Towards the Vision of The Decentralized Web!

decentralization

#1

Yesterday I bumped onto this cool article that will appear in the March 2018 issue of Esquire magazine:

The gist of the article is - as the title says - that Big Tech (especially Amazon, Apple, FB and Google) has effectively monopolized the internet. They dominate and make the rules and, as you all know, have huge effect on our society. Also they mostly just go for the bottom line, in the interests of their shareholders only.

As a result the web has become, sort of, well… broken!

It will be really hard - even when Humane Tech community and other such initiatives thrive - to break the power of these monopolies. One important way to achieve this is by breaking the centralization of the internet.

The web as we currently have it has not evolved in the way the original creators (like Time Berners-Lee) intended: decentralized! The vision of The Decentralized Web still exists, however, and there are many efforts to (re)decentralize our current web.

But first… what is The Decentralized Web? Well, there are many definitions as this article shows: What Is the Decentralized Web? 25 Experts Break it Down

The one definition I like is:

" A Decentralized Web is a network of resources in which no one player can control the conversation or spin it to [his or her] exclusive advantage. "
Simon St. Laurent, O’Reilly Media

Please read on towards the end of this article where you’ll find a number of descriptions of how this decentralized web would affect the average users. Note how closely aligned these views are with the goals that the Humane Tech community likes to achieve!

With the rise of Bitcoin and the blockchain many people have recently heard about the term ‘decentralization’ and the changes it may bring. But realize that this is just an application of decentralized technology, and the field is much broader than that!

Personally I have some experience with Dat Project (and the cool chromium-based, decentralized BeakerBrowser written on top of it. Also see their Github awesome list), Secure Scuttlebot, GunDB (a cool decentralized graph database) and Solid (led by Tim Berners-Lee, and not only decentralized, completely standards-based, but also a semantic web or linked open data technology - or The Web of Meaning, where web resource semantics are machine-readable as well).
But there are many other projects and emerging standards (like IPFS). See for instance this list:

With current decentralized technology projects it is perfectly possible to build production-ready Twitter-like and realtime peer-to-peer messaging systems. But the field is still young and as of yet (as far as I am aware) there do not exist full-blown application frameworks for more complex web-based decentralized products.

Also the big tech companies and even governments are not that keen to promote this technology, mostly due to the risks they pose to their business and practices.

I am really curious to hear your opinions on The Decentralized Web and the role that the Humane Tech community can play in promoting it!


Mozilla just released The Internet Health Report v1.0
Redesigining the Web
Humane Technology reading lists
Investigating privacy-respecting online identity, data ownership & control solutions
Deplatforming, freedom-of-speech and tech monopolies
Mark Zuckerberg as the Greatest Self-Help Guru of All Time?
Mark Zuckerberg as the Greatest Self-Help Guru of All Time?
Mixed-media Communication Problems and sensory overload
Making data work for you
Making data work for you
Idea - Humane advertisement policies: ethical and mutual-beneficial
How do we avoid a Brave New World, reverse dystopic trends?
3 Criteria of Humane Technology
#2

We have been working on a decentralized solution for some time. We now have two sample sites School Libraries Network and WebSonar Libraries Network
We will soon be seeding 2000 schools with this solution.


#3

That is really cool @Cowboy! What kind of tech did you use to build the apps? Also what made you choose to go decentralized? Privacy… control of personal data?


#4

It actually started out in 1987. My father had started a verbatim reporting firm and had a contract with the British Columbia Utilities Commission to transcribe their regulatory hearings. BC Hydro would retype the the transcripts into their IBM STAIRS document archiving system for research. This gave them an advantage over the Intervenors who then asked us if we could provide a similar service. We found Virginia System’s Sonar application for the Mac and started working with him to develop it for the legal market. You can read more history of the process in my library Click Browse the Library and click My Story then PDF and scroll down to 5: The Macintosh. We have an alternative to FaceBook that is not addictive.


#5

Just a followup on the privacy issue. Privacy is sold to finance centralized tech. We have an iPad app that we can sell that will finance our ability to provide the libraries to schools at no cost We do not collect user data and use dynamic conversion on port 8088 to eliminate the robot tracking. We also provide a decentralized communication solution with Mastodon We will be launching in the next week or two so if you know of some schools that have one 2 one iPad programs they will be the ones we are targeting initially.
Link to the Announcement


#6

Thx for sharing @Cowboy. Indeed Mastodon is a good example of a decentralized social network that is very popular / widely used. I’ll add a link to the Reading List thread :slight_smile:


#7

This very interesting article by Ruben Verborgh details 5 paradigm shifts that will result from adopting decentralized technology on the web:

From the article: “Ultimately, decentralization is about choice: we will choose where we store our data, who we give access to which parts of that data, which services we want on top of it, and how we pay for those.


