Task Continuity

April 10th, 2015

At the Mozilla blog there's a fascinating post about research Gemma Petrie & Bill Selman did into how real users handle Task Continuity, i.e. the mechanics of dealing with stuff they want to read online now as opposed to stuff they want to get back to later:

Based on our research, we developed a general model of what the task continuity process looks like for our participants. Task continuity is a behavior cycle with three distinct stages: Discover, Hold/Push, and Recover.

The Discover stage of the task continuity cycle includes tasks or content in an evaluative state. At this stage, the user decides whether or not to (actively or passively) do something with the content.

The Hold/Push stage of the cycle describes the task continuity-enabling action taken by the user. In this stage, users may:

  • Passively hold tasks/content (e.g. By leaving a tab open)
  • Actively hold tasks/content (e.g. By emailing it to themselves)
  • Push tasks/content to others by sharing it (e.g. Posting it on Facebook)


In the Recover stage of the task continuity cycle, the user is reminded of the task/content (e.g. By seeing an open tab) or recalls the task/content (e.g. Through contextual cues). Relying on memory was one of the most common recovery methods we observed. In order to fully recover the task/content, the user may need to perform additional actions like following a link or reconstructing an activity path.

It seems that emailing yourself a link to look at later is still a thing, which I find both amazing and a bit depressing. Also, there's a distinct lack of mentions of just bookmarking the content in the web browser you're using to read the content in the first place, which rather suggests that browser makers have been wasting their time with that whole bookmarking feature they've all been using for the last couple of decades.

For what it's worth, my usual approach to seeing stuff in my web browser that I'll want to come back to depends upon whether I want it as reference material or just don't have time to look at it right now: material I know I'll definitely want to refer to, especially if I'm likely to want to look at it more than once1 goes to Evernote, where I can tag it and set a reminder alarm if need be. Content that I'll want to look at later (possibly with a view to filing it in Evernote if it's worth keeping) gets bookmarked in Pinboard. Text-heavy content that I know I'll want to read but that is lengthy enough to require a significant chunk of my time is likely to be consigned to Instapaper. Whichever way it goes, one click and I'm ready to move on to the next browser tab.2

Interestingly (Well, I find it interesting!)) the only real exception to that general approach comes when I see a YouTube or Vimeo video I want to come back to: if I'm already on the YouTube or Vimeo site I'll generally use those sites' Watch Later features to tag that content, so that next time I'm in the mood to kill some time watching a bunch of videos it'll be to hand.3

[Via Extenuating Circumstances]

  1. e.g. a snippet of Applescript, or a web page showing the opening hours of a local supermarket over a bank holiday weekend
  2. Mac OS X and iOS try hard to encourage users to use Safari's reading list feature to grab content to review later, but that's no good to me if I might want to look at the items when I'm using a non-Apple device, i.e. if I'm browsing from work during my lunch break using Firefox on a Windows PC.
  3. I still tend to do my video-watching in batches rather than visiting YouTube/Vimeo multiple times during a browsing session. The good old days of reading the internet in batches via a metered dial-up connection scarred me for life shaped my browsing habits in ways I still can't shake.

Comments Off on Task Continuity


March 11th, 2015

js;dr = JavaScript required; Didn't Read.

Pages that are empty without JS: dead to history (archive-org), unreliable for search results (despite any search engine claims of JS support, check it yourself), and thus ignorable. No need to waste time reading or responding.

Also known as, if it's not curlable, it's not on the web. […]

Well said!

[Via tommorris.org]

Comments Off on js;dr

Coming to you via Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9) AppleWebKit/537.71 (KHTML, like Gecko) Version/7.0 Safari/537.71

November 9th, 2013

This history of the browser user-agent string evokes times past, when life on the World Wide Web was simpler, yet user-agent strings got more and more complicated.

The pity of it is that my favourite web browser1 ever never got popular enough for anyone else to want to pretend to be it.

[Via The Tao Of Mac]

  1. I do wish Omni Group had kept actively developing OmniWeb. You can still download it, but they stopped doing anything with it beyond maintenance releases years ago. Even so, if they'd ever fixed some weird shortcomings in the program's Applescript support I'd probably still be using it as my main web browser to this day.

Comments Off on Coming to you via Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9) AppleWebKit/537.71 (KHTML, like Gecko) Version/7.0 Safari/537.71

Subcompact Publishing

November 27th, 2012

Craig Mod is excited by the possibilities of Subcompact Publishing:

In 1967 Honda unveiled the N360.

