Decentralized Web Primer

Updated 2 years ago

electronic max (@electronicmax) started discussion #2

2 years ago · 2 comments


Change title of book to "IPFS Primer"

People outside the IPFS community do not see IPFS as the "Decentralized Web". It neither fully embraces Web principles nor is endorsed by the W3C. To avoid misleading people, and also improving the ability for people who actually want to learn about IPFS to find this primer, please consider renaming this book.

Matt Zumwalt @flyingzumwalt commented 2 years ago

You're right that the current contents of this draft don't match the title. I will be doing a lot of writing at the end of this month to get to a 1.0 version of the primer that will actually live up to the name. If the contents of the primer still look like an IPFS Primer rather than a Decentralized Web Primer at the end of January, I will rename it. You can hold me to that.


I started sketching a decentralized web primer before I even looked at IPFS. My attention was primarily on dat and scuttlebutt at that time. Over time I started working on IPFS and creating tutorials for IPFS that conformed with the dweb primer's format. It made sense to situate those tutorials within the overall curriculum of the primer, partially as a way to keep some momentum towards writing the primer and partially to give people a view into the larger picture as the primer evolves -- I think it's important to be able to show how a particular tutorial fits into a bigger context of concepts and technologies (but they don't have to all be in one single primer).

Practicality (workshops, blog posts, etc) drove the creation of the tutorials you see. The result is a set of tutorials that are, I agree, an IPFS primer rather than a decentralized web primer.

Meanwhile a lot of the content that does belong in a decentralized web primer has evolved elsewhere, primarily in presentations and blog posts. I have set aside time at the end of December to gather that information and render it as written sections of the primer. That will also drive re-structuring the curriculum/outline of the primer. I might relocate most of the current tutorials to an IPFS primer, but I do want to include working hands-on tutorials in the Decentralized Web Primer. Most of those tutorials will probably use libp2p and IPFS because I can commit to keep those tutorials up to date as the technologies evolve, but I'm open to incorporating other technologies and protocols where appropriate as long as there is a clear way for me to keep the tutorials up to date.

electronic max @electronicmax commented 2 years ago

Thanks for the thoughtful reply & consideration of my comment (which I didn't intend to be critical).

I think that your ambitions to do a decentralised web primer sound great; perhaps that will be a different book that surveys at a high level, the various efforts in the space (?). Interestingly, terminologically speaking, we (connected to the web foundation, W3C, and greater family) find the term "decentralised Web" weird because the Web was designed to be (and remains) a textbook example of a decentralised system. These days, with the rise of the mega-platforms we are seeing an artificial centralisation of sorts, so you will hear us talking about "re-decentralising the Web" ([](

But I do agree that the class of technologies comprising p2p and various kinds of distributed computing layers on the Web are very interesting and also quite disjoint (unlike the Web, most are not yet designed to be interoperable! - eg dat <-> ipfs, etc - correct me if I'm wrong) and it would be great to see a book introducing these new frontiers as well as what they might mean in a greater Web ecosystem.

I think that any such book should also definitely include work from the W3C (as a semi-democratic union of interested parties who loosely define what the "Web" are), including LDP, SOLID, ldn, among other systems that envision a web in which social protocols allow data to stay in situ, but otherwise do not over-specify the workings in order to maximise interoperability. Would be happy to contribute insight & help in any way I can. I think that distributed computing people would call these approaches "federated" but they're federated to the extreme, where each person's data is on their own web server, so they're definitely in the same class.

Anyway sorry for the ramblings and good luck.

to join this conversation on GitBook. Already have an account? Sign in to comment

You’re not receiving notifications from this thread.

2 participants