Notes from ENMI09

I realize, I haven’t posted anything on my personal blog for a while, so I think it’s time I start doing it again (at least I’ve tried to post more often on WoT), and I’ll use this more as a personal dump of ideas, thoughts, and use WoT for more professional content. That means the content of this blog will not necessarily have the same editorial quality, but well, we’ll see.

I was in Paris a while ago to attend the “Entretiens du nouveau monde industriel“, focused on smart objects. Very interesting talks, I tried to take some notes (so might be filled with errors! beware!) that of my favorite speakers. Although – as always – the excellent Hubert Guillaud captured them much better than I could on the Internet Actu blog here, here, here, here, here, and here (all in french).

Jean-Louis Fréchin

Starts by asking who is interested in Design of Objects (half of the room raised their hands when asked “who is designer?”). Object emancipation, design/branding (”Baudrillard was right”). Big companies do not realize that worlds are changing and do not understand why their product do not work anymore.

Objects are older than IT. Internet is changing, there is more than just information: applications, services: data & processing. With cloud computing we can much better address huge amounts of data. Also objects are changing -> objects/things (choses objets). Things (as opposed to “objects”) can be seen, affects senses. Extension of the objects worlds, new networked objects, and a network of objects. “A marketing craze brings hinders a pragmatic reality.”

Why do that? The why is more important than how (design, usability). For example: prison-objects (jail-objects) iPhone/nespresso that creates dependability. Is it a matter of love of impulsions? A world where users only provide money is over, they provide feedback and it’s essential. “Yeah, Internet sucks because people can criticize our work.” Eh oui, welcome to the world.

Why do we ask things about usability in this area? People are trying to “scientifize” the problem, when it’s not the problem, we’re confusing innovation and research. In a machine to machine world, we’re eliminating the human, and it’s bound to fail.

Hardware is a commodity. In France (ndlr: I’d say in EU at larger), what we can do is to focus on the interface, not hardware (comes from china, etc). IoT is parallel to the internet of data, we can’t separate them. Interface – > objects. People almost to be tagged. Everything connected, tracked. These objects that talk about us, objects that love us.

Domesticated objects. We must learn to transform technologies into products. Stop with user/customer-oriented design. We’re humans above all. It’s also about lovable objects. Interface objects, products become interface. Interfaces also become objects. Solitary objects don’t exist anymore. How do we understand these objects. Do we want to see kwh from smart meters? Do we want something else? Social objects. It’s important: open objects, post-products. Hackable is good, but not enough. Yesterday: objects > commands > services > usages. Today: practices > services > programs > objects = ne-objects.

Nicolas Nova

Users, consumers, participants, contributors, different terms for the same thing, but people is the best word for Nicolas. Objects inscribe predefined things about user (Madeleine Akrich), and this predefines their future uses. For example, there are examples of subscribed information about people in objects (jo/josephine). Sally (PARC), the persona for future computer users, that can use future devices with little learning.

All these preconception were inscribed into objects, and they brought into. The “human processor” by Card and Moran. We represent psychology using the computer analogy, therefore how computers see us (Dan O’Sullivan).

5 archetypes, human representations. What kinds of misconceptions are there?

1. Omnipresence of screens. Designer thinks user wants to use screen to access data (wikireader), or photo frames. This makes sense. But it also goes too much in other things (his previous fridge). Also, it’s often messed up (trains, planes rebooting). A web of screens, it’s everywhere. It makes sense, but should it be really everywhere? Also should it be overcharged for AR, widgets (à la WoW)?

2. Liflogging. Fiat ecodrive, foursquares, life in a camebert diagram? is it is just about a little part of it, the raw data. It’s not just quantitative data. Even small things that happened only once might be very relevant.

3. Persuasive computing. Walk with me on DS, is an example of devices that tells you, attempts to change your behavior. It delegates to the devices.

4. Life simplifying things. No more “where are you”? Again, it’s a great idea. It’s relevant and useful when i do because I need. But not when it’s automatic. Bad automation. But when we ask people if they need automation, they say yes. But when it happens, then no. People want to keep the control.

5. Users to which we should not ask too much. Black box concept. Before, people could.

1. Swings is a little screen-less box that makes you play, engages you. Connects to the net.

2. objects a faire vivre: Autolizum you plug on your iphone or music, and this allows you to make it live. it has it’s own object based on the music you listen to. The object is not inert. you have to make it live

3. Tripit. Go to a place where there is that kind of music. It incites you to transform the music.

4. Customizable objects. Physically. not just like twitter or web. Olinda: Social radios. objects is also modular, little leds, radio, etc…

5. This offers multiple agencing possiblilties. Build objects that wrap around existing (agencés). Ipod access

6. four-squares you can lie about where you are.

Preconceptions are inscribed into objects. To design new objects: be conscious about potential preconceptions, inscribe into objects deep desires (communicate, curious, equilibrium of control on own and data). no paternalism, and teach people what to do.

