How Tim Berners-Lee Will Fix the Internet (reuters.com) 63
"Sir Tim Berners-Lee, the British computer scientist who was knighted for inventing the internet navigation system known as the World Wide Web, wants to re-make cyberspace once again," reports Reuters:
With a new startup called Inrupt, Berners-Lee aims to fix some of the problems that have handicapped the so-called open web in an age of huge, closed platforms such as Facebook. Building on ideas developed by an open-source software project called Solid, Inrupt promises a web where people can use a single sign-on for any service and personal data is stored in "pods," or personal online data stores, controlled by the user.
"People are fed up with the lack of controls, the silos," said Berners-Lee, co-founder and chief technology officer of Inrupt, in an interview at the Reuters Next conference... John Bruce, a veteran technology executive who is CEO of Inrupt, said the company had signed up Britain's National Health Service, the BBC and the government of Flanders in Belgium as pilot customers, and hoped to announce many more by April...
A key aim for Inrupt is to get software developers to write programs for the platform. Inrupt, like the original web, is at its core mostly a set of protocols for how machines talk to one another, meaning that specific applications bring it to life.
"The use cases are so broad, it's like a do-over for the web," Berners-Lee said.
In a video interview, Berners-Lee tells Reuters that what people are worried about isn't privacy per se but "the lack of empowerment" — for example, to collaborate with people. And then he acknowledges that the worldwide web does suffer from limited access control. "I wanted it to be a collaborative space, but in a way that was naive, because collaborative spaces you need to be private. You need to start off with just a limited sharing, and then you allow the sharing to increase."
Social networks provided some features like a unique login and identity. But unfortunately, then "The large social networks will tend to get larger" — and ultimately without a single global signon, users then become trapped in separate silos.
"People are fed up with the lack of controls, the silos," said Berners-Lee, co-founder and chief technology officer of Inrupt, in an interview at the Reuters Next conference... John Bruce, a veteran technology executive who is CEO of Inrupt, said the company had signed up Britain's National Health Service, the BBC and the government of Flanders in Belgium as pilot customers, and hoped to announce many more by April...
A key aim for Inrupt is to get software developers to write programs for the platform. Inrupt, like the original web, is at its core mostly a set of protocols for how machines talk to one another, meaning that specific applications bring it to life.
"The use cases are so broad, it's like a do-over for the web," Berners-Lee said.
In a video interview, Berners-Lee tells Reuters that what people are worried about isn't privacy per se but "the lack of empowerment" — for example, to collaborate with people. And then he acknowledges that the worldwide web does suffer from limited access control. "I wanted it to be a collaborative space, but in a way that was naive, because collaborative spaces you need to be private. You need to start off with just a limited sharing, and then you allow the sharing to increase."
Social networks provided some features like a unique login and identity. But unfortunately, then "The large social networks will tend to get larger" — and ultimately without a single global signon, users then become trapped in separate silos.
NO. Please just don't. (Score:5, Insightful)
He should take up needlepoint or something that doesn't affect most other people, instead. I think he should shut his mouth and finding something else to do.
Re:NO. Please just don't. (Score:4, Insightful)
If he wants to fix something he can fix a sandwich.
Re: (Score:2)
We used to have beautiful things like GTK, MFC, Qt, and the like
[as DJT] They were beautiful, SO beautiful, you wouldn't believe how beautiful they were.
Seriously though, all 3 of those had XML template versions, which is HTML in a nutshell.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
He just needs to remember to use sudo command. ;)
Re:NO. Please just don't. (Score:5, Interesting)
Indeed. A single login for everything is a terrible idea. I prefer to keep my online identity segmented.
I like silos. I don't want my boss reprimanding me for my Slashdot posts.
Re: (Score:2)
Thank you Mr. Bernes-Lee (Score:1)
Re: (Score:3)
Have there been any online personal data stores that did not at some point get hacked and leaked? Any at all? If he thinks I'm going to put the keys to my kingdom in the cloud, he's on crack.
Berners-Lee will always have my admiration and respect for his pioneering work, but he doesn't seem to grok how nasty and dangerous the world he created has become.
