Partly Solipsist Conversational Blogging Over Time

This is blog is one of the places where I have, over the years, chatted to not just to myself, but also to others.

Microblogging tools, as they used to be called — things like Twitter — are now often little more than careless public resharing sites for links and images and “memes” (me-me-me-me, look what I can do in taking this object other people have repurposed and I have repurposed too).

Blog posts, too, are linky things, or can be. Generally, most of my blog posts include links becuase they situate the post in a wider context of “other stuff”. The stuff that prompted me to write the post, the stuff that I’ve written previously that relates to or informs my current understanding, other stuff around the web that I have gone out and discovered to see what other folk have written about the thing I’m writing about, stuff that riffs further on something I don’t have space or time or inclination to cover in any more depth than a casual tease of a reference, or that I’d like to follow up later, stuff that future references posts I haven’t written yet, and so on.

So, this post was inspired by a conversation I heard between Two Old Folks in the Park Waxing On about something or other, and also picks up on something I didn’t mention explicitly in something I posted yesterday, but touched on in passing: social bookmarking.

Social bookmarking is personal link sharing: saving a link to something you think you might want to refer to later in a collection that you deliberately add to, rather than traipse through your browser search history.

At this point, I was going to link to something I thought I’d read on Simon Willison’s blog about using datasette to explore browser search history sqlite files, but I can’t seem to find the thing I perhaps falsely remember? There is a post on Simon’s blog about trawling through Apple photos sqlite db though.

That said, I will drop in this aside about how to find and interrogate your Safari browser history as a SQLite database, and how to find the myriad number of Google Chrome data stashes, which I should probably poke through in a later post. And in aut0-researching another bit of this post, I also note that I have actually posted on searching Google history sqlite files using datasette: Practical DigiSchol – Refinding Lost Recent Web History.

The tool I use for social bookmarking now is pinboard (I note in passing from @simonw’s dogsheep personal analytics page that there is a third-party dogsheep aligned pinboard-to-sqlite exporter, but as one of the old folk muttering, Alan Levine, mentioned, I also have fond memories of de.licio.us (are the .’s in the right place?!), the first social bookmarking tool I got into a daily habit of using.

One of the handy things about those early web apps was that they had easy to use public APIs that didn’t require any tokens or authentication: you just created a URL and pulled back JSON data. This meant it was relatively easy to roll your own applications, often as simple single page web apps. One tool I created off the back of de.licio.us was deliSearch, that let you retrieve a list of tagged links from delicsious for a particular tag, then do an OR’d search over the associated pages, or the domains they were on, using Yahoo search. (For several years now, if you try running more than a couple of hand crafted, advanced searches using logical constructs and/or advanced search limits using Google web search, you get a captcha challenge under the assumption that you’re a robot.)

deliSearch led to several other search experiments, generalised as searchfeedr. This would let you roll a list of links from numerous sources, such as social bookmark feeds, all the links in a set of course materials, or even just the links on a web page such as this one, and roll your own search engine over them. See foer example, this presentation from ILI 2007, back in those happy days when I used to get speaking gigs: Search Hubs and Custom Search Engines (ILI2007).

Something else I picked up from social bookmarking sites was the power of collections recast as graphs over people and/or tags. Here’s an early example of a crude view over folk using the edupunk tag on delicious circa 2008 (eduPunk Chatter):

And here was another experiment from several years later (Dominant Tags in My Delicious Network), looking at the dominant tags used across folk in my delicious network defined as the folk I followed on delicious at the time:

Another experiment used a delicious feature that let you retrieve the 100 most recent bookmarks saved with a particular tag and data on on who bookmarked it an what other tags they used, to give us a snapshot at a very particular point in time around a particular tag (Visualising Delicious Tag Communities Using Gephi). Here’s the activity around the ds106 tag one particular day in 2011:

And finally, pinboard also has a JSON API that I think replicated many of the calls that could be made to the delicious API. I thought I’d done some recipes updating some of my old delicious hacks to use the pinboard API bit offhand, I can’t seem to find them anywhere. Should’a bookmarked them, I guess!

PS in getting on for decades of using WordPress, I don’t think I have never mistakenly published a page rather than a post. With the new WordPress-com UI, perhaps showing me some new features I donlt want, I made that mistake for the first time I can think of today. WordPress-com hosted blog new authoring experience sucks, and if it’s easy to make publishing mistakes like that, I imagine other f**d-up ness can follow along too.