Category: design-and-ideas

The Tubeless Internet – Not bad, maybe too complex
David Reed, TTI/Vanguard Advisory Board

Starts with a quote from Ted Stevenson, which is complete wrong and is so the wrong way of thinking about the network.
Jon Steward effect = when he goes from channel to channel to get the same words from all the channels. Saying the same thing over and over again.
ARPANET was a packet net, the internet emerged at PARC, MIT, etc. At the time there was only the message switch model and phone company model. At the start no one knew what the internet was for so the creators kept their options open. Putting functions at the edge creates value in the form of options. Pervasive computing, now theres computers everywhere. Failed Pervasive computing – Universal Plug and Play, Bluetooth.
Eggtimer model, Intelligence was at the very edges. Reed talks about Phase. Traffic patterns – Rural and suburban traffic (gas), rush hour (liquid), traffic jam (solid), London (semi-conductor). Non-hierarchical, collective behaviours can and do work. Didn't understand the talk anymore, very complex. But David Reed and David Weinburger have been thinking about Beyond Net Netruality

Customized Mobile Virtual Networks – Good
Juha Christensen, CEO, Sonopia

Virtual Mobile operators. The network as a operating system. Its about the individuals not about the collective. Points out that there are credit cards which are interest and brand based. You can build price plans and do things like send out mass market messages to all subscribers. Affinity mobile gives you a revenue share of 5%. Could be used to provide cheaper or free phone calls. Ant and Heilo are other Affinity in America, while in the UK and Europe. Sonopia enterprise is around the corner.

Web 2.0 Architecture: Offline, Freedom, Open and Participation – Good
John Robb, Vice President, Technology Leadership, Zimbra/Yahoo

Things which are changing. Offline browsing, Software as a service, Freedom of Access and the web as a platform.
Offline browsing – why? Experience, lower tco, local backups, sharing, mashups. Offline solutions include Mozilla's Firefox 3.0, Google gears, Sun Java or Adobe Air. John make it clear that Zimbra will interop with many of the Yahoo services.

Mashup on the Fly – Excellent
David Prior, Chief Technologist, Research & Development, General Dynamics

Demystifying Mashups. There is nothing which the marketing department will understand. http://twickrpedia.com. Demostrated a couple of Mashups in less that 5mins.

Enterprise Knowledge Infrastructures, enabling collective knowledge – bad
Ross Button, Vice President, Technology Leadership, CGI

Connecting people and people to knowledge.

PlanetLab: Catalyzing Network Innovation – Good
Larry Peterson, Director, PlanetLab Consortium

Innovation can come from anywhere.
Most of the internet success is due to its support for at the edge development
There is a high barrier to entry for innovating through-out the net.
Planet Lab allows you to define what happens through-out the net, its all distributed virtualsation. So each project could use a slice of a bunch of servers (up to 600). 2500 users.

Therapy Development in a Networked World – Great
Sean Scott, President, ALS Therapy Development Institute

Build a project out of filemaker database to hunt down the correct drugs for his dying mother, simply an amazing story

Layer 8 Is More Interesting Than You Think – Great
Clay Shirky, Writer and Consultant

http://en.wikipedia.org/wiki/Layer_8 – Layer 8, where the users are.
Users will do what you'd never thought someone could do
Individual vs community
Powerlaw signature can be found in many social systems – Links per weblog, tag pairs on digg, edits per user on wikipedia.
Commons based peer production – pluto on wikipedia. adverage of 2.5 edits per user.
80/20 would kill wikipedia, you need the organic division of labor/cooperation without collaboration. If you want good quality work, you need limited collaboration. Jane Jacobs problem – many eyes on one problem, keeps people straight.
the Problem Wikipedia has is down to idenity, but if you make the wiki strict for secuirty you will lock out those who bring good stuff too.

Networks and Social Identity – Excellent
John Clippinger, Senior Fellow, Berkman Center for Internet & Society, Harvard Law School

The state of identity in social networks
Facebook, email address was your context (.edu), now its open and they've lost the context. In secondlife, people wanted to look like yourself with some slighly changes. Persistence is important. In linked in what does the numbers really mean when you can game the system easily? To build effective social networks you need to solve the identity. Identity is more that security, privacy, compliance, unfeathered rights, decisions. Social signaling is happening all the time, look at myspace. Implicit vs Explicit signals in profiles is a interesting idea. User control is centricity. Higgins = manages identity across multiple applications. If people can build relationships together based on trust and express what they want, you could have reverse auctions. Someone suggests that Facebook causes could be used to indicate your identity. Someone suggests that Facebook causes could be used to indicate your identity. Causes clouds which sits out side the closed networks are interesting. Some interesting URLs. http://eclipse.org/higgins and http://www.cloudtripper.com.