The N360 was a kei, or light style car; a subcompact.

I like to imagine the engineers at Honda huddled together, dumping the sum total of all car design and production technology on our worn, wooden table. Around they gathered and together they asked, "What's the simplest thing we can build with this?"


The N360 was something an American car company would never dream of producing. You can't blame them though: they had no incentive by which to dream such dreams. Unlike the American automotive industry, the Japanese automotive industry wasn't beholden to industry momentum or legacy. And when you're not beholden to legacy, you can be excessively brazen.

In the software industry we talk about MVPs, or Minimum Viable Products. The N360 was a Minimum Viable Car.

The N360 didn't make it to the States, but the followup – and near equally cute – N600 did. Next came the Honda Civic, then soon after, the oil crisis. We all know how the story goes from there.


Honda was a nobody in the car industry. But they gained foothold and marketshare by building a car that was more appropriate for many consumers. They had built a subcompact.

So I ask: where are our digital publishing subcompacts?

Mod spends a fair bit of time extolling the virtues of Marco Arment's The Magazine, which I wrote about back when it launched. I've maintained my subscription through the first four issues, but I have to admit that I'm wavering over whether to retain it. The application's virtues remain – it's a beautifully polished application, even if I'd like more control over the presentation of the content that it permits,1 but the content isn't that interesting to me.

In principle, an article extolling the virtues of a wet shave, or the proper way to make a cup of tea could be engaging and fun to read, even to a hirsute guy like me who would quench his thirst with a Diet Coke rather than brew a cup of tea every time; in practice I haven't found them to be so. I'm finding that on average there's one article per issue that I find moderately engaging. It doesn't help that some of the writers, whose work I've read on their own weblogs, are covering very familiar ground. Marco did say early on that he hoped to expand the pool of writers after the first few issues, so I'll probably give it another couple of issues to see if things improve.

Having said all that (and to get back to the ostensible topic of this post), there's no doubt in my mind that the basic model of Subcompact Publishing could well develop in all sorts of interesting ways, freeing up writers to write instead of having to code an application and submit it to someone's app store. It's just a shame that whatever tools people come up with will most likely end up being tied to a specific operating system/hardware type/payment mechanism.

Isn't this a problem the web was supposed to have solved by now?

[Via Marco.org]

  1. In particular, I like pagination in my reading apps, dammit! Marco has explained in one of his podcasts that flexible, high-quality pagination is really complicated to do well so for now he's going with a scrolling view.

Comments Off on Subcompact Publishing


November 3rd, 2012

Ever since Apple introduced the Reader feature to Safari, I've been forced to engage in the same ritual after every update to Safari. The thing is, Reader does quite a good job of rendering a cluttered web page readable, but it insists on doing it using justified text, which looks hideous. The (not very user-friendly) way to fix this was to find the Reader.html file buried inside the Safari application package and add a simple text-align: left; to the CSS embedded in that file and save it. Problem solved, except that after each Safari update you'd almost certainly have to repeat the trick. Better still, in some updates Apple changed the location of the damned file so you'd have to figure out where it lived now before you could apply the fix.

After the update to Safari 6 I found the latest home of the Reader.html file and applied my customary edit, but for some reason Safari ignored the revised CSS and kept on rendering justified text in Reader. In searching for hints as to why this might be happening, I came across a much better answer: CustomReader:

With CustomReader, you can change pretty much any aspect of Safari Reader's appearance. CustomReader's settings panel has a graphical user interface that lets you edit a few basic settings, like body font and background color, with a few clicks. But the true power of CustomReader lies in the Advanced tab, where you can directly edit the custom stylesheet that the extension inserts into Safari Reader. By editing this stylesheet, you can override any of Safari Reader's built-in styles with one of your own.

CustomReader has another feature that may be of interest to some. If you find yourself frequently invoking Safari Reader on a certain kind of page at a specific site – for instance, articles on the New York Times website – you can have CustomReader automatically enter the reader whenever you open that kind of page.

It works!1 And with any luck it'll keep working after the next Safari update.

  1. I do have one small quibble. That 'invoke Reader automatically if you visit a specific site' option requires you to enter an escaped version of the site's address: not www.independent.co.uk/, but //www\.independent\.co\.uk/.+. I understand why it's doing that – using a regexp allows for more flexibility in choosing which subsections of a site should trigger Reader – but surely there could be a simple 'trigger-for-this-entire-domain' option that would do the job for 95% of prospective users. But then, probably 98% of Safari users either don't care about text justification badly enough to see this as a problem that needs resolving, or else wouldn't want to touch the CSS for Reader anyway.