Frédéric Kaplan

Metamorphoses des objets. How to transform something that is worth, into something that has a emotional value (put two coins and turns it into a money).Practical work. his daughter has asked how many things we have. good let’s count. But count what? everything? let’s count what counts. And different objects can have different value profiles (fantastic graphs about value of objects).
All other objects can be replaced, but not the ones that build value over time. Initially a note book is cheap, but value builds over time (you write it). He realized that he didn’t have electronic objects that really matters. He likes his computer, but doesn’t matter if he changes it. For example the first iMacs, with colors, plastic, organic. But then, actually not. People want sexy things, square, 0815 (the black macs). But then we throw away the carapaces. People on the streets. it’s not really the objects that matters, it’s the data they contain and convey. Millions of interfaces for a single machine.
Therefore the IoT word is dangerous, because it makes us believe these objects are little, autonomous and talk to each other. But in fact it becomes a single mega-object, which is really different from the classic vision of a reticular IoT. Therefore, PCs are maybe just an alea in the history of computing. If I break my PC, no big deal. It’s just an interface that i carry to interact with my data that is not stored on the machine but somewhere else.
Object-interfaces. New objects appear that have less and less an value on their own. He shows his latest object, a screen that moves covered with tissue (and not plastic because it’s not convenient for moving things). It has silent motors, it moves, but is not a little man/machine, etc. it’s a point. He has two microphones He has depth perceived camera. and pulsed IR receivers, it can see in 3D people. Can track people, recognize, and you can interact with it remotely, manually. This machine is not a thin client – it gets its data from the cloud, music, sounds, and so on. It’s intimate, the interface comes to you.

He shows another example: his new book has each page 3D-barcoded. This augments the book online, where people can annotate collectively on the digital page. Each page becomes place to start conversations. You can then see the life of the book in real-time, what people like, what they see think about this book. The object keeps living afterwards.
Then Frédéric showed his lamp. When you put the book under it, it’s automated – no more need to manually scan the bardcode. It sees the book it feels. Then, we could embedded that into light bulbs. Imagine, each lamp could summon an physical/virtual augmentation of physical spaces, new interfaces among other. When you leave the library, you can keep working.

The metamorphosis of objects separates historical and functional value of objects. Here it is a little different. objint tend to become valueless objects. They’ll end being products and become services. Their conception could take into account the whole lifecycle of the object. The fact it is enveloped by textile so you can change it, like a physical skin. it can be built with sustaibale. You can change it bit by bit.
Biographical data. We give away huge data, and in the end we get little in exchange. Is it worth a lot? I dunno but i get free things in exchange. am i being ripped off? new actors will emerge that will take care of all this data, biographical agents. Like a bank. The trust will be the major factor. We can decide what they’ll do with our biographic data, we chose how it’s used. We would see the emergence of a memory art, how to decide what we remember, what we want to forget.

Finally, above all, these objects are an invitation to reflect upon oneself. metamorphose-des-objects.com

2009

The clock is ticking. In couple of hours we’ll be in 2009. But another clock is also ticking. Life’s clock. 2008 has been a year filled with tons of things for me, successes, failures, meetings, revelations, fun times and less funny moments. Even some sad moments sometimes. And I’ve learned lots of things, on all levels, and in all I have a very positive feeling about it. But I have much more positive feelings about the years to come. My task in this post, will be to write down my resolutions for 2009 and share them with the world. These resolutions can be viewed as what I’ve learned so far in my almost 27 years of existence, what I got out of them, and what I must change in my life to make it better. I aim to improve as a person, and interact better with others. In a word get closer to nirvana, eternal happiness, übersanctity, whatever you wanna name it (the sense of life I would say).

_MG_3368.jpg

Monkeys in Beppu, Japan. cc Vlad Trifa

  1. Take care of my health shall be my main priority. Indeed, as can be seen in my Amazon wishlist I tend to love food a little too much. Way too much actually. But this has a counterpart: a growing belly. I don’t have eating disorders, but I came to realize that my alimentation is far from being equilibrated, and I generally eat too much. I want to learn to eat less, but better. Especially, I will stop eating processed foods, and crappy meats. I love meat, but I’d rather eat less of it but better one. I would rather eat once a week a good chicken (such as poulet de Bresse), rather than eating cheap chicken engineered on conveyor belts. I’ve learned to enjoy the purity of natural and healthy products (by being in the countryside in Romania), and I’m done putting crappy food in me. Don’t get me wrong; I’m not gonna become a bio-freak, vegetarian, of tofu hippie, but do my best to eat the most natural things out there, and stop eating stuff not made by me (or mom of course). I’ve been cooking a lot at home and enjoy it a lot, but I definitely think I could do much better when it comes to my alimentation. I’ve been reading a lot of great books and will keep learning things about food culture. Oh and also, I’m really looking forward to do lots of sports. Like swimming or running. I’ve never run in my life, but I’m sure I could do a routine out of it. Go run a 5 km at least 3 times a week, and 8 km on sunday morning (okay I stole this idea from Monocle). Now that I got an nike+ipod no more excuses. I’ll post a widget here so that I have more incentives to run often. Finally stop drinking beer and useless things like this (this doesn’t include the Youngs double chocolate), and only drink socially alcohol I really enjoy (fine liquors as scotch, cognac, and wine). Also, no more crappy juices (coke and the shit alike that) and less sweet things. I gave up smoking in August, so I’m fairly positive I can really improve my food habits in the near future.
  2. Learn to organize and prioritize things. Especially the latter. Undoubtedly, my weakest personality trait, is that I’m way too open and curious. I mean it’s pathologic how much I love new things, and how I always have tons of ideas that pop out of my head, and I start things. I have new passion daily, but never do much about them given that it’s simply not possible. I have a huge amount of motivation recently and I feel I have already done lots of good things with it, but I have lots of focus and energy leaks on useless things, that are not really good for me, and this must change. So I need to select a few topics on which I’ll focus most of my energy, and unfortunately cut my time with least significant things. Stop getting involved in side activities and projects that are not my core work & interest, unless I get paid reasonably well for not much of my time & energy. These passions are (in no particular order):

