Web Browsers and Web Weavers

The very nature of the Web Browser is changing at breakneck speed. Because of html5 and cloud synchronization services, Web Browsers are beginning the trek from thin client to thick client. In order to compete with the user experience of mobile os platforms, apps are quickly becoming first class citizens of the web. Soon, apps will be an inextricable part of the Web Browser experience.

Some really cool stuff is about to happen. Unfortunately, the open web is in peril (again.. just like princess peach). If the web is becoming a web of apps rather than documents, then the open web must be about links between open data sets. Instead, we have data silos.

Web Browser data in the cloud is silo’d (think Firefox Sync, Google Chrome sync, iCloud, etc). Uers app dashboard data will also be stuck behind these high walls. The social graph is still stuck in 2007, and we haven’t even gotten around to graphing anything else on the planet.. let alone in a decentralized manner.

Identity is also (finally!) moving into the browser, but not as a first class citizen like Apps will be. This is a tragedy because the Web Browser is supposed to be the User Agent to the web.. not the app agent to the user.

I’d like to start a dialogue on how we can take the best of what is happening in computing right now and converge upon a new product class that rethinks the User Agent.

I’d like to call it the Web Weaver. Lets see if it sticks.

So, what is a Web Weaver? It’s the persistant User Agent in the cloud that acts as a counterbalance to the Web Browser. You can think of it as a personal web server that ties your digital world together into a single experiential whole.

The Web Weaver is also an app! It represents you when you’re offline. It syncs your personal data between all your software, hardware and social connections when you’re both on and offline.. It stores and manages your social graph. In fact, it stores, manages and perpetually extends your graph of graphs.

I may have parts of this equation wrong. Perhaps the Web Weaver doesn’t store all this stuff. Perhaps storage is an extension. In which case, the Web Weaver delegates functionality and acts as the service routing mechanism between applications within a users personal web ecosystem.

Most importantly, the Web Weaver is cross-compatible with every Web Browser and other Web Weavers. This means you can have multiple Web Browsers and multiple Web Weavers running, syncing, and presenting a single coherent identity to the rest of the web.

The Web Weaver isn’t the center of your digital experience, it’s your outer cellular membrane.

Let me be clear, I am not proposing a new software dev project.. there are projects all over the place that are headed in the Web Weaver direction. My goal is to surface where this could all be going so we can get there faster.

Here are some products that are already bound toward Web Weaverdom:

Many of these seem like wildly different product offerings, but there are commonalities between them that seem to hint at key convergence opportunities between them all in the near future.

I’ve identified the Web Weaver as the missing requisite piece to solving global scale problems within the urgent timeframes we face. It is therefore my intent to make sure that this product class is fully defined, specified (specificationified?) and in some form shipped by the end of this year.

8 Comment threads
5 Thread replies
Most reacted comment
Hottest comment thread
4 Comment authors
Captain Lyre CalliopeAugustin BralleyUnhostedWeb Browsers and Web Weavers | The Future of Networking « Jose Murilo’s WeblogSamuel Q. Green Recent comment authors
newest oldest
Notify of
Captain Lyre Calliope

Comment system check.

John Reynolds

Excellent post, Captain Calliope! The Web Weaver sounds like exactly what is needed to put users back in control of their data on the web, and Mozilla certainly has a foot in the right direction with “Personas.” Regarding storage, a decentralized p2p solution would be rad!

For those desiring further illustration of just how much the web is changing with HTML5 and how Mozilla has been fighting to keep it open, check out Christian Heilmann’s talk, “The Web is the Platform,” given at March 24th Mozilla Developer Network Hack Day at in NYC: http://christianheilmann.com/2012/03/25/the-web-is-the-platform-presentation-at-mdn-hackday-in-nyc/


[…] stated in my previous two posts, my focus for the year 2012 is to bring a product I call a Web Weaver into existence. The […]

Sepp Hasslberger

an interesting concept, to be sure. 

If I understand this right, the web weaver will be our own personal user agent out there, exposing exactly as much as we decide it should expose, and keeping some stuff from prying eyes, that we don’t want everyone to know. 

Perhaps a simple way of exchanging value through the web, a “payment” system that is agnostic as to what kind of money or points system or whatever we want to use, and that allows for use of multiple “currencies” of our choice, should also be a part of this?

Captain Lyre Calliope

Yes. Ownership and management of personal data is key as is the ability to set terms for information exchange.

A payment system based on legacy human currency would preferably not be core to this system but perhaps a default app built on top of the Web Weaver platform architecture. A baseline system for authenticating transactions of any kind probably would be a core security feature.

What I’m proposing here isn’t just a glorified server as user agent.. its also a new architecture and ecosystem for web app development. If the developer ecosystem is put together right, Peer to peer based apps would be the norm because all the complicated peer to peer protocol stuff would be built into the web weaver itself.

Think of it this way. If all the apps you used were built for Web Weavers, you could very well decide to use your web weaver as the back end for all your apps which means you could be using facebook’s front end without any of your data touching their servers. However, you’d lose the functionality that Facebook is able to provide because they have everybody’s data to correlate and create extra value for. If you want to in on that value, you can decide to share your data with Facebook in the same way you’d share it with a friend. Facebook becomes just another social node.. albeit one with lots of value to bring to the table if you share your data with them.

And if they want to make money of your data.. you should be able to get a cut of the profits. The web weaver would give us the leverage to share our data if and only if the transaction nets us agreeable benefits.

So, yes. Not necessarily payment.. but some means of setting transaction terms for data exchange are key.

Samuel Q. Green

This may be a bit obvious to some who’ve seen the series, but would I be correct to say that you could use the term “Ghost” as a suitable analog for Web Weaver? It seems to me that this is exactly where all of these trains of thought are converging. And I think that’s why you’ve hit it right on the head with the idea of Web Weaver as a router and connector of sorts.

I’m going to let this marinade so I can contribute more to the discussion, but I just wanted to get that out there, because that’s instantly what came to mind when I read this. It’s an issue I’ve been turning over in my mind for some time as well…

Captain Lyre Calliope

You caught that. 😉


[…] Via captaincalliope.net Like this:LikeBe the first to like this post.   […]


Have you heard of the unhosted.org movement? We write unhosted web apps (i.e. not tied to any server-side code) like http://todomvc.unhosted.5apps.com we then use per-user remoteStorage as a ‘cloud sync for the web’.  http://www.w3.org/community/unhosted/wiki/RemoteStorage Maybe that’s what you are looking for? We’re also planning to make Dropbox, Locker project and OpenPhoto work as remoteStorage for unhosted web apps.

Captain Lyre Calliope

Ooohh!! I completely forgot ya’ll existed! 😀 ‘Unhosted apps’ are actually the ideal best practice implied by the web weaver concept.

Tying you guys into my narrative. I’m especially glad to see that you’re playing with the locker project. As far as I can tell, that’s the project that’s furthest along in running code AND longterm vision for what we need.

A key question I’d like answered: How will Unhosted apps play with Mozilla’s Open Apps architecture? https://www.mozilla.org/en-US/apps/

Augustin Bralley
Augustin Bralley

Sadly the Locker Project has been abandoned. https://github.com/LockerProject/Locker

Captain Lyre Calliope

Perhaps on the surface, but I’ve been tracking the team closely for awhile. The project evolved into https://singly.com/ and I’m pretty sure they still have their eyes on the prize.

Captain Lyre Calliope

“It is therefore my intent to make sure that this product class is fully
defined, specified (specificationified?) and in some form shipped by the
end of this year.”

Wow. That’s a little embarrassing. 😉