 In advance of deleting my Twitter account, I made this web page that lets you search my tweets, link to an archived version, and read whole threads I wrote. https://tinysubversions.com/twitter-archive/ I will eventually release this as a website I host where you drop your Twitter zip archive in and it spits out the 100% static site you see here. Then you can just upload it somewhere and you have an archive that is also easy to style how you like it. 200 comments
 @mcc well its a static site, it works great on localhost and possible even file:// though I haven't checked @darius this is great! I saw something similar on github recently: @rahulbhadani How do you do that? Typically it won't grab your entire archive, just individual pages you tell it to save. @darius You can create an account on archive.org, and then go to account and before archiving, click on “save outlinks”. Without account sign in only a single page is archived. But of course web.archive.org won’t save your private messages but at least it will give you the latest snapshot of the public account to stay on its server as long as the server exists. @rahulbhadani I don't think that would work to back up a person's whole public twitter... if you can link me to a working example I would appreciate it! @darius You can check @potus latest twitter snapshot to get the idea. http://web.archive.org/web/20221216050417/https://twitter.com/potus It even has a chrome/brave/chromium extension. @darius I am not sure on how you have approached but it has been working great for me for many years in archiving snapshots of any webpage. You can try to play with it and should be able to figure out.  @darius this is absolutely amazing. I was looking at Tweetback, which also looks good, but that requires npm/eleventy, which I'm less familiar with. Looking forward to when/If you release this. @darius curious, does it display a full list of tweets, like an archive, or do you have to search for something in particular? @csilverman it doesn't. is there a good way to display hundreds of thousands of things for browsing? I feel like search is the best way to access this stuff but I'm open to suggestions @darius The best approach I can think of would be something like a blog, where there's X number of tweets on the frontpage and pagination below that (personally I've never been a fan of the infinite scrolling approach). Search is useful mainly for me, if I'm trying to find something I remember sharing, but my use case here is a preserved archive of posts that others can browse—again, like a microblog. @darius I stopped tweeting in 2017, grabbed an archive, then deleted everything. I have such regret that my archive is in an old format, and I can’t use your (or other recent) tools to make proper archives including original media and non t.co links @darius is Twitter still functioning well enough to get a new file of our tweets out? I had heard that was broken?  @darius my twitter archive zip is 3GB. You might better do it as a tool.  @darius OMG this is so needed. Thank you. If you want a beta tester, I'm here to help. 
[DATA EXPUNGED]
@darius you might want to keep an eye on this tool ☝️- it may prove very useful for #histodons for archiving your twitter work. @AnarchoDoggo yeah there are a fair number of these but none of them were like... a big zip file of HTML files that you unzip on a web server. they're all like, framework this, github pages that, database blah blah @darius thanks! Sorry one other question, how did you go about dealing with shortened twitter urls for QTs, etc? Especially in case where original link is dead (because of account deletion / suspension), was there a way to utilize internet archive or something else to resurrect the content there? @AnarchoDoggo QTs will look like this: https://tinysubversions.com/twitter-archive/tinysubversions/status/789584846335074306 I leave it up to the user to either go to that link, or to see if wayback machine has it, etc. Otherwise I would be archiving other people's content and that is an ethical can of worms I do not want to open with this project since I cannot get people's consent for that @AnarchoDoggo oh it's built in to the twitter archive itself. the text of the tweets contains the t.co but there is metadata for original URLs so I just find/replace @darius makes a lot of sense, thanks. What is used to translate short links to originals? @darius lol too quick! Thanks again for your responsiveness and work here; very useful model @darius If anyone wants to self-host their tweets, this generates a fairly simple static site https://github.com/donohoe/twitter-archive @donohoe Oh nice. Can you point me to an example of what the output looks like? (In my case I was trying to avoid people having to interact with a scripting language and do it all in the browser.)    Thank you! I'd like to fully deactivate my account next year and this would come in very handy. @darius tried to do something like that but the archive turned to be something like 1.5gb, so not that feasible...  @darius cc: Brewster Kahle @brewsterkahle just tooted that the Internet Archive supports something similar: https://mastodon.archive.org/@brewsterkahle/109530105190586499 @arjen @brewsterkahle immediate problem I see with that implementation: 'tweets.js' only contains a partial record of tweets. for some accounts there can be more files called things like 'tweets-part1.js' in addition to 'tweets.js' I applaud the effort though and look forward to seeing the final version!!
