- Home
- LinkedIn
- GitHub
- Tumblr
- Twitter
- Available April 1st 2012!
- Ask me anything
Candide Kemmler
I am a freelance web expert.
I do software conception and development: backend (J2EE) and frontend (I love Javascript), for my customers both in Belgium and abroad.
-
Info
- SummaryI am a freelance web expert. I do software conception and development: backend (J2EE) and frontend (I love Javascript), for my customers both in Belgium and abroad. I work with a growing network of IT professionals, UI/UX specialists and researchers.
I have worked on big systems with big development teams (eBay, Proximus, Fortis, European Commission) and collaborated on smaller projects with boutique web shops. I have recently launched fluxtream.com, a personal data aggregator. My philosophy is to give priority to the product, where the user experience is king.Specialties: End-to-end development of great user experiencesExperience- Apr 2010 - PresentCEO / Fluxtream.comWe are creating a web and mobile platform that lets users experience a complete visual trove of their digital footprint. Our app focuses on providing users with a compelling and fun way of viewing, comparing and correlating their personal data streams: health data, location, social behavior, media usage, etc.
- Sept 2010 - Mar 2012Senior Ajax Architect / European CommissionWorking on the efp7 programme, helping to improve the Participant Portal which consolidates all public interactions with the EC's online services.
- May 2010 - Jun 2010Architect / UCM TechnologiesResponsibilities document best practices inside the unit understand and enforce these best practices in teams inside the unit
- May 2009 - Mar 2010Architect / Sefol UG (Haftungsbeschränkt)Implementation of social networking enhancements to a University Library Catalogue search engine for HedTek, Ltd ( spin-off University of Manchester). Technologies : PHP, JQuery, Apache SOLR, cobertura, cruise control Responsibilities demonstrate the feasibility of a social networking-enhanced approach to university library search that is scalable to the entire network of UK's university libraries; strong emphasis on tests (both on the server and the client -side), test coverage (i.e. cobertura) and continuous integration (i.e. cruise control).
- Mar 2009 - Jun 2009Researcher / Univ. of OttawaResearch work for Prof. Pierre Lévy Responsibilities define an interactive widget for navigating an IEML dictionary.The results of this research are accessible here develop a basic IEML parser targeting javascript by leveraging Antlr, a parser-generation tool.
- Aug 2008 - Feb 2009Developer / eBay FranceTechnologies: J2EE, eBox, prototype, drools rules engine, Virtuoz intelligent agent technology Responsibilities enhance the OCRA (Optimized Customer Resolution Algorithm) workflow, which was developed to help end users to find answers to their questions before reaching out to the Customer Service via email user Interface enhancements (AJAX) improvements in the logging aspects of the application in order to generate accurate metrics
- May 2008 - Aug 2008Architect / ActiveIDEA SAdevelopment of a generator for custom web admin interfaces Technologies: AndroMDA, J2EE, Oracle Responsibilities develop a custom AndroMDA cartridge for the generation of customizable web administration interfaces to a big and complex Oracle database knowledge transfer for follow-up team, documentation
- Jan 2008 - Apr 2008Architect / Sefol UG (Haftungsbeschränkt)Implementation of a social collaboration framework in Flex, Hibernate and Red5 for HedTek Ltd, spin-off University of Manchester. Technologies: Flex, JPA (Hibernate), Red5, Spring Security Responsibilities refactor a mind-mapping application to allow students to collaborate on the models of their group work assignments integrate Spring Security with an existing social substrate in PHP and implement read/write authorization on service methods implement a persistence engine for collaboratively edited resources; the backend for the flex application is written in Red5.
- 2004 - 2007Architect, Lead Developer / ObjectiveGUI SPRLdevelopment of major upgrades to both the server-side and client-side aspects of ObjectiveGUI WebMS. Technologies : XUL, J2EE, XSLT, Javascript,... Responsibilities design & development test design and implementation knowledge transfer for follow-up developer Since january 2006 I have been responsible for the maintenance and quality assurance of WebMS' entire code base. Technologies : AJAX, J2EE, XSLT, JUnit,... Responsibilities Design & development. Test design and implementation.