    1. Media and culture, which I consider being cinema & music & opera & news & blogs & books & theatre & museums & stuff like that. Read more literature and history books.
    2. Art history and practice (architecture, design, new media, and especially do more photography).
    3. Blogs and all other things related to my work & PhD topics (conferences & papers & basically doing my work). This includes a big project that will be launched soon.
    4. Japanese sword practice & study, and go back to practicing martial arts and sports.
    5. Food culture, history, politics, gardening.
  3. Learn, read, meet. The thing that I always enjoy the most is doing and learning new things. And even more I enjoy meeting new people all the time! I feel I really have lots of things I must improve when it comes to networking, such as keeping in touch and pinging all my friends and past colleagues. This is basically focus and improve everything related to my career.
  4. Blog. REALLY blog. I’ve been wanting to do this for a while, and I really notice that it can help me realize the two previous points. This is one of the things I will do in 2009. I wrote here hundreds of times that I will do it, and now I feel I really must and want to do it. Very soon, with a friend we’ll be launching a big blog (will be announced soon). I’ll blog a lot about my phd work, so I’m very positive that I’ll spend lots of good times blogging this year.
  5. Respect myself and others more. I sometimes don’t finish things as good as I shall. I am aware that it is a weak point of mine, and even though I consider myself trustable usually, sometimes I fail to deliver on my promises sufficiently, especially when it comes to time. I often say: I’ll do that for friday, when it fact I get it done for monday. Even though this happens only for time-insensitive tasks (with soft deadlines), I should do much better on this point and really make sure that from now on, I’ll always deliver everything I promise myself and to others, even if it’s not important. More generally work on my image and how people perceive me, and how this relates to who I am. This implies to stop dressing like a hippie. Yeah, I’m not a teenager anymore, so I need to dress and appear as a real man should, and stop being a bum.
  6. Obtain boat driving license. And car too. I might need that at some point. Especially the boat one. Will do both this summer.
  7. Stop reacting without thinking. I really need to learn to control my impulsions, as sometimes I react too emotively… and might regret it later. Must be due to the fact that I don’t do enough sport. And yoga. This is something I really need to change, and I know that going back to martial arts will help me significantly improve this area of my life.
  8. Spend more time with people I like and spend more time doing things I really enjoy. Learn to deal & collaborate more harmoniously with people I have lots of things to learn from. Stop spending time with übernegative people stop being in environments that drag me down and cut my motivation. Dead simple.
  9. Learn to become much more financially aware. I live within my means, but I am also excellent at spending money without leaving much aside for the winter. I really need to stop spending money randomly for things I don’t really need. Also, I need to set up plans to make a little extra money, put it aside or invest it for my future.
  10. Be happier, nicer to people, and learn to enjoy all the good moments life has to offer. You know, simply enjoy a walk in a forest with my camera and my dog. Well I don’t have a dog, but I have a camera. Keep collecting good jokes about French people, and write a book about my collection of jokes.

That’s it for a start. I feel good writing this down, and see it as a milestone in my life. I wrote it down and made it public with the hope that by having the world as a witness, I’ll really do everything I wrote here (and do what I say I want to do, c.f. point 5).

back this time???

Yeah, sure I’ve said millions of times I’ll go back to blogging, but never did. WHY?? I dunno… But his time I feel like things are different. I started much more actively working on my phd and I really plan to focus all my energy on that.

I’m in Romania now, so I have some time to redo my website better, and really focus on that. I plan to code a lot of things to automate my digital life, in particular how to sync my data with my website automatically, so that I limit the time I lose to maintain my site, and I could really focus on my work.

Besides, with my colleague Dominique, we’re preparing a new project that will be launched very soon. This will be another blogging project that will keep me busy, but I guess it will be much more motivating to work on it.

A little pix for fun.

Autofriedhof

just testing…. stuff…

Autofriedhof

cc Vlad Trifa

Royal Society Discussion: Day 2 – Afternoon Session

Professor Sir Ara Darzi, (not many notes… :)

Today, the complexity of healthcare has tremendous impact on people life’s. Technology is moving us from minimally invasive towards incision-less surgery.


_MG_6450

Sir Darzi

There is a need for a more dynamic and personalized monitoring of patients, in particular the one who suffer from a long disease. If you can monitor them at their home environments we can avoid huge costs due to hospitalization and also prevent earlier needs for doctors. Wearable body sensor networks, allows you to compute directly “on the patient” diagnosis. There are progresses in MEMS that allows to create high-resolution monitoring stations, but there is still a need to power these devices.

Usage of home monitoring during post-operatory recovery, pervasive pH sensors and glucose sensors. Sensor integration concept: combination of wearable and ambient sensor.

Professor Jonathan Zittrain, who gave a very interesting talk (I didn’t really follow everything as it was through a videoconference), unfortunately I took only a few notes. In any case, I think I’ll jump on his latest book “The Future of the Internet – And How to Stop It“, and advise you to do the same.

He started by showing us a set of crowd-sourcing solutions such as innocentive, livebox, or mechanical turk from Amazon, or the ESP game. The idea behind is that you can delegate boring or mechanic tasks to other parties, and humans will do whatever you want without knowing anything about why and for who, but just how much they get if they do the task.

There is something perverted behind that, as an idea of exploitation. “Could one really say that the one laptop per child project be a success if it is used by kids in Nigeria to solve porn website captchas all day long for some pennies?”

