Twitter dropped RSS feeds in early 2013 (Mashable) and never had direct support for RSS feeds from favorites. I still use Twitter favorites as a bookmark or “read it later” service for myself and have been running the Tweetfave service for over 2 years now. Tying these ideas together, I’ve just rolled out Twitter favorites RSS feed support in Tweetfave.
How it Works
Tweetfave periodically scans your account (using the official API) and summarizes all favorited tweets in an email. Now the system will also update a feed which can be used with other systems that accept standard RSS feeds.
The RSS feed will be created with each tweet contained in an item element with the following fields set:
title – text version of the tweet
description – text version of the tweet, plus a link to the original tweet
content:encoded – HTML version of the tweet, plus a link to the original tweet
link – the first link mentioned in the tweet (see Notes below)
guid – same as link
pubDate – date/time of the original tweet
Here’s a screenshot of my favorites feed shown in Firefox:
Sign in to Twitter (if needed) and approve the Tweetfave App
Enter your email and click Submit
Mark a couple of tweets as favorite to get started
Within a couple hours you should get your first email from Tweetfave
Now return using the “existing users” steps above to grab your RSS feed link
What to Do
So what can be done now that you have an RSS feed of your Twitter favorites? I like to use IFTTT which is an awesome application for connecting different services together. (For a great overview, see their About IFTTT page).
I’ve set up my recipes where the source trigger is a new item in the RSS feed and the target is a service like Instapaper, Paper, or Evernote. You can configure the fields and formatting sent to various services, giving you lots of different options.
Here are a few notes and caveats to keep in mind:
If you’re a new Tweetfave user, you need to favorite at least one tweet first, and wait for the first email before your feed link is created
Your RSS feed link is somewhat obfuscated so that people can’t guess it. It’s also only shown to you after login, so you can still keep your favorites stream private.
If a favorited tweet has multiple links, you will have multiple RSS items (because each one can only have one link). This seems like a good compromise even at the expense of some extra entries.
The time between favoriting a tweet and it appearing in your feed is not instant; Tweetfave currently scans everything at 2-hour intervals, then sends emails and updates the feeds.
If you disable your Tweetfave account, the RSS feed file will still be online but won’t be updated. If you’d like the file removed, just email me.
I’d love to hear feedback from anyone trying this out, including any interesting use cases people come up with!
People created some pretty crazy mashups with Pipes. For me the simple ability to combine RSS feeds with some simple logic was very useful (for example: 2014 Tech Advent Calendars). I also recently created some Stack Overflow feed tools for work which will have to be recreated in Python or something similar.
I worked at Yahoo for almost 7 years and the phrase “Product Prioritization” became kind of a good-news/bad-news thing towards the end. It always sounds good (it’s logical to focus resources on the most important products and sites), but if the closure affected the site that you work, that’s not as much fun. (In the end that was the final result for me and much of my team on the Yahoo Developer Network.)
Early in March this site was hit with a WordPress hack that was present for about 10 days until I discovered and fixed it. After the fix I was frequently checking Google Webmaster Tools to make sure search results returned. After about one week, my traffic (from Google at least) had pretty much recovered to the pre-hack levels:
And here is the corresponding timeline from Google Analytics, showing traffic slowly returning to normal:
Last night I had the unpleasant discovery that this site’s WordPress had been “hacked”, with every post redirecting to an uploaded “this site has been hacked” variety of HTML file. I looked back and realized it happened on March 1st and mad at myself for not noticing sooner.
Fortunately it was pretty easy to clean up by zapping the database and restoring from a good backup (thankfully I have daily backups running).
The harder part is going to be recovering in Google’s view. Search queries as shown in Google Webmaster Tools dropped like crazy right away:
And here’s the corresponding crawl errors view:
Hopefully after a little time the Google crawler will see all those pages returned, but I’m guessing whatever page rank I had will be very slow to recover (if it ever does). In the meantime I’ve improved my WordPress security a bit more, updated to the latest of everything, and removed a few unused plugins. Next will be to set something up to notify me more quickly if this happens again.