It’s been three years in the making, but after some gentle prompting and prodding from a few friends (thank you @nhoizey!)—it has finally happened.
We’ve finally open sourced the #Eleventy Twitter Archive project named `tweetback`:
You may have seen this in action on my site https://zachleat.com/twitter/ or the newly shipped Eleventy Archive https://twitter.11ty.dev
You can find the source code for the project here:
https://github.com/tweetback/tweetback
Try it out!
45 replies
@zachleat You’re awesome!
@eduardoboucas I LEARNED IT FROM YOU!!!
@jaapstronks some people got their archive after 8 days, patience is required. 😅
@zachleat@nhoizey @jaapstronks yeah the @eleven_ty one took over a week
-
@zachleat Finally! Thank you for yet another open source project. ❤️😊
@autiomaa feels like 100 years since I first said I would open source this thing 😅
@zachleat It's me, explaining your projects to myself again! You'll never escape!
This creates a linkable archive of your Twitter presence, the idea of which is to canonize THOSE links as the "correct" links for your tweets going forward. So if you have a lot of content in a blog, like Zach does, that links to tweets, this replaces Twitter as a dependency for that data.
Is that right?
@mwdelaney Yeah! And even better, you can use `@tweetback/canonical` to change any twitter URLs on your site to point to the archives we know about and route around twitter altogether: https://fediverse.zachleat.com/@zachleat/109382945371213697
@zachleat @nhoizey @shanselman cool! sad that twitter archive export stopped working weeks ago
@martinjuhasz people are still getting their archives these days, but it can take a week.
@nhoizey @zachleat @shanselman good to know. crossing the two weeks mark soon 🤞
@martinjuhasz wow, that’s a lot indeed!
@zachleat Dang, I could of used this. I recently built something with Eleventy to archive my girlfriend’s tweets and images. Mostly the images, because there are hundreds. But, it’s not nearly as robust as this.
@nhoizey @zachleat This was my favorite thing to use this week. Thank you! 🤩 Was really impressed with how well it worked (across 25K tweets).
My only feedback was uncertainty navigating beyond Recent/Popular/tweet URL seemed a gap. Perhaps by design, but otherwise lookups by tweet URL worked flawlessly 💜
@addyosmani great!
There’s an issue already open for monthly archives navigation: https://github.com/tweetback/tweetback/issues/15
@nhoizey @zachleat OCI formerly known as Docker ;-)
OCI (Open Container Initiative) is the standard for defining container images that you can run with Docker or other runtimes (containerd, podman,...)
The benefit here would be, if you have stored your public tweets in such a container, you then could just use something like
docker run -p 8080:8080 rhuss/tweets
And then point your browser to http://localhost:8080 Or run it in a Kubernetes cluster. Or whatever :-)
@ro14nd oh, I didn’t know about this, thanks.
You can already run `npm start` if you want to browse you #tweetback archive without sharing it. Isn’t that enough?
@nhoizey @zachleat true, but then you need Node installed and fetch all dependencies. And git. Things might break in the future (left-pad anyone? :-).
OCI containers are fully self contained and contain all those dependencies within the image.You can run it anytime in the future anywhere without rebuilding or checking dependencies. Not so bad for an archive ;-)
Let me do a quick prototype to make it more clear what I mean. Will send you a PR.
@nhoizey @zachleat Here's a first prototype --> https://github.com/tweetback/tweetback/pull/26
You can try it out with my tweets by running `docker run -p 8080:8080 rhuss/tweets`
Thanks a lot for creating and open sourcing this package, this is very helpful for us immigrants ;-)
-
-