Microtransactions: the ability to cut task into small slices and be able to distribute them here and there that would require new methods for pricing…

Internet governance forum: taking a bunch of people to discuss on the future of the internet is not the best way to change things. Rather, it’s a small set of people who just start doing something (anything, but actually do it and stop just talking about it) that usually have the biggest impact.

Mr Adam Greenfield, gave a nice introduction to his vision depicted in his book Everyware: The Dawning Age of Ubiquitous Computing. There are many ubiquitous computing, not only one single type. Embedded in architectural place, wirelessly connected, imperceptible, multiple types of devices and services, post-GUI, deployed in everyday life, and vastly expanded “user” base. With the advent of the internet of things, a global “mnemotechnic” system, and the ways we interact with these techs are changing. Information processing (design) dissolves in behavior, Naoto Fukusawa (Muji’s idol designer). As an example of such dissolving behavior he discussed how girls “created” new physical interaction rituals to swipe their RFID Octopus card in HK.


_MG_6457

Adam Greenfield

A class of systems that tends to colonize everyday life, like the example he quotes “denki toire” (EDIT: I’m not sure about the correct name of this things, please send me correct name or link) that analyzes our poo and transmits its over the internet (hopefully encrypted) to our doctor.

Everyware is can be engaged even without active conscious decision to do so. Unknowingness (I didn’t know), unwillingness (you have to do even if you didn’t want to do so).

Differential permissionning without effective recourse: it’s the sensor in the floor (physical access schemes), problems of authority: “who to ask for permission”. In everywhere the presence of one component may trigger functionality in another. Also, the notion of property becomes fuzzier (we don’t really know what devices belongs to who), and through this fading notion of property, the locus of control is obscured. We don’t really know why the system doesn’t work (what has caused the malfunction), and it is even harder to find about who to complain to when something is not working. Worse the system doesn’t present to ourself in a way that makes it easy to determine who are the stakeholders.

Most of the people in ubicomp are just building little components without even knowing they are actually all contributing to this global network of interconnected devices. What if people were actually aware of that?

We have to take everyware seriously, and here are five rules Adam proposes to implement when we design everyware:

  1. Default to harmlessness: concepts of risk and safety are very different across cultures. For example in Japan, there are many warnings everywhere, because they attach a strong importance to safety. Even if ubicomp system present themselves as harmless and neutral, there might not be so…
  2. Be self-disclosing: should be transparent to the user, seamfulness vs. seamlessness. People must know what data is being collected, so Adam and his wife design a set of logos that can be used to show people that everyware devices are present and they collect data (the problem with these signs is that they didn’t say what kind of data is being collected).
  3. Be conservative of face: should not be embarrassing for people to use them. No society could be totally transparent, as it wouldn’t work. Social status is also essential and depends on the cultures (eg. Japan where language itself encodes social status). The systems must be able to find hierarchies (who is responsible for what)
  4. Be conservative of time: don’t bother the user with useless questions if he doesn’t want to use the technology at that time.
  5. Be deniable: it should be possible to volunountarily opt-out, and not use the technology at anytime.

Royal Society Discussion: Day 2 – Morning Session

Professor Anupam Joshi, “Trust, security, and privacy in Ubiquitous Computing”.

Security is essential because we interact with devices that are not in our home or office, but most of the work done in security for distributed system is not directly applicable to unicomp because these systems are open and dynamic. For that we can create policies and and sanctions for their violations. Autonomous entities need norms of behavior: declarative and dynamically adapt and explicitly manage trust. He offers the example of Asimov laws because they are abstract and understandable, and are build using unambiguous terms such as “must, can, oblige, refuse.” Entities will have multiple authorities so there’s a risk of being over-constrained, and then how to deal with failure. (Rei is an example of declarative policies declared with RDF/OWL-S.)

_MG_6407

Prof. Joshi

From 911 the moving from “need to know” to “need to share,” and that implies to explore the feasibility and desirability of this model, but also the risks and benefits associated with it. For having a shared policy, the parties must agree on the semantics of the language and of the domain ontology to avoid ambiguities and misunderstandings. Access control based on authentication simply won’t work in pervasive environments. Also, what is allowed in one context might be forbidden conditions.

First thing you can do is to secure the routing layer (cryptography), but you can also do it above, at the application level. You can build trust and reputation notions in ubiquitous applications based on what the other entities are saying, and by dynamically adapting your trust levels in peers depending on whether they give you appropriate answers or not.

Professor Mogens Nielsen, “Trust models in ubiquitous computing.”

He discussed about the role of trust and gave a short introduction about how trust systems can be modeled in computational systems. Interesting talk and topic, unfortunately I did not take many notes. In ubicomp are applications where decisions are made between two actors information providers and requesters. The problem is that these decision are made by autonomous agents and not humans, and often based on incomplete information, therefore new security principles are required. Two main classes of computational trust system exist: credential based (Keynote system, delegation logic) and reputation based (EigenTrust, beta reputation). The abstraction in trust systems is the concept of reputation and you represent trust by using a mathematical notation of the quality of interaction between peers. However, these computational systems need formal models that allow to ask (and hopefully answer) questions related to computational robustness and performance.

Samuel Karlin “The purpose of models is not to fit the data, but to sharpen the question”.

Professor Gary Marsden, “Mobile interaction design for developing nations”, who recently wrote the amazing book “Mobile Interaction Design“.

There aren’t many Internet users in Africa nowadays, but the mobile users are growing amazingly and the market is huge up to the point that many people will use their whole cash for mobiles (they would even skip meals just to save some extra money for that). The PC has been leapfrogged: macro-ubiquitous technology. HCI teaches humility and you need to deploy system, observe what people do, and then evaluate that.