- Jan 1999 - Jan 2004Owner / Fractals SPRLdevelopment of a website devoted to Belgian politics. Development of a CMS focusing on video metadata editing development of several websites dedicated to sports: soccer, formula-1, basket-ball fund raising for the creation of a web portal centered on a cartographic metaphor and development of a Flash-based portal consisting of a cultural agenda, "golden pages" and PIM based on a cartographic metaphor. development of web cartography tools targeted at the Flash player and exploiting geographic coordinate data from TeleAtlas.
- Jan 1996 - Jan 1997Developer / Emakina SAediting work on the "Magritte" CD-ROM: localisation in english animation for the "Galaxians" CD-ROM development of an interactive website on the Horta Museum (sponsored by CBR Group)
Education-
2010 - 2010Founders InstituteGraduate in Business
-
1991 - 1994Université libre de BruxellesBS in Philosophy
- Lycéee Dachsbeck
Additional Information-
Websites:
-
Interests:User Experience, Web Development, Entrepreneurship
-
Repositories
- Command-Line Interface to the Fluxtream Backend3 forks / 3 watchers / Pushed May 10, 12
- a micro-project just to check that we are able to retrieve photos+videos with a minimum set of libraries. The purpose of this project is to debug problems with the picasa connector in fluxtream1 fork / 1 watcher / Pushed Mar 13, 12
- Command-Line Interface to the Fluxtream Backend3 forks / 3 watchers / Pushed May 10, 12
- a micro-project just to check that we are able to retrieve photos+videos with a minimum set of libraries. The purpose of this project is to debug problems with the picasa connector in fluxtream1 fork / 1 watcher / Pushed Mar 13, 12
-
Posts
- How to survive your first year as a Quantified Self entrepreneur
I have graduated from the Founder Institute a little more than a year ago. Exactly why I decided to undergo this torture is still a mystery to me, but it’s a mystery with a happy ending. I consider myself to be an idealist inventor. For as long as I can remember, I have dreamed up systems. I have imagined software that’s way too large, way too complex for one man alone to implement, much less a man who needs to earn his daily pittance to sustain a living. I had my ass kicked. A year after graduation, I did launch fluxtream.com. First time in my life, I experienced having actual users.
That was great (and it still is). Only, if I’m looking closer, I can see that I am still struggling with the same demons. As it stands today, Fluxtream.com only scratches the surface of my original
plans for world dominationideas. I did (almost) everything on my own, including some (totally pathetic) attempts at pitching my idea to investors. I lost money. As an entrepreneur, to a certain degree, I made a fool of myself.Clearly, by today’s standards, I do not fit into the Entrepreneur’s archetype: I’m a terrible pitcher; I hack most of the time and I don’t earn money: I spend it. Strangely though, last year doesn’t feel like failure, not at all. Like one of my friends likes to say, it’s adequacy we need to aim for, not conformity. For an idealist inventor like me, pioneering a totally new space is totally adequate. Contributing to the creation of a new software ecosystem even more so.
Fluxtream is now open source software. We also just published a first draft export API. Fluxtream provides a fantastic work bench to experiment with personal data and imagine new quantified-self experiences. It is simple to operate within a user community and works with commonplace technologies like MySQL, jQuery and JSP. The next version of the frontend will offer a Twitter Bootstrap-based framework that makes it easy to create new widgets, and it works on tablets and laptops or desktops alike. Today, Fluxtream is already cross-pollinating with another great quantified self open source project called BodyTrack (video).
All this stuff is pretty geeky, right? But, hear me out, on the Quantified Self planet, it’s perfectly adequate to be a geek.
I am confident that many fascinating business models will be uncovered in this space. My conviction today is that we need to work hard on a thriving personal data ecosystem before that happens.
It’s too easy to get lured by the one-size-fits-all standard of the entrepreneur role-model. Innovation goes way beyond business stereotypes.
