I had been using Feedly on my iPad for some time, before that dies irae. Thus, the announcement of Reader’s execution was not much of a concern to me: Feedly promised that the transition from Google Reader to Feedly (which already used Google oAuth) would be smooth — and so it was. After Reader was finally earthed, reading my feeds with Feedly’s native iPad app continues to be a comfortable experience. But managing my feeds with Feedly’s Chrome extension, meanwhile has proven to be a way better experience than Reader’s outdated UI ever had on offer. Some features (or rather the usabilty of their implementation) remain to be desired still, though. For one: importing a few dozens of feeds at once.
Last week, my (partial) employment as a teacher brought me a very practical use case. I commissioned my students to set up a blog, where I have them report on my classes each week. There are about fifty of them, and hence as much blogs I need to keep track of (for grading :()). Instead of copy/pasting the URLs in Feedly’s “Add Content” input field, one by one, I wanted to give Feedly a list, and get them in there all at once.
Google returns no quick solution. But I dugg this clue on a Uservoice.com forum. In brief:
- Create an OPML file (some all-purpose XML schema) from your list of feed urls.
- Import that OPML file into Feedly (via an unfindable UI element).
Surely, I wouldn’t go through the pain of typing all that XML bloat by hand. Luckily, I also dugg a (browser-based) “OPMLBuilder” — it’s quite a handy tool. There’s a “List of RSS feeds url” input field where we can paste our urls, one per line. Click “Create OPML”, et voilà, you’re directed to the generated (raw) XML.
Now select/copy/paste into your code editor of choice and edit the XML — it’s not required, but hey, when we’re at it, we want to do things right! OPMLBuilder outputted each url in an
<outline> tag, with the literal url repeated thrice: as the value of the
Using Sublime’s awesome multiple selection cursor feature, we can edit these values in one go and remove the clutter in the
text properties. The value of both properties should be identical — one of them will be used by Feedly as the name of the corresponding feed’s source (i.e. the name of the blog or website).
OPMLBuilder put all the
<outline> tags holding our feeds within a common
<outline> parent, which has but one property
text="Main Folder". Now change “Main Folder” into whatever name you want those urls to be grouped under, once we’re in back Feedly.
Save the edited OPML file as
[mylistoffeeds].opml. Next, point your browser to http://cloud.feedly.com/#cortex, and upload your
.opml file. Done.
<?xml version="1.0" encoding="UTF-8"?> <opml version="1.0"> <head> <title>My Awesome Feed Collection</title> <dateCreated> Tue, 01 Oct 2013 18:56:47 +0200 </dateCreated> </head> <body> <outline text="Typography" title="Typography"> <outline type="rss" text="Rhythmus.be" title="Rhythmus.be" xmlUrl="http://www.rhythmus.be/blog.rss" jsonUrl="http://www.rhythmus.be/blog.json" htmlUrl="http://www.rhythmus.be/blog/" /> </outline> </body> </opml>
Say we want to share our entire Feedly collection of awesome content sources. Then we reverse the process, and ask Feedly to spit out an OPML file, which, by now, we have become familiar with. Feedly indeed has an export feature, which you can access over http://cloud.feedly.com/#opml. Here’s mine. Enjoy!
One more thing
Now, I wouldn’t be a good hacker if I wouldn’t got an idea by now for yet another cool
app tool (one could hardly call it an app).
There’s a few cool people who regularly share some awesome content, on Twitter, Hacker News, and what have you. They surely do have an exquisite collection of feeds they’re following. I want those, too!
Not unlikely, they also have a neat collection of bookmarks over on Pinboard. I can have those bookmarks already, but I really want the entire RSS feed of those sources (when applicable).
Now suppose that instead of curating a collection of RSS feeds and bookmarks, I could rather curate my personal meta collection of awesome content curators, who’d do the heavy lifting for me.
The app would allow me to collect Twitter handles, HN usernames, Pinboard users, and so on. I could link someone’s HN username with his Twitter handle. I could group those people (with their respective online IDs) into folders, categories, or tag them.
On the back-end, the app would go over my collection of cool people, go to their respective online outlets, would grab all the urls they post, resolve those into rss feed urls, would remove duplicates, collect the uniques, assign them values for popularity, get tags, and use those for auto classification. That app would also — well, you get the picture.
Let’s see if we can put up something like that. Shall we? Let me know what you think via Twitter.