Price sensitivity: there are 3 operators (A is the cheapest for local calls, B is the cheapest to receive calls, and C has the biggest coverage), so when people receive a local call, they change the SIM card from B to A to do a call. We don’t do that here, because the small price difference are not worth the pain of switching sim cards for each operation.

Camera phones have a huge impact, people are keeping a life diary on their mobile phones. Storytelling, sharing stories is a big things in Africa and they developed a mobile phone software for sharing (by broadcasting) pictures with the other people in neighborhood – a very interesting way to use technology as a support tool for the perpetuation of traditions.

MXit = like IRC but on mobile phones, which is so much cheaper than SMS, so many people use it. Kids even starting using it to collaboratively do their homework. It’s ugly, nobody in Europe would use that, but as there is no alternative in Africa, it is very successful over there.

Two problems are important: we cannot learn the culture in just a few months, and also there are not enough of people doing this kind of work. To develop ubicomp, we need to find the “bridging persons” who can understand both the local African context and out technology.

We don’t help nurses but doctors because, and we empower them to create and customize and develop their own version, because they know the problems and the local context, and they also know how do these people work.

You learn a lot about computer science by working there (ndlr. in Africa), because the solution is not use more technology or some other “common” shortcut/trick, because the constraints are totally different over there. Mobile phones are an appropriate technology because it has value over there, and we all need to find ways to leverage that.

Royal Society Discussion: Day 1 – Afternoon Session

In the afternoon, professor Jeannette Wing presented her talk entitled “Computational thinking and thinking about computation” (slides). She wrote a short essay “J.M. Wing, “Computational Thinking, CACM Viewpoint, March 2006, pp. 33-35 (paper) and introduced her big vision for our future.

_MG_6382

Prof. Wing

Computational thinking (CT) should become a fundamental skill for everybody in the world. CT is about taking a computer science approach to solve problems, and it is not just algorithmic thinking, but also engineering thinking, AI thinking. Abstractions are our mental tools, and it is important to choose the right abstractions, operate simultaneous at multiple layers of abstractions, and define the relationships between the different layers.

Examples: taking your kids to soccer, gymnastics, and swim practice (traveling salesman problem), cooking a gourmet meal (parallel processing), cleaning out your garage (data management).

Research implications: how this has already influenced other scientific fields and beyond? Examples: caluculus for biology to model behavior of cells, brain science, chemistry, geology, astronomy, maths, engineering, etc… in economists, game theoretic notions for pricing (economics), all phd go to wall street instead of becoming professors, social science to explain social networks. Visualization enables new toolsÉ sports (ai tools for view tapes), arts,

What concepts can children learn when? What should we teach when? How to best integrate the computer with learning and teaching the concepts? There are different drivers for that.
First technological trends as for example alternative computing substrates such as biocomputing, quantum, nanocomputing, bio-nano-quantum computing, or new types of devices: mobiles, robots for disabled, cars are networked (you drive a computer)! Virtual worlds (NASA even has 2nd life meetings), brainy machines, blue brain project, web 3.0

Then societal drivers, which are related to our expectations: everything, everywhere, anytime, for ever, for everybody, different layers of things… hmmm.

There are still five deep questions in computing

  • P=NP? Could drive to develop new mathematical techniques.
  • What is computable? Power and limits of computation? What is a computer? A comp is not just a PC anymore: the net, server farms. Consider a machine and a computer computing together (TOGETHER), complement each other? Combine both powers.
  • What is intelligence? (and consciousness)?
  • What is information? Not just 0 and 1’s…
  • (How) can we build complex systems simply? Is there a complexity theory for systems as there is for computing? Meaning of sys cplx that spans theory and practice. Do these systems need to be so complex???

So spread the word, make it a common place, explain it everyone!

Professor Tom Henzinger, “Challenges in embedded systems design“.

There exist two cultures of models: engineering (differential equations, linear algebra, and probability) and computer science (logic, discrete structures, automata theory). All systems we build are extremely complex, and 1/3 of the boeing development cost was for integration and validation cost. It is impossible to try every possible line of code and see where are bugs.

_MG_6385

Prof. Henzinger

What went wrong? Engineering theories of estimation/ robustness, CS theory of correctness. build reliable systems. Temptation: programs are mathematical objects, but we should avoid the boolean (true/false) vision of the correctness of programs.

Cyberphysical systems 3 type of constraints for embedded systems.
Execution, Reaction, and Computation. We left the physicality in CS, we remove time and resource constraints. Software is the most costly and least flex part of embedded systems. We should take back the embedded systems design and not let it fully to EE guys. Also, we need new formal foundation for computational, which remarry compuations and physicality. Two main aspects: performance (what is computed) and robustness (change of resources, failures, attacks).

Subchallenge 1: integrate analytical and operational modeling
We need both and need engineers that understand both worlds

Subchallenge 2: build predictable system: predict = deterministic (internal we don’t care), but behavior needs to include nonfunctional aspects such as time

Non-determinism is central to complexity theory, to abstractions and concurrency. There is useful non-determinism, and the one we don’t care and that we don’t see. As long as the result is correct we don’t care how we got it.

Can we build languages that treat time in the same way as high level programming languages treat memory? The compiler (or a runtime) can choose when to execute tasks in order to be optimal. (SET – execution time model).