Comments [Comments]
Trackbacks [0]

So I'm going to attempt to Live blog the conference but mainly for myself

Leanord Kleinrock, TTI/Vanguard Advisory Board
The conference isn't about plumbing, its about the services and ecosystem which sits on top of it.

The Wealth of Networks – Excellent

Yochai Benkler, Co-director, Berkman Center for Internet & Society, Harvard Law School
Critical change is decentralisation of inputs and processes. Cites Participatory culture. Talks about the incumbents who want to fight over copyright, there last breath of air. Talks about the battle for Teraflops, SeTI@home beats IBM and NEC, radically decentralised capitalisation beats the centralised. This makes us all involved.Yochai cites the encyclopedia ecosystem, from books to cheap cds to free and wikipedia .Our authority models have been challenged. Brititannica and Nature are challenged. There
are new Opportunities. IBM's revenue jumped when they moved from old patents to there open linux stratrergy. Commons based production = production without exclusion, decentralises authority. Some guy, cuts Yochai mid flow and asks the question, what does this mean for big business? Yochai makes it clear that business needs to count all this into there business models, if they don't they will simply die. Someone else kicks in another question about how deep should people should be allowed in to collaborate. Someone
else asks a question about boundreys and hints at the trust. Whats to stop businesses to be open at the start then close everything up once they have an advantage? Yochai points out that GPL keeps people and businesses in check for the Free software world. Although Yochai isn't say everything should be Free and Open. But the legal framework keeps everyone on the right side of the line. Fear is the enemy of collaboration, fear wants you to know all the pieces and non open and collabroative. Fear puts up walls
and limits your scope. Yochai is hoping to write a book about fear soon as he has tons to say about it.
Someone asks the question if there is such as thing as too much collaboration, says Firefox stopped becoming creative and innovative due to the amount of people collaborating. However Yochai, points out that the plugin framework allows for creativity and innovation. You wouldn't see plugins like operator and seatbelt if you had a team tasked with building a browser.
Yochai says its not all about the money. If you go to dinner with friends and leave a tip for 100 pounds, you wouldn't improve your status with your friends. Not everyone is motovated by money all the time. People change their priorities a depending on many things, the system and markets need to reflect this. We need to learn how to manage and work with distributed social

Application Design, trust and the End-to-End arguments – Good
David Clark, Senior Research Scientist, MIT Laboratory for Computer Science
E2EA (the end to end arguement) – http://en.wikipedia.org/wiki/End-to-end_principle
Is Email end to end? Answer 1. No because its not wrapped around the outside. Answer 2. Yes End to end doesn't apply to apps. All of the servers are part of the rest. Answer 3. Yes we have used reliable application-aware parts to compensate for unreliable ones – we trust the servers, we trust mine and you trust yours. Answer 4. No we trust more that we need to. Question we should be asking is Email reliable? Reframing End to end in Trust instead of end of the transport mechanisms. Trust to Trust – Tim Moors.
This all fits in with net netrailty. In the Email example, the protocals allow functions to be assigned to different servers, can re-arraged to match trust assumption. So if your ISP blocks SMTP servers you have a broken trust model. Users trusts IM servers whether or not they trust it. We don't trust each other, hence in the real world we have deeds and escrow, registries. We can't really trust our own computer/servers no more due to bots, virus, etc. Trust to trust, third party servers and services are not
intrinsically in violation of End to end. How can people trust Google to look at your email along delivering your mail? What levels of trust will people put up with? Someone ask the question about a motorway of drivers and do people trust each other or is it something else?

Emerging Global Networks – Very good
Henry Tirri, Research Fellow, Nokia Research Center
Sensor planet. Wireless Grid, the worlds largest grid but how will people interact and work with the grid? A goal to have 50% access for the world by 2015. 500 Trillion text messages sent in 2004 alone. 1 in 5 are illterate.
Nokia have a app which can read a menu and translate it into english or other languages but what if your illterate? Concerns. Privacy and incentives, licences and non-licenced,centralised vs viral distributed, green vs survival, literacy (interaction methods), multi-lingual environoment, native innovation (home land innovation) vs innovation by others and transactions. America and Canada have the disadvantage of having minute based charging and charging both ends of the connections. Were still in the voice
to data transaction and that seems to be true for the rest of the world. 