#8

I saw that article too. I too hold promise for a more decentralized Internet. That said, I’ve been around long enough to also have a healthy dose of skepticism as well.

For example, I see a lot of enthusiasm for blockchain and DAOs (Distributed Autonomous Organizations) which borders on the misplaced. There is often an air of “Governments would be so much more efficient if it weren’t for all the people”.

I say that half jokingly. The facts are that no matter how you slice it, the efficacy of all business and organizations comes down to human relationships. And what we most value from them comes in the inherent messiness of those relationships. Automating these relationships out of existence makes for nice science fiction fantasy, but it is anything but a humane technology future.


#9

Hi @greg! I share your skepticism about blockchain… It was not mentioned in the article, only in the comments, and in my first post in this thread I mentioned it because most people nowadays think that decentralization == blockchain, which is not true of course.
Personally I’ll wait till the technology matures a bit further. Right now it is at the peak of its hype cycle… a solution for everything. :slight_smile:

Having control of your own data, little or no vendor lock-in, and privacy by default still needs a paradigm shift. It won’t however directly impact the nature of human relationships, but decentralization may be better in supporting the inherent ‘messiness’ of them.

But we are a long way from realizing The Decentralized Web, and big tech does not want to do away with their walled gardens any time soon. Adoption will have to come from a grassroots movement, I think…


#10

Hey @aschrijver

Don’t get me wrong. I think there are a lot of exciting opportunities for blockchain – even if I really just think of it more as a public domain database. But there are people extolling the “revolutionary virtues” of DAOs in the belief that automated computer systems will save the day if only we get rid of all the people. Well, they’re not saying it in those words exactly, but it is essentially the underlying premise that excites them that they haven’t acknowledged or come to terms with yet.

The web’s original promise was about decentralization - how that with typing a public URL you could access any information published anywhere without gatekeepers. But social self-organizing, interest in aggregation and scale, etc., and reality sets in. You cannot upend a human legacy of consolidation and control and not expect those urges would merely sit passively in the spectator seats all the while.

I don’t necessarily feel that makes me a cynic per se. But more a realist that any technology we deploy is inherently human, and we can’t escape our humanness for very long and prevent it from seeping back in just because we deploy a newly disruptive technology. Meet the new boss: same as the old boss.

So what we get to hope for are windows of decentralized opportunity followed by motivation to aggregate and re-centralize on the new paradigm all over again.


#11

Maybe… possibly… probably… but new, better models could arise from this. E.g. having your own devices be ‘thick clients’ once again (even smartphones have enough processing power for that now), paying for services to be installed and managing your data on the device. Cutting out the man-in-the-middle cloud services (don’t see cloud disappearing… they will co-exist, but there may be a better balance).
I am not really optimistic that this will become a big thing anytime soon, but not trying will only worsen current situation :slight_smile: Thanks for your views @greg!


#12

Btw, if you haven’t seen this Aeon post, it’s a great read on this topic:


#13

This is not really The decentralized web? Or was The silk road based on bittorrent… don’t know, never used it.
But when you want to do something really private on the internet - not talking illegal stuff, but maybe searching on a disease, legal advice, etc. - it might be a good idea not to type it directly into Google search, but use Tor browser and/or a VPN.
Anyway, this article then is also a good read:


#14

I see it as quite analogous. It’s another case where the idealism and egalitarian aspirations of a decentralized “rule of code” (as opposed to rule of law) go awry because we live under societal rules. We are inherently among people-based systems of organization – it ultimately comes down to human relationships in the end, no matter how many layers of decentralized code you throw on it.

No coding in the world is going to eliminate the fact that we’re dealing with humans and human nature underneath it all. As I jokingly referenced wrt DAOs, the only way to get past that is to eliminate the people.


#15

I’m not a tech so I can’t respond the same way at as most people here but… just today I said to someone I’m growing weary of google and amazon s influence on our subconscious thinking- they distract me from regular tasks I have to do. It takes real discipline to get anything done like pay bills. It’s almost like if you have ADHD a case could be made for interfereing with a persons ability to function- making attention deficit worse.