- The secret technology for absolute mind control
I have been following Prof. Pierre Lévy’s work for about 20 years now and I just ordered his latest book: The Semantic Sphere: Computation, Cognition and Information Economy; I can’t wait to have it in my hands. Pierre is the Chair of Research for Collective Intelligence at the University of Ottawa. He wrote numerous books on computers and their impact on human cognition and culture. As an (Human) Intelligence Technology, Prof. Levy places the invention of the computer in a long lineage of other such technologies, starting with hieroglyphs, then the alphabet, much later the movable type. For Prof Levy, while computer theory will likely not change very much – computers will only get much more powerful – there is however a huge theoretical work to be done on the next evolution of Intelligence Technologies that they will finally enable. His last book, on which he has been working relentlessly for the best of 10 years, is an attempt to sketch the broad strokes of this extraordinary endeavor. There is another book coming next, however – which will try to describe a concrete attempt at the construction of such a technology: IEML. In 2009, I was lucky enough to work with Prof. Pierre Levy (and even produce a minor contribution to the project), which gave me the opportunity to gain some insight as of the potentially incredibly disruptive potential of this formidable invention.
IEML is a universal language. Although it is possible to translate pretty much any English sentence in IEML, it is not meant to be used as a language for human-to-human communication. IEML is a language of computable concepts. There is a base IEML dictionary that is very carefully constructed, and starting there, you should be able to express any imaginable concept, possibly creating new IEML words of your own. The IEML dictionary (or, I should say, IEML dictionaries, because you can theoretically have many of them), is a network of interrelated concepts where every node or word or sentence receives its meaning from the nodes it is related to. Now, and here is where things get interesting, all these relationships are mathematically computable. What it means is that IEML makes it possible to make computations on meaning.
From the very beginning, Fluxtream was conceived in part as a way to experiment with IEML’s core concepts. Since IEML’s technology is not yet mature enough to be incorporated into a real product, all we can do is create an experience that is at least suggestive of what would be achieved if we could use the “real thing”.
Fluxtream is an ideal testbed for IEML for multiple reasons:
- If you’re using Fluxtream, it’s because you are interested to know the story behind the raw data, what it means;
- when we become aware of our behavior, we are unlocking our power to act on it; that is the single most compelling incentive for users to annotate their data on a daily basis - and IEML can do nothing for you if you don’t annotate something with it
- Fluxtream represents a slew of data streams that together describe human behavior in an extremely detailed and vivid way. A great deal of what constitutes the “human experience” can be deduced just by looking at Fluxtream’s data streams with the right interpretation key
We are currently working on a set of features to help you categorize your daily actions. First we will let you create events of your own so you can describe what you were doing at some point during the day. Then we will let you define goals, both long-term and short-term — short-term goals contribute to long-term goals. We will let you associate goals with a choice of “conceptual maps” with percentages. A conceptual map is a set of concepts, like:
- family / work / leisure
- health / money / relationships / love / spirituality
- earth / water / fire / air
- …
For example you could create a long-term goal of doing more sport, which is 100% health related, and a “Find a Yoga class” goal that would be (for you) associated 80% with health and 20% with spirituality. When you have created some goals, you are able to associate any event in Fluxtream – including events that you have created yourself – with these goals. We are now able to give you back some extremely interesting information about your behavior:
- how much time are you spending on different goals
- what’s the percentage breakdown of your activities in terms of the initial “conceptual maps”
- what are your “most popular” goals of the day/week/month
How is that related to IEML? First there’s the initial impulse: you gave some simple meaning (‘this’ is related to ‘that’) to a dead piece of data. With that, Fluxtream can automatically compute some other, more general, meaning that you can learn from.
Of course, when we will be able to use IEML, the interpretation possibilities will be virtually infinite. In the silicon brains of our next-gen AI-powered friends, our most minute actions will not only make sense in the constrained economy of our private, self-absorbed whereabouts; they will automatically be linked to the most advanced concepts and theories in medicine, behavioral theory or philosophy. IEML will offer us a clear path to go from the former to the latter.
- 4 Unknown Aspects of the Quantified Self Movement
The Quantified Self is, yes, a Movement. QS Meetups pop up everywhere in the world. There are already 45 Meetups worldwide, (22 in the US, 12 in Europe, then Tokyo, Beirut, Sydney, Rio,…). Yet QS is still very hard to explain to non-initiates.
Problem is, “Quantified Self” is a vastly misleading label; by no way is it a description of the movement’s true meaning. I will hereunder expose 4 hidden aspects of what the Movement is really about.
