447 karma · joined April 20, 2010
I used an algorithm that splits along text into tweets from another project that ensures that the sentences never slice and the thread is easy to read.
I've built it using Preact, CSS, Firebase (Firestore/Functions).
Chirr does not update the content but removes the edited part of the thread and republishes it, so it's impossible to abuse.
Another thing I working on is delayed publishing ala Gmail.
- First of all, threads are often long-living i.e., authors update it over time. It allows them to surface the content every time they update it.
- Secondly, often every tweet's a self-containing idea. Blog posts usually are worried and need more time to grasp.
- Also, it's not only easy to consume but also to share.
I like, write and consume both Twitter threads and blog posts.
I'm also working on adding a delayed publishing like in GMail, so you can edit typos even before the thread got published.
- https://twitter.com/dannypostmaa/status/1282952961857970176
- https://twitter.com/robhope/status/1265278107088347136
Of course, you can make a blog post from such threads (and probably should, if the thread got traction), but it would look and read differently.
I released the initial version of the app in 2017 as a free and open-source tool (https://github.com/kossnocorp/chirrapp) even before Twitter threads become a thing.
Later that year, Twitter gave me a lot of trouble when they added 280-chars support but didn't update API and the twitter-text library. I was about to close it, but thanks to Firebase, the running cost was $0, so I fixed the issues and kept it alive.
With time, the app accumulated authority and users. Since 2017 when I first launched the app, it has been used by 9.5K people to publish 22K threads and 330K tweets! Happy users started coming to me with thanks and features requests more and more so I decided to rewrite the app and added tons of new features, like:
- Drag-n-drop images
- Editing published threads (yes, the edit button!)
- Drafts
- Scheduling
- And much more!
In case if anyone curious, the app is built with TypeScript, Preact, on top Firebase platform.
My goal wasn't to create a 100% secure diary. If you need that, you probably should not store your data in the cloud. That's unlikely that I will ever try to make it happen. But even if I would, there're so many ways to screw it over anyway, so I won't ever try to make this promise.
However, I will consider open-sourcing it.
Also see my comment where I addressed the privacy issue: https://news.ycombinator.com/item?id=22045670
> But once the data leaves the browser there is no way to know, wouldn't you consider to partner with Gmail(or others) and appear as an addon to an already trusted company in order to start off the business ?
Also, I understand the concern and that Paul most likely will not trust their secrets to anyone. The problem is that's not a business, but a beautiful hobby project that I honestly love, so it's unlikely that I will ever spend time rewriting it and then paying Google $15K (https://www.gmass.co/blog/google-oauth-verification-security...) so they could vet me.
> But I can probably set that up as a cron job
Almost every app could be replaced with email, a cron job or a spreadsheet ;-)
Here're some facts that could help you to find peace of mind. First of all, I operate in the EU, so I can't simply sell your data. Also, it costs me virtually $0 to maintain the service (thanks to Firebase and Mailgun), so I won't be forced to sell out to keep it afloat. At last but not least I use it myself with my close friends so we're in the same boat.