Subchallenge 3: build robust systems, more continuous systems: check values in function of time. Less continuous: read sensor if x then, else…

We need high-level programming models for building deterministic systems from non-deterministic parts. We need system preference metrics for building continuous systems from non-continuous parts.

Royal Society Discussion: Day 1 – Morning Session

This is one long post here…

The day started with Professor Gaetano Borriello, from University of Washington and his talk entitled “Invisible Computing.”

Prof. Bariello started with the classical Weiser’s quote “the most remarkable technologies are those that disappear”, but remarked that currently computers are not really disappearing. He stressed an important aspect in the metaphor of the disappearing computer: it is not about the “visually” disappearing, but “cognitively” (you can still see the computer and know it works fine, but his role is to assist you and not to frustrate you). Also, he presented the Labscape project, which is a system for automatic work capture for cell biologists. In science, methodological procedures and research results are hard to annotate, to index, and to share, as there is no “standard language” to formally describe them. Automatically capturing the methodology and organizing the data of experiments, search all that, would be extremely helpful to share data and eventually automatically write documentation about the experiment. Labscape is here to assist the researcher in doing science and not spending time on meta-science (sensors on what the guy is doing, how much liquid is poured where, etc). It’s visually visible, but not cognitively, however visibility is still necessary (need to know it’s there and working).

Also, he presented the RFID ecosystem, which is an large scale experiment of the pervasive use of tagging (people and objects are both chipped). The goal was to create a microcosm of a world saturated with identifiable objects, 160 tag readers, and thousands of tags (every interaction is recorded, all read events are stored in a DB specially made for users, with probabilistic event detection, particle filters for location estimation). Twitter was also used to track last location and movements of colleagues. They also setup digital diaries for when/who/where/how long you meet. Challenges: noisy sensors (water in the body affects, data management, and security). Limited scenarios only for now in the lab, not real deployments.

He also discussed the panopticon – asymmetric visibility (guard tower in the middle of prison and everybody can be seen). Institutional privacy: what is socially appropriate. Physical access control is used, where only viewable things that anybody else could see anyway can be accessed (from your physical presence) – no superman Xray vision is allowed by the system. Tag IDs are sent in clear, provenance trail (they can revoke/delete anything anytime, incl. the composite events), transparent queries. He also discussed about Joshua Smith from Intel who is working on adding accelerometer and light sensors to passive tags, which can sense without any external power (which would be cool to know if a chicken has been exposed to dangerously high temperatures).

Philosophy of both projects: future world technologies will really disappear, process of gathering many bits of created data/interactions. Not physical invisibility, just cognitive! For now, benefits are only small tricks and hacks that can significantly change little things in our lives, but it is still too early for seeing large benefits.

Two cities in China are doing exactly that: monitoring people (he didn’t mention which city). Also can be used to detect anomalies in mobility patterns (2 people not moving for long, etc…).

Afterwards, Prof Andy Hopper FRS from Cambridge University presented his talk.

How can developing countries become successful without using too much resources (as we did), computer tool for enabling improvement in developing countries.

Optimal digital infrastructures

Provisioning appropriate availability: redundancy of data centers doubles costs, so develop new tools to reduce the redundancy. E efficient computing: adaptive datacenters. Scale energy use with useful work done at all levels. Develop principles: switch off if not use, don’t send data if not wanted, know where traffic is coming from, use tech that linearly scale energy (PROPORTIANAL consumption to task). Servers consume 50% when idle, and as server are sometimes idle for a long time it’s quite bad (Energy balancing in cloud computing – start/stop more or less servers based on predicted load, XEN virtualization, jobs are moved around in 250ms, non-interactive jobs are delay-tolerant). Virtual battery something spatial about computing tasks, can be virtually done anywhere in the world, where energy is available and cheaper (because it’s cheaper to “ship” the task anywhere else) (quote article for supercomputers in wired) – > what granularity should be the jobs, what to ship? Data, program, or both? I like the idea of “shipping/outsourcing” computing. Where to put the farms? Look at wind maps to use energy that would be otherwise lost. Global goal: energy proportional computing/communications at all levels. BAN THE WORKSTATION, because it’s very inappropriate.


_MG_6364

Prof. Hopper.

Sense and optimize

Sensor based digital model of the world: Googling earth, space-time. Sensing: already many sensors, and people publish/share their data (incentives to do so). Storing: create a global repository, data & comp models. Indexing. Sensing indoors location (using beacons and fixed infrastructure) or outdoors with cars & mobile phone. Mapping 3D cell reception. Human sensing. Reward for content creation enticing and wealth creating for developing world?

Thermal maps, personal energy meters. Collect info about individual consumption. Dilemmas: who to trust, of values, governance… Also, which surveillance scenario will prevail?

Predict and react: prediction of travel times and routing. Why a particular route is chosen? Gives several choices (www.camvit.com).

Global computing standards for digital alternatives to physical activities: growing tendency to move bits rather than people and product. Do webcams make us travel less, or more? Can we create virtual worlds where we can conduct our lives?

Control through price, where you empower people, and they can make choices to spare money. How much computation is really “necessary”? Evidence: the more computing is there, the more energy is used. Having more computers is really actually going to save energy?

Professor Margaret Martonosi, from “Zebranet and beyond: exploiting the unique characteristics of mobile sparse networks“.

All computing is mobile, distributed, as lots of devices became computers. So we have now a heterogeneous network of different computing platforms. How to program/manage them?


_MG_6368

Prof. Martonosi.

