
articles
nowadays, I learned.
tags
contact
Tagged with cultural press, modern survival, stable websites, and other topics.
Posted on December 8, 2025
I grew up alongside sociable internet, as it was changing from dude interest to contemporary tradition. I made friends and companions, pursued a career in social identity, and discovered my gay identification as a result of those events. In the early 2010s, I signed up for Twitter and Tumblr, and pussy pics I did so for more than ten years.
That effect did last a long time. The messages themselves, Not at all.

Social multimedia is delicate, and it can vanish immediately. Websites are sold, disabled, or locked down. Persons log out of the Internet or shut their records. We live in an era of vast engineering and storeroom, but the normal report of our lives is disappearing before our eyes. Websites that don’t care about stability delete, censor, or lose articles.
I want to retain social advertising as a hazy recollection rather than just one. I want to realize simply what I read, what I saw, what I wrote. These vestiges are now electric. If I were born 50 years ago, I’d be the type of person to stay a diary full of words and postcards, natural reminders of the people who mattered to me.
I don’t believe the Internet to realize for me, but I’ve built my own book of social press. It’s a spot where I may archive content that have shaped, overjoyed, or simply stuck in my head.
I may use my own keywords to search through dialogues from various solutions, enjoy them in my internet website, and keep them to a dynamic website. This blog is the first in a three-part collection about preserving sociable advertising, based on both my professional and personal expertise. Although it is less than two years old, it previously seems more in-depth than some social media sites.
The extended route to a long-term library
How does a dynamic website operate When viewed in a computer
Interactions as a store product
A unique data concept and configuration for each webpage
Every chat uses phrase labelling.
Metadata in JSON/JavaScript, interpreted as a curve
A huge collection of testing
The first-party libraries of Twitter’s stable site
Flickr Foundation Data Lifeboat
My net tabs
Iconfactory’s Tapestry is available now.
Social internet links are embedded on this website.
Before I actually heard the phrase ”digital preservation\
To locate and correct identifiers, I use soft wire coordinating.
Here’s a quick rundown of how my documents and data are organized on device:
This information creates a simple diagram:
My comments contain all of my information. jquery, which contains elements similar to the preceding Twitter instance.
Posts does recommend to media files, which I party and retailer in the media/directory using the first notice of their filename, allowing for a tolerable number of files in each subdirectory.
Messages indicate their authors in clients. jquery. My customer unit is small, with a show name and the journey of an game picture in avatars/, if the site allows it.
Customers are presently broken down by webpage, and I didn’t compare users across different websites. For instance, I can’t verify that @alexwlchan on Twitter and @alex@alexwlchan .net on Mastodon are the same person. That’s everything I want to do in the future.
I have a Py and pytest evaluation hotel that inspections the accuracy and consistency of my information. Among these are items like:
– My information concept and information documents correspond.
Every press file described in the data is saved on device, and every advertising file described in the data is described in the information.
– For each article I’ve saved, I have a status graphic for the creator of each.
– Every date uses a standardized style
– My videos are not encoded in AV1 ( which is not supported by my iPhone ).
I’m performing a lot of regular information enhancing, and these tests provide me with a health nets against mistakes. Since they move quite quickly, I use them each occasion I change.
Every social media website has a way to import your information, but some export are superior to another. Ample to cooperate with info import regulations, but nothing more. Some websites clearly only provide it with a zip archive that is jam-packed with JSON documents and little documents or justification.
The Twitter archives was little better. The first thing you would discover when you downloaded your library was a HTML record called Your archives. xhtml. A dynamic site with full-text lookup for your posts did appear if you first did this.
This strategy was a great motivation for me, and it led me to choose to use dynamic sites instead of little files. These files did continue to exist for a long time after Twitter or X have vanished from the web because it’s a amazingly solid element of architecture.
Because it only contains my posts, the Twitter library isn’t specifically what I want. Back-and-forth meetings were one of my favorite on Twitter, and my specific archives solely contains my side of the conversation. I does record both women’s accomplishments in my personalized book.
The Flickr Foundation’s venture, Data Lifeboat, aims to create archive versions of Flickr. I spent nearly two years building the earliest designs of the Data Lifeboat at the Foundation. I joined because I was interested in archiving interpersonal advertising, and the thoughts developed in both ways: my research informed my function, and vice versa.
Although some specifics about Data Lifeboat and my diary differ, the fundamental ideas remain the equal.
One of my favorite things about that project was pushing stable webpages more than I’ve ever been before with little archives. A spectator software for browsing the contents, which is a dynamic site created in a stale Script style, is included with each Data Lifeboat deal, which is very similar to the Twitter archive. The most challenging dynamic website I’ve actually created was so difficult that I had to create a check set using Playwright.
What I learned from that encounter made me more creative about what I can do with dynamic, self-sufficient websites.
I wrote earlier this year about my bookmark selection, which I likewise keep on a dynamic website. My favorite books come primarily from long-form language and video-reference substance with personal information. The diary is normally short-form material, frequently using visible press, and frequently featuring conversations I participated in. Both offer me access to reliable, up-to-date files of items I don’t want to reduce.
I created my own tabs website because I didn’t confidence a bookmarks assistance to final, and I also created my societal press scrapbook because I don’t think social media platforms will continue to exist. They’re two distinct incarnations of the same concept.
An phone software called Tapestry combines content from various channels into a single timetable, including journals, RSS feeds, and social media. The software uses site-specific” connectors” that were created using fundamental website technologies like JavaScript and JSON to pull in articles.
Although I don’t personally utilize Tapestry, I was pleasantly surprised by the pattern, particularly the plugs. I was thinking about distinct information versions for each webpage when I was inspired to think about this concept, and of course I enjoy using old-fashioned website technology.
When I embed social media posts on this site, I don’t use the platform-provided indigenous connects, which consume gigabyte of JavaScript and recording. Rather, I use simple Browser excerpts that I created using my own CSS, which I initially discovered on Dr. Drang’s website more than thirteen centuries before.
These fragments’ sensory demeanor isn’t ideal for the authentic page, but they’re still there in good plenty ways to be used. My book used the CSS and HTML themes as a good starting point.
You you create everything related with a fraction of the work, but I spent a lot of time and effort on this task, and I had fun doing it. There are many simpler ways to save an online page’s internet storage, such as a copy, a word document, or a picture.
Preserve your unique replicate if there is something net that you care about but wouldn’t want to drop. According to the past of the Internet, it will almost surely vanish at some point.
The Internet can forget, but it doesn’t have to bring your recollections with it.
This website was created by Alex Chan ( they/she ) with.
You can use, reveal, and conform what you’ve read as long as you link up to me because it’s licensed under CC BY 4.0. If you enjoy the website, would you kindly contemplate giving something.
No listing found.
Compare listings
Compare