An Outline for Using Hypothesis for Owning your Annotations and Highlights

I was taken with Ian O’Byrne’s righteous excitement in his video the other day over the realization that he could potentially own his online annotations using Hypothesis, that I thought I’d take a moment to outline a few methods I’ve used.

There are certainly variations of ways for attempting to own one’s own annotations using Hypothesis and syndicating them to one’s website (via a PESOS workflow), but I thought I’d outline the quickest version I’m aware of that requires little to no programming or code, but also allows some relatively pretty results. While some of the portions below are WordPress specific, there’s certainly no reason they couldn’t be implemented for other systems.

Saving individual annotations one at a time

Here’s an easy method for taking each individual annotation you create on Hypothesis and quickly porting it to your site:

Create an IFTTT.com recipe to port your Hypothesis RSS feed into WordPress posts. Generally chose an “If RSS, then WordPress” setup and use the following data to build the recipe:

  • Input feed: https://hypothes.is/stream.atom?user=username (change username to your user name)
  • Optional title: 📑 {{EntryTitle}}
  • Body: {{EntryContent}} from {{EntryUrl}} <br />{{EntryPublished}}
  • Categories: Highlight (use whatever categories you prefer, but be aware they’ll apply to all your future posts from this feed)
  • Tags: hypothes.is
  • Post status (optional): I set mine to “Draft” so I have the option to keep it privately or to publish it publicly at a later date.

Modify any of the above fields as necessary for your needs. IFTTT.com usually polls your feed every 10-15 minutes. You can usually pretty quickly take this data and turn it into your post kind of preference–suggestions include read, bookmark, like, favorite, or even reply. Add additional categories, tags, or other metadata as necessary for easier searching at a later time.

Here’s an example of one on my website that uses this method. I’ve obviously created a custom highlight post kind of my own for more specific presentation as well as microformats markup.

A highlight from Hypothesis posted on my own website using some customized code to create a “Highlight post” using the Post Kinds Plugin.

Aggregating lots of annotations on a single page

If you do a lot of annotations on Hypothesis and prefer to create a bookmark or read post that aggregates all of your annotations on a given post, the quickest way I’ve seen on WordPress to export your data is to use the Hypothesis Aggregator plugin [GitHub].

  • Create a tag “key” for a particular article by creating an acronym from the article title followed by the date and then the author’s initials. This will allow you to quickly conglomerate all the annotations for a particular article or web page. As an example for this article I’d use: OUHOAH062218CA. In addition to any other necessary tags, I’ll tag each of my annotations on the particular article with this somewhat random, yet specific key for which there are unlikely to be any other similar tags in my account.
  • Create a bookmark, read, reply or other post kind to which you’ll attach your annotations. I often use a bookmarklet for speed here.
  • Use the Hypothesis Aggregator’s short code for your tag and username to pull your annotations for the particular tag. It will look like this:
    [hypothesis user = 'username' tags = 'tagname']

    If you’re clever, you could include this shortcode in the body of your IFTTT recipe (if you’re using drafts) and simply change the tag name to the appropriate one to save half a step or need to remember the shortcode format each time.

If you’re worried that Hypothes.is may eventually shut down, the plugin quits working (leaving you with ugly short codes in your post) or all of the above, you can add the following steps as a quick work-around.

  • Input the shortcode as above, click on the “Preview” button in WordPress’s Publish meta box which will open a new window and let you view your post.
  • Copy the preview of the annotations you’d like to keep in your post and paste them over your shortcode in the Visual editor tab on your draft post. (This will maintain the simple HTML formatting tags, which you can also edit or supplement if you like.)
  • I also strip out the additional unnecessary data from Hypothesis Aggregator about the article it’s from as well as the line about who created the annotation which isn’t necessary as my post will implicitly have that data. Depending on how you make your post (i.e. not using the Post Kinds Plugin), you may want to keep it.

As Greg McVerry kindly points out, Jon Udell has created a simple web-tool for inputting a few bits of data about a set of annotations to export them variously in HTML, CSV, or JSON format. If you’re not a developer and don’t want to fuss with Hypothesis’ API, this is also a reasonably solid method of quickly exporting subsections of your annotations and cutting and pasting them onto your website. It does export a lot more data that one might want for their site and could require some additional clean up, particularly in HTML format.

Perhaps with some elbow grease and coding skill, sometime in the future, we’ll have a simple way to implement a POSSE workflow that will allow you to post your annotations to your own website and syndicate them to services like Hypothesis. In the erstwhile, hopefully this will help close a little of the data gap for those using their websites as their commonplace books or digital notebooks.

Syndicated copies to:

Reply to chenoehart’s tweet about community

Replied to a tweet by Chenoe HartChenoe Hart (Twitter)
Whenever I find myself actively seeking something to RT it always feels like there’s nothing but noise. Seem to find the most interesting things to share after I’ve already found too many things to RT at once.

Sometimes I find myself wanting to tweet just in general, and wish there was an easy way to just have casual conversations on here, tweet about the weather or something. It’s often really just a proxy for trying to meet people anyway.