She presented the Zebranet project, that focuses on analyzing the interactions within species and between species, and to measure the impact of human development on animal behavior. In Kenya, there is no communication infrastructure. Triangulation with VHF signals to locate animals is good but not really robust. Also, you have to be out actively monitoring. On the other hand GPS devices are totally restrained (either local storing of data or sent, neither of which is optimal). So what to do when no good infrastructure is available? Mesh p2p data forwarding!

Mobile sensor network, not just a fixed mesh, but by encounters between nodes, for sparse networks. This requires a very different protocol/system design, including custom hardware and software that allows reprogramming of nodes at runtime through viral propagation to the neighbors. Data compression (very aggressive) and look at tradeoff between compression and communication costs, data abstraction layers for fast communication & queries.

Efficient non-GPS localization: why does it matter? Localization is essential for many applications, eg. social networks to locate your friends, geographical routing for optimized communication, spatially aware computation, etc. GPS is expensive and power hungry, so she proposed a collaborative localization algorithm called LOCALE (ISPN 08 Zhang and Martonosi) which emphasizes low power, accurate low device cost, low infrastructure, etc. Also, how can the aggregation of multiple position estimates increase the confidence in localization results, tracking position between encounters.

Each node keeps an idea of where it is with a cloud of confidence. If devices have high confidence, then they can reduce the cloud size. Use fixed beacons with GPS and accurate location, and they can share their info with neighbors. Collaboration is a powerful mechanism for improving accuracy in localization.

Then what if we replace the zebras with people? Store and forward to replace phys, wired infrastructure. SARANA: system to support collaborative, low-infrastructure computation

SA – Spatially Aware where are services located? New languages are being developed for location aware computing (SpatialViews) The language allow the programmer to create quality assessments as well, RA – Resource Aware, NA – Network Aware.

Sparse mobile networks are here and growing, time sync is still a big issue when you don’t have cheap access to a global clock (as the GPS would provide).

Finally, Professor Timothy Roscoe, from ETH Zurich closed the morning session with his talk entitled “Network architecture for ubiquitous computing”.


_MG_6374

Prof. Roscoe

Obstacles to progress: past – what the internet is, present – internet problems, roadblocks for progress and a research agenda, future – new architecture, and implications for ubicomp.

Very few ubicomp, long-lived, or large deployments have been made, and only few of them addressed real-world real challenges such as heterogeneity, scalability, and evolvability. Besides it is very difficult to evaluate these systems. Unfortunately, there’s not much motivation for these issues in Ubicomp.

IP is our “universal” protocol, end-2-end arguments, and edge vs core. There are things it supports but it shouldn’t (spam), and the other way around. The net was not designed with a security or resource control of mind for example, and there are many other things the net doesn’t also have, and this should be a research opportunity! But how to know if your idea works, because usually the problem is bigger than your lab network? How can anyone else reuse your idea? According to a report in 2001 (US NRC), the Internet is too hard to change and too important to change (“ossification” is the elegant term he used for that) because of the impact a failure would have. A suggestion was to use overlay nets, but this leads to other problems: how to deploy, evaluate, and access overlay networks?

Large scale internet services: akamai, google, with many geographical locations. P2P nets: millions of nodes, increased variety of applications. Lots of simulations, but who believes them? And then you also need to validate against something. Emulation on large clusters has also be done, but would you really meet the real-world problems?

Lots of small scale deployments. A meeting for building a collective platform (march 2002 – paper came out blueprint for introducing disruptive technology into the internet), remarkable consensus was reached, community-built platform, address both issues: enable a real wide-area distributed systems tesbed, and deploy new nets overlays with real users. Key ideas: slices! Like Wuala, you give some, get some. PlanetLab had huge impact and now is made of over 800 machines throughout the world! It even changed the publishing climate by raising the bar of validation, and no more excuses for lack of deployments. You can virtualize links as well as servers! GENI.net large-scale project – for research for the future of the internet. FIRE FP7 program in EU. Now the problem: how do you share resources between multiple users, with resource guarantees, over short timescale, and securely?

He suggests that it is time to rethink two basic principles of the internet: core vs. edge applications increasingly resemble overlays and links are more important than the end-2-end (also often there are more than one end…). Is really a new network that we need? Or rather a different approach when using the same, existing infrastructure?

Key ideas

- Low-level resource provisioning VM, links, radios. etc

- Overlay networks that can adapt in an application-specific manner

- Rich resource descriptions knowledge representation.

A much more “re-configurable” infrastructure is therefore required. Mixture of net technologies! Not just a homogeneous set of techs, but mix of mobile low power and high-end computers.

Intervention from Jeanette Wing: First decide the agenda and research topics, then define the experiments to validate models, and that will finally drive the infrastructure choice, rather than the other way around. It’s not just about the internet, but about a much wider substrate that spans different physical layers, “internet is just an artefact”.

Prof. Ozalp Babaoglu: self-* properties of complex systems

Professor Ozalp Babaoglu: “self-* properties of complex systems“.

Current information systems have reached a level of complexity that makes it very complicated to manage and deploy distributed software using traditional techniques. Nowadays, most costs are related to maintain and fix existing systems rather than buying new equipment. To lower the total cost of ownerships of IT systems, humans should be removed from most operations that could be automated, therefore there is huge need for such system to posses self-{configuring, optimizing, healing, protecting, managing, etc…} properties (denoted self-*).