[DATA EXPUNGED]
I am getting various responses to this like, "Looking forward to your twitter archive creator! What happens with my twitter data when I upload it to your service? How do you expect me to upload a multi-gigabyte zip file?" You don't know me very well. There is no upload at all. There's no service, no server. The data stays on your computer. Everything happens in the browser client :AngelDevil: We are all so trained to assume that "web application" means "a server does the work"!  @darius Maybe you can explain on the page your data isn't actually uploaded. The people at https://loudnesspenalty.com also do this for example :) @Tijn I have this notice on the "make it yourself" tool that is not yet released. People are just making the assumptions from the text in my tweet about how I'm going to make a service @jonas I have! It's great but it requires you to run node.js, and do a bunch of other dev-type stuff. I wanted something that is 100% in a browser and just involves you uploading a file and unzipping it, which is possible even if you just have cPanel access @darius What if we want to post our Twitter archive as a searchable database online? @steventdennis the thing I am making is something you can upload to any web host. But I'm not going to set up a service to host other people's data @darius This made me think, because it's a thing that's currently impossible to prove to users, but what if browsers had an irreversible `navigator.disconnect()`, plus a UI indicator? @kosinus that would be neat! right now the only way a user has to prove it is to basically go into airplane mode before running the code and close the browser tab before reconnecting to the internet  That IS super useful Darius!!!! I'll be sure to check this out! 👍👍👍👍 @darius I know its kinda not your goal, but not everyone is able to self-host. I would certainly be interested in hosting a way to allow users to upload their archive and be able to link to it later. @HMHackMaster this may be relevant for people who just want a backup somewhere: https://help.archive.org/help/how-to-archive-your-tweets-with-the-wayback-machine/  Would you care to write/link a short howto? I'd like to do something similar, but I don't have the skills to do this 100% independently. I did save my twitter-posts, though. @iju I am going to release a web app where you can do this yourself, as I say in the post you are replying to Ah, I think I misread the original post (for my defence: it's 2 in the morning here). Thank-you for both your work and kindness! @darius I think this is the tweet where I realised I no longer have an effin’clue about how the internet works, about storage and archiving, hosting, etc. I’m now at the age where the internet is just a thing that’s there. Something I consume and enjoy but I never contribute to or change. Basically, I’ve become “a grandad with his first ‘so-called smart phone (so what’s so bloody smart about it, can it cook pizzas?)’”. An observer. @darius nice, but what I value most are the conversations. It's a truth of social media that no matter how much various partners value a combined work, just one can just delete sizeable portions that leant it meaning. @darius I don’t know how I forgot to post yesterday, but I’d love to test this out if you’re still looking. I was trying to figure out how best to upload my backup. @darius — I wish I knew how to do this. Too late for me— but sharing for others! 👏🏻 My tool for making a simple, searchable, themeable archive of your public tweets and threads is now live: https://tinysubversions.com/twitter-archive/make-your-own/ The tool runs entirely on your computer, in your browser. None of your data is uploaded anywhere in the process. The output is a zip file of a basic HTML website that you can upload to a web host if you choose. The site also answers most of the common questions I get. Please read it first before asking me questions here!  @darius Never mind the folks saying, gosh, I wanted you to do a different thing (reminds me so much of academic reviewers who critique by saying you didn’t write the article THEY would’ve written). I am delighted to be able to turn my twitter archive into something I can actually use! Thank you so much! @darius |
Gregory's Server
I tried to make sure that the link previews for pages look nice over here in fediverse-land. Hopefully this renders well??
https://tinysubversions.com/twitter-archive/tinysubversions/status/470670223901986816