A few words from Alex about the Pastry Box and content preservation

At the Pastry Box, I have introduced a few features to help you share and preserve (and cherish) the content we publish, and consume it in the ways that suit you best.

This document outlines how those features work, and the philosophy behind them.

Decentralizing content consumption

Preserving content is an obsession of mine.

While content on the web can take a myriad of forms, from full-blown applications and their many versions to binary files, I’m going to focus here on the type of content that the Pastry Box publishes: texts.

Applying the “worst-case scenario check” to various forms of content to see how fit for survival they are

When I think about texts published in the form of web pages, I can’t help but immediately make a comparison with texts in the form of books and magazines. When I acquire books and magazines and store them on my shelves, they are mine to take care of. And I take good care of them.

If all the authors of the books I have on my shelves suddenly died and all the publishing houses editing those books went bankrupt and all their warehouses (and stocks) simultaneously burned down and the libraries selling those books all disappeared at once along with all the material used to print those books, digital or otherwise―if all that happened, my copies would still sit quietly on my shelves. That worst-case scenario would have absolutely no impact on them.

Let’s go one step further. Let’s now imagine that a burglar broke into my home and stole all my books. What would happen to the content that I lost? Would it be impossible to access it ever again? Would it remain forever lost to me?


All I would have to do is find some other people who owned the content that I had lost and duplicate it. I could go one step further and create my own publishing house and re-edit those books to make them accessible to the world again. Although this would be no small task, it’s something that could undeniably be achieved.

This is all made possible because the content stored in books and magazines is duplicated and decentralized: it’s not stored in one single place. You don’t go to the publishing house to read books, then go back home empty-handed. You bring that content to your home.

Let’s now apply our worst-case scenario to texts published in the form of web pages. Let’s imagine that all the owners of all the websites that I love disappeared all at once, the servers were shut down and all the warehouses holding those servers burned down. Psssht. No more databases to serve content.

What would happen?

Well, in this case you’re pretty much left with one single option: cherish the memory of the content you lost, because it won’t come back. Even if people saved HTML pages on their computers to archive them (no one actually does that, I’m fairly certain), it’s unlikely that you’d be able to rebuild the whole content of a website that dramatically vanished.

What we can do about it

To put it simply, we can do what books do. We should make it easy for people to duplicate the content published on our projects so it can be stored in various places (as opposed to a single server). Many blogs can be duplicated nowadays if they are stored on, say, GitHub. If you like a blog that runs on Jekyll, I encourage you to keep a local copy of it and let people know about it (I personally archive Divya Manian and Harry Roberts’ blogs).

However, depending on the size of your project and your editorial constraints, Jekyll may not be a viable option. I personally can’t contemplate working on the Pastry Box via git commands.

So here’s what I came up with to decentralize the content of the Pastry Box.

  • You can download all the thoughts published on this project through a simple API endpoint. Although you can specify various formats (JSON, HTML, Markdown), my idea with this endpoint is to let you grab a JSON string that could be used to rebuild the project if the website ever disappeared.
  • That JSON string, containing all the articles published on this website (as well as relevant additional information like names, slugs and pubdates), can also be found on GitHub.
  • Each article published on our website exists as a standalone HTML file. They’re all stored on GitHub.
  • You can use a nodejs module to print pdf versions of the articles published on this website.

I’m also working at storing our content on AWS S3 and Dropbox.

Decentralizing content consumption is only one step toward content preservation. The other important step is to keep articles visible and accessible over time.

Fighting the abysses of time

When you stumble upon an article you want to share, you usually hit your favorite social network and let your followers/friends know about that article. This is a great way to support authors and publications. Really. But a given user usually shares a given link only once and, while it may certainly lead friends/followers to re-share the link in question, the model of social networks gives only a short lifespan to the spreading and discovering of web pages: a new article replaces an older one, and so on, and links quickly disappear at the bottom of pages and look “old” or “irrelevant”. This is an undeniable fact.

Any webmaster who has made sense of her projects’ stats has noticed that traffic focuses only on the newest articles. Even when traffic constantly grows, which means that fresh, new readers connect to your project every day, only the newest articles are, in fact, read. Older articles are not read. That state of affairs (new readers reading only new articles and not checking older articles) makes it easy to deduce that older articles are left aside not because they have already been read (if that were the case, there would be no new readers, and traffic wouldn’t grow), but because they are old (and by old I mean they were published something like two weeks ago).

I don’t want the articles published on the Pastry Box to disappear into the abyss of time. We are lucky enough to generate material that can be labeled as “timeless”. So I decided to build a community around the content we publish, where people can share and store articles they like away from the buzzing fray of social networks.

Your Pastry Box account, and what you can do with it

You have the possibility of creating an account on the Pastry Box, and I’d like to tell you a few words about it.

Registering for a service often means receiving lots of unwanted emails. By default, we don’t send you any emails. If you want to receive emails from us, you will have to explicitly tell us by ticking a checkbox. We’re not trying to sell you anything. We don’t monetize our traffic.

So, identifying you gives us the possibility to let you do cool things with our website: customize how pages are rendered, recommend and store articles, etc. Once you have registered, you become the owner of a “profile page” located at Simply put, this page is an interface to share texts published on our website with other people (of course, you can choose not to display this page if that is your preference). It’s like some sort of feed dedicated only to our content.


The idea behind our “recommend” feature is quite similar to tweeting/sharing a link. When a text published on the Pastry Box hits a string, just click on the star at the bottom of the article to recommend it. You will see that this text is added to your personal page, making it easy for others to see a “history” of all the texts you recommended/liked. It may disappear from social networks, but it will be stored on the Pastry Box.


We have all stumbled upon articles that we want to share again, and again, and again, every day until the end of time. In actual fact, this usually translates into a “must-read” written in CAPITAL LETTERS, followed by the link to the article which caused an epistemological break in our conception of things.

We certainly all agree that a few words in capital letters don’t do justice to the awesomeness of such articles. But, on the other hand, we’re not going to actually tweet again and again about the same article, as if our timeline got hacked.

Enter “badges”. A badge is an article you think should echo in eternity. A badge represents content whose ideas you would live by, whose concepts, in one way or another, define you and what you believe in. Badges are texts that are important to you, that you care about.

Badges are placed at the top of your profile page. When people check your Pastry Box page and see your badges, they will know that they should absolutely not miss those texts.

Isn’t vouching for awesome content a thing we’ve all wanted to do? That’s what I’d like to try here.


Another thing that I wanted to focus on is to let you customize how you’re going to consume this website. You can check the customize page to see how you can alter this layout. The main idea is to let you download only the bites that you actually need. For example, if you’re connecting from a mobile device, you may want to improve performance by not downloading images. You can do that.

When you customize the Pastry Box, your settings are stored in a cookie (read our tos). This means that modifications are device-dependent (as opposed to “account-dependent”): you can customize the Pastry Box on your mobile phone in one way and on your desktop in another.

Happy reading!