Comments Off on CustomReader


June 20th, 2012

I love the fact that Tim Bray is so keen that his proposal for a new HTTP status code for cases where access to a resource is blocked for legal reasons incorporates a Latin example that is both grammatical and historically accurate:

One of the things in the proposal is that the 451 Unavailable for Legal Reasons status is sup­posed to be ac­companied by an explanation of what the legal restrictions are, and what class of sites they apply to. The proposal has an ex­ample, and since obviously you don't want to use any real legal author­i­ties in this situation, I decided to pick on the Roman Empire:

This re­quest may not be serviced in the Roman Province of Judea due to Lex3515, the Legem Ne Subversionem Act of AUC755, which dis­al­lows access to re­sources hosted on servers deemed to be operated by the Judean Liberation Front.


But I made up the name of the Roman law by typing some­thing into Google Trans­late. So… does any­one read­ing this know what a plausible Latin name would be for such a law, and how it would be cited? Roman his­tory is full of law­suits, so I assume it must have been a fairly routine operation. Thanks in ad­vance.

Attention to detail1 being very much the mark of the Alpha Geek.

  1. And yes, he's already acknowledged that he muffed the Monty Python joke, and will be inserting a reference to the People's Front of Judea in a future revision.

Comments Off on 451

The Future of Publishing v5.0 beta 1

May 7th, 2012

Technology Review publisher Jason Pontin learned the hard way that Apps weren't the future of publishing after all:

[…] Tablets and smart phones seemed to promise a return to simpler days. Digital replicas of print newspapers and magazines (which could be read inside Web browsers or proprietary software like Adobe PDF readers) had never been popular with readers; but publishers reasoned that replicas were unpleasant to read on desktop computers and laptops.

The forms of tablets and smart phones were a little like a magazine or newspaper. Couldn't publishers delight readers by delivering something similar to existing digital replicas, suitably enhanced with interactive features, which would run in applications on tablets and smart phones? […]

Here's hoping that the magazine's solution – transitioning to an enhanced web site built to accommodate all sorts of screen sizes, complete with an RSS feed to let users keep track of all the content they publish – ends up netting them enough income to keep publishing.

[Via Scripting News]

Comments Off on The Future of Publishing v5.0 beta 1

Haters gonna Hate, Hat-Tippers gonna Tip their Hats

March 21st, 2012

Following on from the flurry of comment on the Curator's Code the other week1, the Code's creator Maria Popova has responded. In a manner of speaking.

Unfortunately, Popova has responded not so much by addressing the points people have made – be they about why the term 'curation' is inappropriate or about how unsuitable obscure Unicode symbols are as substitutes for the phrases 'Hat Tip' or 'Via' at the end of a post – but by spending three quarters of her post quoting paragraph-long passages from the essays of Albert Einstein on 'the ties of sympathy', 'public opinion', 'our interconnectedness, interdependency, and shared existence', 'good and evil, creative bravery, and human value', and 'life's highest ideals' before alluding to the way that some commenters have responded to her suggestion with 'venom and mean-spirited derision' before pivoting away from the substance of the arguments being made about her project and talking instead about how unacceptable cruelty is and how disappointed she is in many of those who have criticised her suggestion.

To be clear: Maria Popova is perfectly entitled to be offended and upset at criticism she feels to be other than 'constructive' and to call out the community accordingly.2 For what it's worth, I don't think that most of the commentary she linked to (or that I've seen for myself elsewhere) was particularly aggressive or derogatory or bullying. Sceptical as to the benefits of her suggested approach? For sure. Put off by what they saw as the misapplication of the term 'curation'? Absolutely. But with one exception3 they weren't particularly personal or bullying, let alone 'sinister'. But I also recognise that I've almost certainly seen only a small portion of the total response, and in any case it's not my call to make; if Popova felt attacked then of course it's for her to respond as she sees fit. I'm just finding it really hard to square the discussion that I saw going on in various corners of the web with the vicious debate Popova is describing.

It's a shame that she devotes so much space in her post to inspirational quotations and so little to addressing the arguments people made in response to her suggestion, given that she's making a post on the same site where she announced the launch of the Code? Why accuse critics of factual inaccuracies but not address them right there?