_MG_6401

Prof. Babaoglu

In the blueprint for autonomic computing of IBM (quote), it is suggested that self-* would require an intelligent control loop. However, an alternative would be what he calls “grassroots” approach, which consist to interconnect many agents (rather than having a central intelligent control loop) where needed functionality would be an emergent property of the system. Emergence is unavoidable and is found everywhere: power grids, telephone switching nets, retail supply chains. The whole point is to control the interactions between agents such that the global behaviors which emerge are the ones that we want. Actions based on local information (swarm intelligence), with small number of components (I guess he means small as in relation to all components in the system)?

Prof. Babaoglu proposed the gossip method where interactions only happen with known peers, and all peers act identically. Peer sampling service of nodes random from the whole population. Overlay networks that satisfies desired topological properties: scalable, robust, decentralized, self-organized cell formation, some cells like or dislike each other.

He also discussed about the phenomenon of synchronization (eg. heart beats) and classical modeling using coupled oscillators, and how these can be modeled using the gossip framework – state (phi of the oscillator) that is communicated to a small set of neighbors. Finally, he proposed a model for formation creation (devices that self-arrange in a 2D ring formation) in a classic ubicomp scenario (does such a thing exist?), basically an ad-hoc network of mobile devices formed based on physical proximity and with multicast communication abilities.

My personal comments

This was a nice summary of the basics of swarm intelligence, with some insights on coupled oscillating systems that synchronize. Having worked for a while in both these fields, I would have been very happy to see more applicable description of these systems in the particular context of physical computing and actual ubicomp scenarios. In particular, I would have been glad to know more about (if any) attempts to formalize such complex systems (especially useful for verification and evaluation) and how to prove that a given (physical) system can behave deterministically in a certain way in a given amount of time (thus could be actually applied to concrete industrial settings having hard real-time constraints).

Also, I would be curious to know what are the global effects of using physical devices as opposed to simulated ones – can the emergent global behavior still be guaranteed with noisy sensors and communication channels, failing hardware, and so on? How would the wide range of possible failures be incorporated in more formal model for complex systems? These are important topics that are worth studying as I am sure that even if the system as a whole is robust to individual nodes “die”, I doubt it is also robust when the devices behave inappropriately.

Finally, as a question in the public mentioned, ubicomp scenarios are heterogeneous by “nature”, yet the gossip model presented assumes that all devices are identical and have the same properties/capabilities, which clearly limits the applicability of these models outside computer simulations. This is a very important question to my current research, how to efficiently and smoothly use a dynamic network of devices that can have very different capabilities (be it computational, sensing, communication, you name it)? I really look forward for new work in this direction.

Back from london

I just got back from London today, and… what a city! Hugely diverse, an amazing mash-up of cultures, fun, and food packed in a funky city. I took the time to visit (of course) the Tate Modern (an totally amazing building by the way), and a cool exhibition of Banksy in there. It’s crazy to see how good his marketing scheme is, but rest assured he certainly is sipping Pina Coladas somewhere in the Bahamas. It’s not that I think he wouldn’t deserve it, because I love his stuff and I definitely think he’s an amazing artist (and certainly one of the most sharpened one in today’s art world), but when I see the prices of his stuff it makes me a little dizzy…

_MG_6379

The Royal Society Building in London

Other than that, my impression of London it’s that it is a very “bricky” city. I mean everywhere you see bricks… I thought that the brick houses were just a cliché, but it’s totally true.


_MG_6270

Bricks, bricks, and other bricks..

I loved walking around randomly in soho streets, crazy mix of restaurant and bars and clubs, and especially LOTS of TOTALLY amazing BILBs (Bookstores I’d Like to Buy) filled with freakish design and artsy books which usually hurts very bad my credit card.

_MG_6331

Bars in Soho

Finally, I attended this amazing meeting on the last two days (see previous post), and took lots of notes that I’ll post here very soon, just once I’ll make them human-readable. The discussion was quite interesting by the diversity of the speakers and their thoughts on ubiquitous computing. There weren’t very “new” things but more perspectives from “non-ubicomp” people. I’ll put very soon the different talks online, and as some of the talks were much more interesting to me and related to my work, I’ll put them in independent posts with my thoughts on them, while the other will be just a bunch of notes I took and I don’t feel like spending hours just to make a post out of each (not that they don’t deserve, but it’s just that I have several other things in my mind for now).

So stay tuned!

Off to london

I’m off for London tomorrow as there is this this awesome meeting next week entitled “From computers to ubiquitous computing, by 2020“, organized by the Royal Society. I’ll try to take plenty of notes, but it’ll be hard as it’s packed with great people. Like I’m gonna feel quite insignificant with such a line-up:

Professor Ozalp Babaoglu, Professor Gaetano Borriello, Professor Jon Crowcroft, Professor Sir Ara Darzi, Professor Rocco de Nicola, Mr Adam Greenfield, Professor Wendy Hall, Professor Tom Henzinger, Professor Andy Hopper FRS, Professor Anupam Joshi, Professor Marta Kwiatkowska, Professor Gary Marsden, Professor Margaret Martonosi, Professor Robin Milner FRS, Professor Morgens Nielsen, Professor Ronald Rivest, Professor Tom Rodden, Professor Timothy Roscoe, Professor Vladimiro Sassone, Professor Morris Sloman, Professor Jeannette Wing and Professor Jonathan Zittrain.

I’ll try to liveblog the conference and see what happens, never done that before, but a good excuse to blog more often ;)

Seriously! Ping me if anyone is in london so that we could catch up, have some beers, food, and rebuild the world ;)

This blog is hosted by Dreamhost, and is proudly powered by made with Wordpress.
© 2008 Vlad Trifa