Page 1 of 2
 It has been just over a week since I started using the new /updates view for /reader rather than the more typical river.

It's been great! It make a refreshing change to visit each person's site again and read their stuff in its home environment.

I'm still informed when someone has written something new (the automatic refresh is 6 hours) but it's not right there in front of me to scroll through (consumption by default.) Instead, I have to make the conscious decision to click out to another site.

Site by site, rather than post by post. It gives me a better feel for things.

I've been leaving a few more comments than I previously would have done, and I seem to take more time with each post when not consuming it in the river view. I also, obviously, get to see the comments left by other people.

The visual variety, rather than river homogeneity, keeps things interesting making me want to visit and read posts in situ. I highly recommend this approach if you're bored with your current consumption habits.


The 'Now' namespace for RSS

 Just before bed last night I had a flood of inspiration following on from the /updates page and Bix's idea of followable /now pages.

With RSS being my preferred method of distribution I thought about how this could be achieved.

Introducing the 'Now' namespace for RSS

After initially reading Robin Sloan's proposal for the Spring '83 protocol 1 I suggested that RSS feeds could have "a 'block' that shows under certain circumstances". This, combined with the new 'updates' view for /reader, gave me the idea of creating a namespace (itself inspired by the /now page) to specify the elements required for such a block.

The namespace is documented on the site here or on GitHub (feel free to comment by creating an issue). The (channel level) elements defined are:

  • now:title – (optional) a title for the update

  • now:link – (optional) - the link to a /now page or other source of updates

  • now:content – (required) the update content itself, plain text or CDATA block

  • now:markdown – (optional) markdown version of the update content

  • now:timestamp – (optional) the time of the update

The source of update for the 'block' can be anything, not just a /now page; I am using the contents of /now as a proof of concept. When the RSS feed is built it pulls in the contents of the now page from the database.

As an example, the information in my feed is consumed in /updates after having modified the feed parser library to support the namespace elements. It looks as below:

Updates with Now text

Comments? Suggestions? Is this a viable means of making 'now-like' updates shareable and followable?

  1. where updates would be shared as small 'boards' and a resulting collection of boards would look like a magazine rack or the small ads from a newspaper 

 An initial version of the 'feed updates' page is live.

(See this post for the background.)

It's nice and simple, just listing those sites with unread updates, how many and since when. A 'mark as read' check mark is shown next to each when logged in.

Feed updates

This is, obviously, only intended for me as it reflects my unread status in /reader. It's an experiment to see how if affects my reading behaviour.

An extra option has been added to /admin to choose the default view of /reader so that I can switch between this and the standard 'items' view. This doesn't affect the view when not logged in.

 Back in May I started looking at making feed fetching in /reader more efficient and explored the options for running PHP processes in parallel.

Unfortunately, the common methods (async, pthreads, parallel) are not supported by my hosting.

The library I'm using to consume feeds uses cURL (if installed, file_get_contents() if not) to grab the feed contents but doing this repeatedly for multiple feeds takes time — with the 29 feeds I had it was taking around 18 seconds to do a full refresh.

I recently came across 'cURL multi' 1 which allows the processing of multiple cURL handles asynchronously so wondered if I could make that work here. On Wednesday evening, as is my way, I started overcomplicating matters thinking I was going to have to rewrite the library. Tired and confused I went to bed wondering how I was going to achieve this.

Yesterday, however, I realised that, rather than need to rewrite things, I could just make a couple of the private functions public and call the library one stage further in. Instead of the library making single cURL requests, I could do the multi requests first and pass the results to the (now public) later functions.

The upshot is that a full refresh now takes less than three seconds for 31 feeds — a sixth of the time it was previously taking! As I add more feeds the time savings will only get larger.

  1. I'm still learning my way around PHP. I haven't got to classes yet, using them confuses my old brain. 

 I keep going back to the thought that "the most focus you'll ever have" on something is when it first grabs you.

I've had a number of items left as unread in Reader, a number of notes left unfinished, a number of tabs left open, all just waiting for me to return to them and write something.

Two years ago, Drew Coffman wrote that notes can be a graveyard for your thoughts, written down then lost. I remarked that it felt my blog could be like this: "that I'm not doing enough with what I've got, making do instead of making strides."

If that is the case, if it is a graveyard, then I want to speak to the dead, build on their ideas. (And by 'the dead' I mean past instances of me who thought something was worth linking to or writing about.) That's why I've made all the recent changes to the site. Reader, Start, it's all about extending blogging, getting more from it, hopefully doing more with it by creating pathways through my thoughts over time.

Tom Critchlow wrote about increasing the surface area of blogging; he was talking about blogging as a whole but it can equally apply to a single blog. Creating a larger surface area allows for more scope, more references, more links, more connections – with some work and a bit of luck that leads to more ideas. Critchlow wrote:

"I think there’s something quietly radical about making your feed reader open by default. It increases the surface area of RSS so others can discover content more easily. It makes blogging more visible."

That's true but, in a setup like mine, it also allows for easier consumption and reuse.

The desire to turn the blog into a database, rather than just be held in one, has been long standing, five years at least. My interest was, therefore, piqued when CJ Eller quoted Justin Murphy's post Personal Knowledge Management is Bullshit. Murphy argues that having an ever expanding dataset is "oppressive not impressive. It’s not useful, and it’s not illuminating."

"Individuals blessed with high degrees of industriousness and orderliness will build sophisticated media diets, note-taking systems, and automated archiving pipelines much more effectively than those less blessed with these traits."

Then they try to sell these systems to the rest of us.

I am definitely more of the latter, the 'less blessed' and always have been; organisation has never been a strength. To that end, I've always been a bad curator. Start is an attempt at improving things, taking them out of my hands to a degree, with a minimum of effort.

It seems a waste to have years of material, of my personal history, sat there for its own sake rather than be able to take advantage of the potential held within. But Eller likens PKM to a garbage heap, "a perpetually expanding web of hyperlinked notes" asking how we can prevent one from developing and save ourselves from such a fate.

Maybe the answer is simply to be selective.

For various reasons, I have been very bad at feeding the machine, very few posts have been marked with labels that would place them within my system. Maybe that's a good thing. By being slow to add items I am preventing an unwieldy build up. I originally intended to throw all sorts in but find that something really has to resonate before it gets labelled.

I also feel that integration is key. Having tried a wiki-like system in the Garden I decided it was better to have any form of PKM as integral to the blog, directly fed by it rather than as a separate repository. Thoughts and ideas are posted once to the blog, why duplicate the effort? I don't want it to be what most would consider a traditional PKM system. Yes, there are bi-directional links in some cases and, between posts, I have the ability to indicate these links, but there are few. Instead, I mainly rely on internal webmentions to indicate related posts. Start deals more with mini-hierarchies originating from individual starting points, threads rather than a web.

Perhaps the blog is finally ready to be the "digital public commonplace book" or thought space alluded to by Chris Aldrich almost two and a half years ago.

Past, present and future me are in a far better position to communicate across the years than ever without having to shoulder too heavy a burden, or continually sort through (and possibly dispose of) a garbage heap. The blog exists in its own right, and will continue to do so for years to come. Everything needed is held within and I now have a way to unlock it.

What could be easier than that?

Colin Walker Colin Walker