I’ve had this feeling before and often long for the earlier days of Twitter when it functioned more like this. The popularization of Twitter in 2009 and the subsequent iteration on the platform and its community killed all the original spirit. It also reminds me of a piece I’d read recently by John Naughton1 about how toxic the retweet functionality (and other gamification like likes/favorites) can be.

I’ve seen the type of interaction you’re describing in smaller pockets of the internet on services like App.net (aka ADN, now defunct), pnut, and 10centuries, and a few corners of the Mastodon sphere.

The place I’ve seen it done well most recently is on Manton Reece‘s awesome micro.blog service, which I think has some strong community spirit and a greater chance of longevity. They’ve specifically left off “features” like follower counts, number of likes, and made conversation front and center. As a result it is a much more solid and welcoming community. I’m curious, as always, if they can maintain it as they scale, but the fact that they encourage people to have their own website and own their own data mean that you can take it all with you somewhere else if they ever cease meeting your needs in the future–something that certainly can’t be easily done on Twitter.

I hope you find the connections with the types of people you’d like to meet.

Originally bookmarked on April 01, 2018 at 09:22PM

References

1.
Naughton J. How to stay sane on Twitter: ignore retweets. Memex 1.1: John Naughton’s online diary. http://memex.naughtons.org/archives/2018/03/11/25409. Published March 11, 2018. Accessed April 12, 2018.
Syndicated copies to:

👓 Librarian tweetstorm by @green_grainger

Read Librarian tweetstorm by Georgia | Saoirse (Twitter)
So there was a MYSTERY at the library today.

A wee old women came in and said "I've a question. Why does page 7 in all the books I take out have the 7 underlined in pen? It seems odd."
"What?" I say, thinking she might be a bit off her rocker. She showed me, and they did.

I asked if she was doing it, she said she wasnt and showed me the new book she was getting out that she hadnt even had yet. It also had the 7 underlined! "I don't know, maybe someone really likes page 7?" I said, assuming of course that there is a serial killer in the library.

I checked some other books. Most didn't have it, but a lot in this genre did - they're "wee old women" books (romances set in wartime Britain etc). Lots of underlined 7s. The woman who pointed it out shrugged and went on her way, "just thought you should know".

My manager came back from doing arts and crafts with some of the kids and I decide to tell her about the serial killer in the library.
And that’s how I found out that a lot of our elderly clientele have secret codes to mark which books they’ve read before.

Our computers do it automatically but many have been doing it since before that was possible, so Esther might underline page 7, while Anne might draw a little star on the last page, and Fred might put an “f” on the title page. Then when they pick it up, they can check!

It’s quite clever really but now I’m dying to just underline page 7 of every new wee old women book we get in.

So, good news: there’s not a serial killer in the library whose MO include the number 7 and wartime romances. Bad news: people are defacing books rather than just asking us to scan them (smiling face with smiling eyes)

I'm now concerned that the amount of people enjoying this thread means there's going to be a new spate of readers using secret codes - apologies to librarians everywhere!
(although, in truth, I find it hard to be annoyed about it - better than torn pages and felt pen graffiti!)

(Also, I am new to the library job, hence why I hadn't seen it before! The library and our customers are great though (smiling face with smiling eyes))

Just had another victim of the page 7 vandal returned!!!
(Now checking every book that looks like it might be their taste...)

This is such an interesting little story including some cultural anthropology.

Syndicated copies to:

🎵 “Natural Blues” by Moby feat. Vera Hall

Listened to Natural Blues by Moby feat. Vera Hall from Play (on Mute, V2)
"Natural Blues" is a song by American electronica musician Moby, released as the fifth single from his 1999 studio album Play. It samples "Trouble So Hard" by American folk singer Vera Hall. It was first released in the United Kingdom, where it peaked at number 11. The song received remixes by Paul Oakenfold, Mike D, Peace Division, Katcha and the Olmec Heads.

🎵 “Somebody That I Used To Know” by Gotye, Kimbra

Listened to Somebody That I Used to Know by Gotye, Kimbra from Making Mirrors (Eleven)
A song written by Belgian-Australian singer-songwriter Gotye, featuring New Zealand singer Kimbra. The song was released in Australia and New Zealand by Eleven Music on 5 July 2011 as the second single from Gotye's third studio album, Making Mirrors (2011). It was later released by Universal Music in December 2011 in the United Kingdom, and in January 2012 in the United States and Ireland. "Somebody That I Used To Know" was written and recorded by Gotye at his parents' house on the Mornington Peninsula in Victoria and is lyrically related to the experiences he has had with relationships.

🎵 “I Need You” by America

Listened to I Need You by America from Warner Bros.
"I Need You," released in 1972, is the second single by the band America from their eponymous debut album America. The song was written by Gerry Beckley. It appears on the live albums Live (1977), In Concert (1985), In Concert (King Biscuit), Horse With No Name - Live! (1995), and The Grand Cayman Concert (2002). The studio version is included on the compilation albums Highway (2000) and The Complete Greatest Hits (2001). George Martin remixed the studio recording for inclusion on History: America's Greatest Hits (1975). An alternate mix from 1971 appears on the 2015 release Archives, Vol. 1.