To be fair, Popova does mention and link to one site where there's some discussion of the pros and cons of her idea, but it's mostly commentary from third parties and the comments from her that they cite only addresses the issues to the extent that she argues that the Curator's Code site (which, remember, offers bookmarklets for download, all set for users to install so that bloggers content curators can easily insert appropriately-formatted links including her chosen Unicode symbols to their posts) wasn't really about the Unicode symbols or even about her site, it was about 'the bigger point' of why 'curation' matters. If you make specific proposals with accompanying blocks of Javascript code, I think it's incumbent upon you to address issues people raise in detail, not just lament the incivility of those who raise questions about your proposal and airily refer to notions about how now the details aren't important.

[Via swissmiss. Given the context, I can't believe that I forgot to add a 'Via' block to my first draft of this post!]

  1. See my post on the subject here, and a trio of posts at Pop Loser including links to some of the commentary elsewhere here, here and here.
  2. And, to clarify still further, I'm not writing this because I feel that she's directly, or even implicitly, criticising what I wrote about her proposal. First because I'm approximately 99.753% certain that she'll never have noticed what I wrote, and second because I don't think I was in any way venomous or mean-spirited in my post. If you think otherwise, please tell me so.
  3. A tweet linking to an extremely juvenile animated .GIF. Which is at best an impolite but snarky comment on the amount of intellectual masturbation going on over this topic – to which I plead guilty to adding my portion right here in this post! – and is not anyone's idea of a civil contribution to the debate. But it's also atypical of the level of commentary out there.

1 Comment »

The Twungle

March 18th, 2012

Margaret Atwood has posted another1 tribute to the Twungle:

[On Twitter…] you find yourself doing all sorts of things you wouldn't otherwise do. And once you've entered the Enchanted E-Forest, lured in there by cute bunnies and playful kittens, you can find yourself wandering around in it for quite some time. You might even find yourself climbing the odd tree – the very odd tree – or taking refuge in the odd hollow log – the very odd hollow log – because cute bunnies and playful kittens are not the only things alive in the mirkwoods of the Web. Or the webs of the mirkwoods. Paths can get tangled there. Plots can get thickened. Games are afoot.

  1. Previously.

Comments Off on The Twungle

[Via Pop Loser, H/T The Curator's Code]

March 11th, 2012

I've read Introducing The Curator's Code: A Standard for Honoring Attribution of Discovery Across the Web three times now and I still don't see what it accomplishes.

Part of the problem, I'll admit, is that I just don't like the use of the word "curation" to describe the process of writing a linklog. I've been posting links and adding a 'Via' link showing where I got the main link from for twelve years now,1 but I'd never dream of suggesting that I've been 'curating' a collection. It seems to me that the term 'curation' implies a distinctly strategic approach, putting together collections of objects that are somehow related to one another, or which comment on one another in some respect. The only conscious strategy I've applied is that of posting links to the things I've found interesting/amusing online. I don't doubt that looking at my 'body of work' will reveal some clues as to what sort of subject matter I'm interested in, and probably also some shifts over time in what I post about, but I'm under no illusions that my interests are different to those of dozens hundreds thousands millions of ageing English-speaking geeks out there. I do this not because I'm trying to build up a coherent collection, but because sharing links is, at some level, what the World Wide Web was made for. I don't want to just be reading the web without giving something back, and both writing a weblog and attributing my sources are part of that.

Anyway, setting aside my doubts about the use of the term 'curation', I don't see what the use of a special Unicode symbol to mark a 'Via' or a 'Hat Tip' link adds to the web. Those Unicode symbols presumably have other uses, so you can't rely on them a semantic indicators: they're just a text decoration that will mean absolutely nothing to anyone unfamiliar with the concept of the Curator's Code. It'd be immensely helpful if there was a <via> HTML tag that denoted the source of an item and could be used both to style the text on a web page and to allow web tools to latch onto 'Via' links and make some use of them, but really all this is it's a side issue.

The important principle is the question of a weblog author's willingness to attribute the source of a post. Most people who write linklogs (or post to Tumblr, or maintain publicly accessible lists of links at Pinboard or Delicious or wherever) decided a long time ago whether they wanted to go to the trouble of attributing the source of the items they found. I suspect that their decision had very little to do with whether there was a universally recognised Unicode character to use to tag their 'Via' links.

[Via Pop Loser]

  1. Whatever you do, don't be surprised if some of the internal links on those pages don't work. I really should have put thebeard.org to some other use once I migrated the weblog to soreeyes.org years ago now, but somehow I've repeatedly got to the stage of installing a new CMS but then not knowing what I want to use it to publish.

Comments Off on [Via Pop Loser, H/T The Curator's Code]

Page 1 of 1012345...10...Last »