Re:NO. Please just don't. (Score:5, Informative)
Re: (Score:2)
But what stops data simply being copied? If I give permission to access my name and address to say eBay so I can receive a package, what stops them simply copying it and selling it to others?
I'll tell you one thing that stops it: GDPR. Legal consequences to not obtaining affirmative, opt-in permission. Inrupt, not so much.
Re: (Score:1)
When my colleague brought this project to my attention, I was hit with the same concern. How can we trust that a so-called secure pod is not going to be hacked/back-doored/etc? We cannot. It looks like there is no single solution to privacy security, especially this project. Too many questions to be answered, and this project falls short on addressing them.
Storage Security: If the data is stored on the cloud or hosted, how can we be sure that the entity doing the actual storage is not going to scrape a
This is just deananomizing (Score:3, Interesting)
Re: (Score:1)
Re: (Score:2)
This reminds me of the brain-dead shit from the 90's that parents would put on their teenagers computer that would block the world "breast" or "penis" and similar nonsense, and thus blocking information about cancer, etc.
Slashdot can be VERY Catholic at times.
Re:This is just deananomizing (Score:4, Insightful)
It also brings out the best.
Re: (Score:1)
Re: (Score:3)
I don't know/understand the technical details of how he plans for things to work, but I think there are a couple of different concepts that people conflate:
Spoiler Alert: He Won't (Score:5, Insightful)
TBL enabled the current state of the Internet, where tech companies have built data-driven walled gardens, and baited unsuspecting Luddites with trails of poisoned candy. From the mid-2000s, it was clear that "developer interests" mentioned in any Working Group correspondence referred to big tech, not individual workers.
The W3C should have been absorbed into a more stable, formidable, and respect-worthy organization a decade ago, such as IEEE or IETF.
Big Tech has exactly the environment they want, and there's no incentive for them to change any of it.
Users are no longer the consumer, they are the product source. If the Internet was a dairy farm, we are the cows, and the farmers only care about bringing the milk to market.
Re: (Score:2)
Re: Spoiler Alert: He Won't (Score:2)
Re: (Score:1)
How long? (Score:2, Interesting)
Re: (Score:3)
Yeah, I agree that Tim Berners-Lee seems to have little to show recently for so much resources.
Likewise, I too have worked in this space on and off over the last decade or so, also self-funded and in my spare time. Example: http://twirlip.net/ [twirlip.net]
The video there is from when I was considering doing a kickstarter a decade ago -- but ironically I ended up instead working for years for a major media company (NBCUniversal) as a contractor instead to pay the bills.
I'm hoping in a few more years to be able to "retire
Re: (Score:2)
Press releases make more money (through venture capital) than a shipped product that no big tech company will touch with a 10ft pole, and nobody else will even care about.
oh wow (Score:1)
Re: (Score:2)
Data vs. Applications vs. Identity (Score:4, Informative)
The goal is to separate data from applications and to provide a decentralised way for authentication. Users can store and manage their data in so-called pods (personal online datastores).
Pods are essentially souped-up HTTP servers (based on Linked Data Platform containers) that allow the management of graph-structured data (RDF) via HTTP. Authentication for accessing and manipulating data is done via WebID, which builds on OpenID and OAuth.
Overall, the technologies (unsurprisingly) fit well into web architecture and are simpler and more developer-friendly than the previous technologies around the Semantic Web. With a focus on running code and practical applications, I see that his effort is gaining traction. I hope he succeeds, it is time to get alternatives to the centralised cloud systems run by big companies.
Best ... (Score:2, Informative)
If he wanted to "save the webs" then he would want to ban JavaScript and the execution of arbitrary code in the Web Browser, and ban DRM.
He doesn't want these things. He is part of the problem, not part of the solution.
Re: (Score:2)
ALL RIGHT, let's jump on the JavaScript hate bandwagon!!!
Do you know who Guy L. Steele is?
Toxic anonymity (Score:2)
Deanonymizing users of social media would do a lot to solve the toxicity problem. If Twitter could blue-check every user, it would be a much more tolerable forum of discussion.
Privacy (Score:3, Informative)
>"Inrupt promises a web where people can use a single sign-on for any service and personal data is stored in "pods," or personal online data stores, controlled by the user."