Anyways… I come on here to be validated- I cannot stand seeing people looking at their phones- I need a bumper sticker akin to “kill your television”.


#16

I’m not up to date on this post, but I’ve been advised that an article I’ve posted elsewhere is more relevant to this topic. So I’m reposting here and for now, I just want to get the link up. Over the course of the weekend I’ll integrate my thoughts with what’s already been posted.

One element I do want to highlight in the article, now that I’m rereading it, is the discussion about Storj and its similarity to Pied Piper on the sitcom Silicon Valley. I have no idea which came first, Storj or the tv show, but Tom Simonite, the article’s writer, expresses something important, I think in the palette of human behavior and human thought. The idea of expressing something unknown in terms of something familiar.

Going as far back as Aristotle, this proficiency at metaphor was deemed that which makes us uniquely human. More recently, though, in the work of behavioral economists like Dan Ariely, Dan Gilbert, Daniel Kaufman and Nicholas Nassim Taleb, we’re looking at the propensity to metaphor as something more dangerous in that it promotes recessive thinking strategies inadequate to new challenges.

For the moment, here’s the article:


#18

You need to find services whose incentives are aligned with yours. What asks do you have from your social media platforms? It probably starts with basic things:

  • I want to keep up with my family and friends
  • I want my family and friends to be able to keep up with me

But if you’re smart, you might have some deeper asks:

  • I don’t want my personal information sold to others
  • I don’t want to be manipulated into spending my money

We might even have some asks as a society, too:

  • We don’t want to be manipulated into hating our countrymen
  • We don’t want to have our people’s opinions radicalized

https://drewdevault.com/2018/03/24/Decentralize-decentralize-decentralize.html

and the interesting discussions on HN:

https://news.ycombinator.com/item?id=16673091


#19

Let’s be realistic. I actually run a tiny internet company that deals with billions of tiny pieces of information so I have some real experience.

With information tech, centralization usually works much better than decentralization. It’s simply too hard to manage all this ** information! Databases are complicated, and managing data quality is even harder. A decentralized web search engine would never be able to come even close to the quality of Google. A decentralized social network would probably have all of its “independant” hubs secretly run by the world’s powerful governments, waiting on the opportune moment to unleash them for hybrid warfare or to influence an election. Though I’m sure there are a few possibilities for decentralized internet and I hope very much they do succeed, but let’s think of some things which can actually compete head-on with the monopolies in terms of quality.

I think the more likely scenario is that we’ll soon (hopefully) see more choice in the future, caused by the big monopolies competing against each other and against new rivals. This will happen as tech industry “matures” and becomes more like a normal industry. The big tech companies are starting to compete more in each other’s product “territories”, and more of the developing world’s tech products are going to go head-to-head with Western products. Actually all of this is already happening or is already in development if you watch the product development pipeline. So instead of decentralization we’ll see more options to choose from different centralized networks, and this competition and “maturing” will drastically change “business as usual” and that will be for the better.


#20

Yes @anon76657042, you make lots of valid points, and for sure some fields of IT - like search - will be very hard to decentralize. Also decentralized systems tend to become more centralized over time, like the often mentioned example of email. While email is inherently decentralized, people have flocked en masse to using Gmail, because of its spam protection or just convenience, and effectively centralizing it.

The core protocols of the internet like TCP/IP and DNS are decentralized, the centralization happens in the upper layers of the technology stack. Here the monopolists drag you into their walled gardens - vendor lock-in results - making it really hard and costly to ever leave once you are invested. And you are subjected to the rules of the provider, which may change at any time (like e.g. FB news feed algorithm recently). They may even decide to shut down your business on a whim if they so decide, block you from their network… which occasionally occurs.

For decentralization to be successful there is a need for more decentralized, federated protocols to become standardized and widely adopted. While this is against the interests of the monopolists, it would open up the markets giving more chances for smaller players to actually succeed. Currently you have innovative startups that are just targeting to be gobbled up by the big players for a couple hundred million dollars, after which the founders either take a position in that company or move on to a new startup venture.
There have been interesting discussions on Hacker News that companies like e.g. Google are not all that innovative anymore (except in some specific areas, like AI) and just buy up the competition. Thereby stifling overall innovation.