The Sociology of Networks
– Very Good

Duncan Watts, Professor, Department of Sociology, Columbia University
Interactions between individuals is paramount, how they collect and aggregate (emergence) is the problem. Hence why Sociology is difficult. The Web has brought together a suite of interactions viable and measurable – email, online communities, mmorpgs, etc. Individual and group behavior can be observed over any period of time. Now there is tons of data when previously there was little. A science of networks is coming. Most data is being mined by corporations, not collected by research and the design
of sites have been largely evolutionary (mashing new features in). Recent experiments. Small-World Experiment, verified Milgram's Six Degrees result. Music-Lab Experiment, dumbness of crowds, markets don't necessarily. There is no evidence that there are leaders in a social network. Dominators, leaders, influences don't exist and if they did they would be replaced a few weeks later. After the fact, that person looks influencial. Before that they were just a person. Duncan talks about how the influencers
are just people who figure out where the crowd is going and jump in front at the right time. So in actual fact they spend most of their time working out what the crowd whats to hear.

Leveraging Networks
Michael Furdyk, Co-founder and Director of Technology, TakingITGlobal.org
Social network for young people who give them skills and resources to set themselves up for new challenges and opptunities. It acts as a platform for action. Inspire > inform > involved. Does Facebook and other social networks encorage you to meet people in other countries other cultures? A lot of social networks are banned in schools. Should do some work with geekdinners and girlgeekdinners?

The Emergence of Enterprise 2.0
Andrew McAfee, Associate Professor, Business Administration, Harvard Business School

Intranet need a kick up the ass. Enterprises 2.0 is about how its used not how its built or deployed. Andrew says IT is a differentiator. Underlying trends are social and network effects. Andrew adds Free platforms, lack of structure and tools that emerge.

Email = Channels which are point to point, invisible to others and be consulted.

Website = Platform which is universal, visible, consultable.

Blog = Free and Easy Platform

New apps have been getting out of the way of the user. Uses Wikipedia vs Newpedia as example. Then Delicious.

Tools should let structure emerge. Uses Google and tags clouds as example a example. Taking snapshots of tag clouds is interesting. Mentions how Flickr uses Clusters to apply structure.

Someone asks about complience. Transparancy helps.

Networks and the Semantic Web (Weaving a web of Data) – Very Good
Eric Miller, President, Zepheira

The art of data.

The web is unique objects (pages and links) data is hidden.

How did people work before the web? Next generation just can't imagine anything before.

Eric talks about how the cuting and pasting needs to stop and points at the TTI Vanguard site as example of rich data.

The web is evolving, its not versions

Talks about community driven ontologies using it the way they want it to be.

Shows off Piggy Bank to pull together his own list of hotels which made sense via a google map.

Talks about dbpedia.org – mines for details out of wikipedia

freebase.com and mashmaker.intel.com (intel's new pipes type system)

Shows a bunch of javascript tools to build new views on data

note to myself – BBC.co.uk as timeline

Web framework for resource description, talks about wrapping data sources like excel spreadsheets so they appear on the web.

Lego is interopable from 50 years ago, if Lego can do it, why can't we?

Eric answered my question about Microformats and he talked about GRDDL as a way to make use of microformats beyond their basic use.

Generally the questions surrounded the concept of a semantic web

Peer-Production of Rich Media (or What Happens when YouTube Meets Wikipedia) – Bad
Shay David, CTO Kaltura.com

Shows a porno on the wikipedia page for Pornography.

Rich media collaboration is harder that text based collaboration. Been little peer production of rich media / instead lots of aggregation.

Thinks we will see real time feeds of networked video and audio. Quotes Reed's law.

Shows off Kaltura.com

Shows off Kaltura.com in facebook

 

The Big Switch – Good

Nicholas Carr, Author, Does IT Matter? Information Technology and the Corrosion of Competitive Advantage

Radical centralist

Nicholas thinks we're moving to a world of centralisation/control

Preview of the next book – The Big Switch

Uses the Westinghouse engine station (AC electricity) as a example of centralisation with a service over a network. Local power went out of fashion and out of business.

Someone challenges Nick about the black outs on the electricity grids. Nick says utilitys always get treated badly. And now there cueing up to take a bite out of nick. Nick points out that the mainframes use to run at 90% of capitacy while the move to the server client model has given rise to high costs and higher wastes of processes. Someone suggests Virtualisation is fixing this problem and Nick actually agrees. Someone points out that work productivity went up due to IT investment along with the server client
model. Shows off the new Google server station in Columbia and says its the new Westinghouse engine station. Someone suggests some regulation doesn't allow there data be placed outside of their firewall, nick comes back and says regulation will catch up. Plus sharing data between companies and people will be lighting quick because they will be on highspeed inner network, instead of a companies T1 connection.
Nick certainly lived up to his rep and caused a stir. But now its time to fix my machine and go to dinner.

Comments [Comments]
Trackbacks [0]