I don't want "single signons" with different systems, tied back to some central place. Unless I can still control which logins I want to use. If something like this becomes popular, governments will move in, trying to mandate ID. And/or big business will force unified logins and force tying it to something like your phone (in the name of 2FA), which reveals identity. This is already happening.
>"Lee tells Reuters that what people are worried about isn't privacy per se but "the lack of empowerment"
I don't care what users THINK they want. *I* want privacy, a key part of security, more than I want convenience. "People" probably want privacy and security also, they just don't realize it or realize it YET. And once they do, it is too late.
Re: (Score:1)
WebID provides a "single signon" but based on a decentralised way of managing user ids.
With WebID it is in principle possible to host your own identity provider (identified via a HTTP URI), which takes care of authentication. So you can control yourself what login to use.
If somebody wants to give you access to a resource, they add your WebID (the HTTP URI) to the access control list of the resource, which takes care of authorisation.
A pity that WebID+TLS, with which you can easily set up WebID via yo
Please rest on your laurels (Score:1)
Re: (Score:1)
No (Score:1)
I would like more "silos" please.
Sites that don't interoperate with each other and share my information with each other. Sites that don't require me to log into a different site just to get access. Sites that are truly independent of each other.
Farmers invented silos for a reason, dude! That's my grain, not yours.
been done (Score:1)
Sorry Tim, cats out of the bag (Score:2)
To change the internet, you don't just need an technological argument, but an economic one as well. The tech giants aren't going to change and give up thier money. Much of the internet is supported by ad revenue and that's not going to change anytime soon
Re: (Score:2)
The "Internet" is not supported in any way by advertising revenue. Only the "criminal class" is supported by advertising and spying revenue. Unfortunately most of the "Web" is inhabited by the "criminal class" -- and the "Web" is an application which runs on the Internet.
Re: Sorry Tim, cats out of the bag (Score:2)
Some internet *content* is supported by ad revenue. The internet infrastructure is largely not ad-supported. It is supported by ~2 billion people (give or take a billion) paying ~15 dollars/month (give or take 10 dollars) on average to telcos (including mobile companies and ISPs).
Centralization, Decentralization cycle (Score:1)
Most comments here seem to be dismissive without acknowledging the core problem.
Before 1995, at the being of a public internet, power was centralized. CompuServe or AOL. Walled gardens.
But they could not keep up with services offered by the larger forming internet, and eventually just became competing ISP's to give you internet access. The rest became irrelevant.
In 1995, there was no centralized power on the internet.
Protocols were all decentralized for:
- personal web presence.
- chat
- newsgroups
- email
Now
Focusing on the wrong part (Score:2)
You guys are all focusing on the "single sign-on" part, which isn't the important part. There's nothing that would prevent you from creating alternate profiles to keep parts of your life separate if you like. The important part of his plan is that the data associated with a profile is kept in a single location, under your control. Access control is very granular, probably down to the field level. Grant Company A access to your name and email only. Grant company B access to your name, email, street addres
Sounds like more 'cloud'-based shit to me (Score:2)
Re: (Score:1)
Until...
a) Your computer dies (and you don't have an updated backup ready) or gets corrupted,
b) Your computer is hacked/hijacked,
c) Your access to the Internet is compromised/banned,
d) You accidentally delete your pod,
e) You butcher your pod because you are bored and everything in it is a lie,
f) You do not have the experience to properly maintain a computer,
g) You use an operating system that does not give you complete control over your equipment (and will likely shove everything into the cloud in the near
Networks don't work if everyone does their own (Score:2)
This will never take off, if everyone does their own. Berners-Lee should have done his research and just joined an existing project instead of starting his own. Hubzilla, for example.
Brochureware and source-code, no midde-ground? (Score:2)
I seem to often run across projects that sound potentially interesting, but (as it appears in this case) if I want to know what the heck they're really doing there's either a site with a handful of graphics and maybe an incomprehensible diagram or two with some marketing phrases, or a source-code repository. If I wanted to know what a "pod" is and how it works, I have a choice between Inrupt's awful brochureware site where I can watch vague buzzwords zoom in and
Get rid of distributed client-server (Score:2)