How I supercharged Telegram Group with RSS feeds and collaboration for bookmarks (Raindrop)

Telegram introduced a nifty feature called “Topics“. These are forum style interactions, which dynamically switch its user interface.

First, look at this:

I reached out to someone and offered to manage his group using chat management bots. However, he had constraints on time to answer specific queries. We then brainstormed and decided to convert the group to forum style navigation. As you can see in the attached image below, we converted the group into topics. Each topic has an individual link (more on that later).

The immediate upside? Owner can respond to specific queries based on topics. The entire chat is streamlined. I had earlier proposed every query have specific hashtags, which is easier to bunch (or even have the bot auto-reply in busy groups). However, that was impractical, because users were unaware of Telegram’s capabilities. The bots can be configured to respond to specific hashtags (or queries) based on notes. That requires extensive writing and effort (in effect, creating a FAQ). For simple queries, I could possibly involve ChatGPT in future.

Nevertheless, we were also faced with another quandary-links were not allowed to be posted. The reason? They end up derailing the discussions. However, I soon noticed that users were sending direct messages to each other; based on mutual permission, of course. What if we allow links, but have a workaround without derailing the discussion flow?

I was searching for a solution to automate another private group with information from other websites. IFTTT is my preferred bot for connecting various services, but that requires manual intervention on those websites. I needed a solution that would “scrape” the information and then deliver content automatically. We found that solution in AximoBot. I won’t get into details at present, but suffice to state that Aximo can deliver to Telegram topics. This saves time, as I can consume content (on scale), in specific channels, and without any algorithm determining my watch time (e.g. Twitter/YouTube). I follow specific Twitter accounts (in a Twitter list) and read them on Inoreader or fetch high value accounts in Telegram through bots, which helps me reflect on their content. (As an aside, my Telegram account is categorised in folders) which allows me even more fine tuned control over how I consume information by granular control over notifications.

This was a mini-eureka moment.

I created a new folder on RainDrop, which allows collaborative bookmarking. Collaborative folders require email invites. Each collaborative “public” folder on Raindrop has its unique RSS feed that I could connect to Aximobot. Aximobot further delivers to each topic in its configuration!

Aximobot fetches resources from numerous sources (e.g. Reddit, Coub, YouTube etc) and through RSS feeds. The delivery of each source can be tuned in the bot settings through a simple syntax.

What are the potential applications?

• For healthcare, you can have patient support groups. Dedicated volunteers and admins can watch/respond to topics.
• General purpose groups can be divided into topics; e.g. you can have an image repository topic by using the bot to deliver content automatically and keep discussions switched on.
• Collaborating with bookmarks helps preserve the archive. The best part of Raindrop is that only one individual requires “pro subscription” and that the folder is visible to all participants depending on their access. Some users can be set to view bookmarks alone. A group participation ensures best content surfaces appropriate to discussion, and granular permissions helps preserve privacy.
• You are only limited by your imagination on segregating topics!
• Here’s for the oncology professionals: Create a group; for e.g. Lung Cancer. It can be divided on “Pathology Discussion” (for pathologists), Surgery for surgical oncologists, Radiation for Radiation Oncologists. I can set up highly topical RSS feeds from PubMed to deliver content straight into topics. Let’s say, for example, a new DNA array for lung cancer detection has been published. It can be delivered straight to pathologists. A hypofractionation using SBRT has been published, which can be piped to Radiation Oncologists. You can have a common mutual discussion in any forum you wish to pursue. Anyone topic can have specific AMA’s from respective experts and close the other forums for maximal attendance. It is like a conference with multiple ongoing sessions and leaving the open hall for discussions.

I am only placing this on record because I could finally implement it on scale. There are other nifty tricks you can do with automation! They represent endless possibilities! (What if you get plain language summaries for every abstract? It is doable and has been demonstrated! That helps filter out relevance quickly).