Like everyone else, I have feelings about the rumoured algorithm coming to change our Twitter timelines. But maybe unlike everyone else, I have long been working on secret plan to keep my Twitter timeline intact.

Ever since Twitter started making life difficult for third-party clients and developers to create (or even make use of existing) features, I’ve been stocking the bunker with the supplies. I’ve even had to start from scratch when Twitter (officially) removed some key features and changed their API. But as of this moment, stores are replenished. And if I had to take shelter in the bunker tomorrow, I believe I could survive.

Here’s my plan for surviving an (the?) eventual Twitter apocolypse…


This wasn’t such a bold idea back when this was an officially supported feature of Twitter, but my idea was to use an RSS reader to read Twitter. At the time, I wanted two things from twitter:

  • To be able to read every tweet, no matter how long it had been since I last checked Twitter
  • To be able to read tweets offline

A feed reader was the best tool I could think of to accomplish this. Feed readers could handle large volumes of unread items, and at least some of them could downlad items and function offline. Feed readers also allowed you to create folders of grouped RSS feeds, thereby seeing unread items from several feeds grouped together in chronological order. This would, in essence, produce a “Twitter timeline” with individual feeds (ie users) interspersed.

And so, my task became to figure out how to get streams of tweets as RSS feeds.


It wasn’t completely obvious how to find an RSS feed of a particular user’s tweets, but neither was it difficult. Although we typically think of identifying a Twitter user by their Twitter ID (e.g. sarah11918 or jbrains), each Twitter user has a unique identifying number. With a little sleuthing, it was possible to figure out the number of any user, and from that number you could create an RSS feed of that user’s tweets.


Somewhere along the line, Twitter changed their API and officially removed support for creating RSS feeds based on an individual user’s tweets. So, my old RSS feeds were all broken.


Here’s where creating RSS feeds got all backdoor-y. People started writing executable scripts in Google Doc files that you could copy to your own drive, and use to run custom twitter “apps” you created by entering your app’s unique API key. This meant…


At, you can sign in with your Twitter account then scroll to the tiny font at the bottom where you can select Manage My Apps. Here, you can register a new twitter app, which basically just means that now you have a thing that can access Twitter and you have some codes that give you the permission to log in and do so. Now, I had the information I needed to plug into the Google script I had downloaded which would convert a user’s tweets into an RSS feed.

People wrote other Google scripts that could also archive these tweets into a Google spreadsheet in your Drive account. I played around with those just to see how they worked, but, um, have you SEEN how much Joe and I tweet? Those files started to get large and I decided I didn’t really need them.

Anyway, this was all well and good until Twitter changed AGAIN and removed the ability to pull data based on the syntax of that feed, which incidentally was now username based, not a Twitter number.


While Twitter removed the ability to create RSS feeds (and then, intentionally or not, removed the ability to create sneaky RSS workarounds), they added in a user’s settings the ability to create “widgets” to embed on a website. Embedded widgets weren’t really a new thing. Established blogging platforms had had them for a while. But making it so that just anyone could easily create code for their own widget was new.

Widgets are mini-windows you can put on a web page that display one particular part of twitter: a single user’s tweets (or favourites/likes), all the tweets from a Twitter List (a list of users you want grouped together), or Twitter Search Results for a particular keyword. And guess what? Each individual widget you create has a unique identification number…


After months of broken RSS feeds, I stumbled upon Amit Agarwal’s post at where he details how to use that unique widget number to, once again, create an RSS feed from an executable Google script. The nice thing about this solution is that you no longer actually need to create and register a Twitter App to have access to Twitter’s data. The widget collects all the data and the Google Script converts it to an RSS feed that can be read by any feed reader.

Now, all you need to do is log into Twitter, go into Settings, create a widget (to display tweets for an individual user, an entire list, or a search term) and copy that number into the Google Script to generate an RSS feed. You can repeat the same step multiple times with multiple widget ID numbers… one for each user you follow, if you really wanted to! Or, you can create a list of users and then create a single widget for this Twitter list.


But, the point of Twitter is to see tweets in chronological order, so conversations don’t get confusing, so you can see events unfolding, so you can distinguish new from old news… Here’s where configuring your feed reader comes in.

Most feed readers allow you to create folders of feeds, adding any number of feeds you want into a particular folder. While you can view any feed individually, you can also simply choose to view “the folder.” This will show you, in chronological order, every post from all the feeds in that folder. This somewhat replicates your Twitter timeline in that you see the tweets of people you follow in the order they were tweeted, not separated out by user. This is my strategy for coming close to replicating a chronological Twitter timeline, no matter what Twitter eventually decides to do with their timeline.


Obviously, this isn’t actually a “Twitter client.” You can’t post tweets and you can’t reply; this is simply for reading/tracking tweets of people, lists and keyword searches you have chosen to follow.

Similarly, it will take some extra clicks to get the full context of tweets. There is no “conversation view” (although, Quote Tweeting has somewhat broken that on Twitter’s end, itself), and depending on your feed reader, you may see Quoted Tweets, pictures and other media as simply a link rather than as an embedded item.

However, feed readers are capable of handling links (and you can often specify whether you’d prefer to use an internal or external browser), so viewing the original tweet, or the links/media with it, as well as any context/conversation/replies, is just a click away.

For me, this system doesn’t have to handle posting of tweets. I will have one Twitter app or another installed on my phone and laptop, and I can always visit the website simply to click the big Tweet button to post. I don’t actually need to read anything on the website itself.

And, one can always hope that third-party clients will remain immune to Twitter’s whims. Currently, I don’t see ANY promoted tweets or add on any of the clients I use, so those tweets simply aren’t making it to me outside of “official Twitter.”

This is really nothing more than the Survivalist Guide To Viewing Your Twitter Timeline, not a full-fledged Twitter solution. I currently only have a few RSS feeds created myself, not a full 700 or so to represent everyone I follow. I more just wanted to set this up as a proof of concept, and to feel confident that I had a backup plan.

And until Twitter changes things YET AGAIN, I do.