Even the recent Cambridge Analytica scandal may eventually be to the benefit of Facebook. Mark Zuckerberg has apologized and said FB is open for more government regulation. This regulation - when it comes - will not be targeted to a single company however, but all across the board. FB - as large as it is - can easily afford dealing with this, while small startups will have a much larger struggle to comply (see: TechCrunch: Regulation could protect Facebook, not punish it and techdirt: How ‘Regulating Facebook’ Could Make Everyone’s Concerns Worse, Not Better).

Like you, I hope there will be more competition amongst centralized IT competitors as well, with preferably different business models, like subscription-based rather than ad-based models. Have users become the customer again and good privacy policies and ensurances in place. Its just that current monopolists are so incredibly entrenched, having revenues larger than many individual countries and just growing ever larger, which will make this harder and harder.

Now, decentralization based on good standardized protocols give you choice and interopability, avoiding walled gardens. As an example: The W3C last January made ActivityPub a recommendation - a decentralized social networking protocol. Already a number of open-source projects are creating new social networking apps based on it, or are rewriting their stack to support it.

A notable example here is Mastodon (“Social networking, back in your hands”) - a decentralized microblogging network comparable to Twitter. With Mastodon anyone can spin up a server and attract users to it. These micro communities are self-contained, but users can decide to follow users on other servers, thus creating a larger interconnected network structure. Unlike Twitter - where you have a single moderation policy - on Mastodon every server can set their own rules, topics and guidelines and you are free to choose based on your preferences.
Note that Mastodon - based as it is on ActivityPub - interoperates with other software products, something you won’t find e.g. between Twitter and Facebook.
Setting up a server - while requiring some knowledge - is not all that hard and does not require you to have intricate knowledge of database management and such. This is all managed by the application, which is open-source and has more than 450 contributors as of now and many more people watching and reviewing the code.

I think the argument of “all hubs” being compromised by governments or otherwise foul actors doesn’t hold much sway (in fact this argument is often used in scare-mongering tactics against The Decentralized Web). Sure, a server can be compromised, but so can the servers of a centralized application provider. I believe American companies (and any company doing business or hosting server in the USA) are already required by law to provide backdoor access to their information to government agents, like the NSA, or at least there is regulation like this in the works (there are even conspiracy theorists that claim that FB was founded by the CIA - which I don’t believe, but it is not impossible).
Large social networks may be just as vulnerable to attacks, riddled with zero-day exploits in closed-source services, and big companies like Yahoo and LinkedIn have been hacked already leading to millions of accounts to be stolen. This has arguably a much larger impact than having to hack each individual decentralized server or even individual user accounts (depending on how encryption policies are applied).

Note that there is a form of decentralization that goes even deeper than a federated system (like Mastodon) that needs servers (or ‘hubs’), and that is a fully distributed, peer-to-peer system where every device is both the client and the server. Our laptops - and even our smartphones - have become very powerful computing platforms that are up to this job. Example implementations (earlier mentioned by me in this thread) include Dat Project, IPFS and Beaker Browser (based on Dat Project). Beaker allows you to host your own decentralized website, and as people visit your site, it will get more peers hosting it, thereby increasing performance and availability. Very interesting.

ActivityPub and other decentralized protocols offer new opportunities to companies, new business models. Like any commercial entity they can ensure they operate with integraty, good ethics, and have good compliance regulation in place, certifications, thereby gaining the trust of their user base. There is huge innovation in encryption technologies, not just in encryption of data, but also in having strong authentication and digital identities (see for instance Rebooting The Web of Trust, here and here).

In short, I think, The Decentralized Web has the potential to become the web of the people again, rather than the playground and cash cow of the few :slight_smile:


#21

There is already a project in the works to use open standards for chat. This would use existing email servers. I hope this works!

Meet the open sorcerers who have vowed to make Facebook history

Note the title is a bit misleading (sensationalised / fake news) as they are actually taking on chat (Messenger, Whatsapp) and not the Facebook social network itself.

As you can see from the article, the biggest hope is in replacing simple things like chat. I think it can be done! But then there is the problem of no money, app developers being unpaid or themselves having to resort to ads. I imagine most people will still be drawn to the kinds of shiny features (such as Apple’s unicorns with rainbow horns) offered by the big monopolies.

I would love to see the open web, a dream of mine for years. But what is theoretically possible differs from what is practically possible given technology and business models. The best approach I can think of is a foundation, a charity. There, people would donate tens or hundreds of millions of dollars and that would be used to create a perhaps centralized but open and democratic alternative to Facebook or Google. However that has been tried before and failed:

Quaero