Owning my Online Reading Status Updates

As of October 30, 2016, I’ve slowly but surely begun posting what I’m actively reading online to my blog.

I’ve refined the process a bit in the last couple of weeks, and am becoming relatively happy with the overall output. For those interested, below is the general process/workflow I’m using:

  1. As I read a website, I use a browser extension (there’s also a bookmarklet available) linked to my Reading.am account to indicate that I’m currently reading a particular article.
  2. I have an IFTTT.com applet that scrapes the RSS feed of my Reading account for new entries (in near real-time) and this creates a new WordPress draft post on my blog. I did have to change my IFTTT.com settings not to use their custom URL shortener to make things easier and to prevent future potential link-rot.
  3. Shortly after I’m done reading, I receive a notification of the creation of the draft post to remind me to (optionally) post my comments/thoughts to the draft post. If necessary, I make any additional modifications or add tags to the post.
  4. I publish the post; and
  5. Optionally, I send POSSE copies to other silos like Facebook, Twitter, or Google+ to engage with other parts of my network.

Status updates of this type also have a pre-included O-embed with a synopsis of the content if the bookmarked site supports it, otherwise, a blockquoted synopsis stripped from the site’s meta-data is included.

Other near-term improvements may include custom coding something via the available Reading.am hooks to directly integrate with the WordPress Post Kinds plugin to use the URL post pattern http://www.yoursite.com/wp-admin/post-new.php?kind=read&kindurl=@url to shorten the workflow even further. Post Kinds automatically handles the wrapping of the post data in the appropriate microformats automatically. I also want to add a tidbit so that when I make my post I ping the Internet archive with the URL of the article I read so that it will be archived for future potential reference (hat tip to Jeremy Kieth for giving me the idea at IndieWebCamp LA a few weeks ago.)

I had originally played around with using the Post Kinds bookmarklet method directly, but this got in the way of the immediacy of reading the particular article for me. Using a PESOS method allows me to read and process the article a bit first before writing commentary or other details. I may also integrate a Hypothes.is based workflow into this process in which I use the hypothes.is browser etension to highlight and annotate the article and then use the Hypothes.is Aggregator Plugin to embed those thoughts into the post via shortcodes. The following post serves as a rough example of this, though the CSS for it could stand a bit of work: Chris Aldrich is reading WordPress Without Shame.

I was a bit surprised that Reading.am didn’t already natively support a WordPress pathway though it has a custom set up for Tumblr as well as a half a dozen other silos. Perhaps they’ll support WordPress in the future?

These new read post types can be found at the following URL: http://boffosocko.com/kind/read/?type=status?type=link.

Syndicated copies to:

Web-based Push Notifications with Pushpad

Push Notifications

A push notification (AKA client notification) is a notification that shows up on one or more of your client devices without you having to explicitly request it — it’s “pushed” to you, instead of you having to poll for it. –Source: IndieWeb.org

Pushpad

Today I came across a beta web service called Pushpad that provides easy-to-install push notifications. As a result, for people who spend a lot of time in front of their screens, they can now subscribe to updates on the site here via web browser push notifications. Subscribers will get a small toaster-like pop up notification in real time on their screen to indicate that new content was published.

My first push notification
My first push notification

 

Set up

The service was quick and simple to set up with lots of documentation. While geared at large corporations looking for a simple turnkey implementation for push notifications on most major web browsers, it’s also easily usable by smaller sites. Even better it’s free for providing less than 10,000 notifications a month, which covers most small sites.

They provide an “Express” version that requires no serious technical skills and sets up in just a few minutes and a separate “Pro” version which provides a lot of additional customization (including a white labeled version) for those with the development skills to implement it.

For those on WordPress, they also have an easy to use plugin.

Pushpad supports the Push API for Chrome and Firefox and APNs for Safari.

Automation

Pushpad also supports integration with Zapier (currently in beta), which means that any of the hundreds of applications that are integrated with Zapier can be used to create push notifications on the desktop. Hopefully they include IFTTT.com soon too. I’m already using Pushbullet with IFTTT for integration between my Android phone and my desktop, but additional integrations for personalized notifications could be cool.

Roll Your Own

But maybe you’re hard core? If you prefer not relying on outside services, you can always build your own push notifications! In particular, IndieWeb.org provides some thoughts and tips about how to implement these for yourself based on open web standards.

Push Notifications for BoffoSocko.com

Now that we’ve been talking about them, would you like to try receiving them in the future?  You can subscribe to push notifications for my blog by simply clicking on the icon below and then authenticating your subscription:

Not into push notifications? Maybe this isn’t your favorite way to find out about my content? If not, I offer a number of other ways to subscribe and consume my content.

Syndicated copies to: