Webmention spec allowing a site to modify and/or update its replies/webmentions, but in practice I’m not sure how many sites actually implement this functionality, so from an archiveal standpoint it’s probably pretty solid/stable at the moment.

    Separately, I also find myself looking at your small example and how you’ve expanded it out a level or two within your network to see how it spread. This reminds me of Ryan Barrrett’s work from earlier this year on the IndieWeb network in creating the Indie Map tool which he used to show the interconnections between over three thousand people (or their websites) using links like Webmentions. Depending on your broader study, it might make an interesting example to look at and/or perhaps some code to extend?

    With particular regard to your paragraph under “Wait! Aren’t you researching Twitter?” I thought I’d point you to a hybrid approach of melding some of Twitter and older/traditional blogs together. I personally post everything to my own website first and syndicate it to Twitter and then backfeed all of the replies, comments, and reactions via using webmentions. While there aren’t a lot of users on the internet doing something like this at the moment, it may provide a very different microcosm for you to take a look at. I’ve even patched together a means to allow people to @mention me on Twitter that sends the data to my personal website as a means of communication.

    After a bit of poking around, I was also glad to find a fellow netizen who is also consciously using their website as a commonplace book of sorts.

    Originally posted at:


  2. Chris,

    Thanks so much for thinking about my post, responding in such detail, and extending my knowledge. I only had a vague notion of webmentions from somewhere, but never really looked into what they are or what they do. Your post has certainly moved me forward in that respect and I can see why you would refer to them as an iterative step forward from pingbacks and trackbacks.

    In the interests of personal interest and further exploration, I’ve followed the instructions on using and enabled them on my blog. It may not go much further than that at the moment as I’m up to my eyeballs in analysis and thesis writing, having moved on from data collection. I’m also lacking in any coding/scripting skills, so the entry point to many Indieweb projects is often somewhat high for me.

    You’re right to wonder whether I used the Internet Archive to resolve the 404s and I must confess, I didn’t think of it at the time. As you say, it might have extended the vis somewhat, but I guess what I was looking for the vis to do, is to provide a time-bound snapshot of the assemblage as it expands and contracts. If sites do disappear, at least on the surface, then that’s a natural part of the learning assemblage. Learning is growth and reconfiguration; sites disappearing might be like forgetting information. If I’ve understood webmentions correctly, they perform better in a learning assemblage than pingbacks, because the information they capture has a greater longevity. They make knowledge more resilient?

    Plenty for me to think about. Thanks Chris.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s