Epistemology - QS is citizen science. It encourages experimentation, abductive reasoning and peer review at the most basic level. Everyone becomes a highly complex sensor on a global mesh. It is an opportunity to uncover knowledge in a way that has never been seen before.
Society - experiments in QS are only instrumental to well-being, not to power or money. Also, it is very humbling to think of yourself as a laboratory rat.
Religion/Spirituality - In today’s world, science and technology can be threatening at times. QS is a way to reconcile with these wrathful gods. At the personal level, self-quantification is a commitment to let go of one’s personal judgments and interpretations and to faithfully record information on a contiuous basis. It’s a behavior that can be linked to prayer. The meetups is where proper hermeneutics is performed and where a sense of community is achieved, along with a proper ritual and a code of conducts.
Politics - Quantified Self suggests a rather egoistic endeavour. The opposite is true: Quantified Self is nothing without the Meetups. There, private experiences become an opportunity for the public to question their validity, to reject or embrace their outcomes. It is the most basic form of democracy, based on science and on experiences that are (mostly) of direct interest to one’s well-being. No ideology, only practical, applicable and useful science.
- Opening Up
With the advent of cloud technologies, Open Source sounds a bit passé. The promise of Open Source is pervasiveness of ideas; the promise of the Cloud is to create pervasive experiences. Most people like experiences better than ideas. We love both.
Running a web application is hard. Software is just one aspect of it. The rest is performance, security, upgrades management… lots of operational chores. As an application grows more popular, those aspects tend to become prevalent. After a while, it becomes difficult to draw a line between the core software and the rest of what makes up a global, highly available and distributed application. Facebook or Google cannot be open sourced.
Given Fluxtream is dealing with extremely private and sensitive data, we feel like it is not possible for us to follow Facebook’s model of a completely walled garden: people should be able to conduct experiences with our software in their private cloud, on their own servers, or even on their local machine.
Fluxtream.com remains our mission number one: to provide the best lifelogging experience around to everyone with a web browser; no compiler, no debugger, no bash scripting required.
Fluxtream.com’s goal is to be a cloud’s first-class citizen. As such, it will provide both a client-side javascript widget framework and a push API centered around user behavior. We believe in the emergence of an ecosystem of cloud-based services that will leverage personal data to help individuals become healthier and happier, learn new skills, develop consuming habits that are environment-friendly, etc.
However, for those of you who would like their own instance of fluxtream, we will soon provide a Community Edition (CE) that you will be able to download and run where ever you like. You will be able to clone the source on github and hack it, too. Fluxtream CE uses totally commonplace technologies and frameworks: java, spring framework, JPA, jQuery,… so the majority of you should feel right at home.
Fluxtream CE is the stripped down, open source, easy-to-build, easy-to-deploy version of fluxtream.com. It is a testament to our dedication to an open web and to the free circulation of ideas.
- Generation Gap
I came to visit my Godmother recently. It was a long time I hadn’t seen her. To me, my godmother is the incarnation of good taste; when I was a teenager she was, amongst other family friends and relatives, the kind of person who made me think I was special. Of course, I was just a jerk; or at least it is what my age group thought of me — and they were right, for the most part. But, sadly for me, I am pretty sure my godmother herself didn’t think too much of me either. It’s a generation thing. Let me explain.
Clairette was part of a philosophical movement called “situationism”. From the outside, it looked mainly like a modern form of epicurism. If you asked, there were some highly technical books (“La Société du Spectacle”, “Le Livre des Plaisirs”,…), but it looked like nobody had really read them. What counted was the good life. Yeah, that’s right. Situationism was mainly about being cool: eating and drinking, be opposed to any kind of hard work, be creative… that sort of stuff. And they were dead serious about it.
Of course, as a teenager, I was very admirative of this way of life, given the particularly intriguing language, the immoderate taste for baroque musique and fine cooking. Sadly though, as it turned out, for the young adult in seek of a professional future that I was, this particular form of wisdom they had come to practice was not very appropriate. Things turned awry for me, and I became a computer nerd. From then on, when meeting Clairette, I would forever have to experience the display of a peculiar mix of pity and a self-imposed sense of god-motherly benevolence. To me, it looked and tasted like despair.
It is hard to imagine the degree of contempt that a situationist feels inside her guts when she thinks of a computer. After all, in their eyes, computers are the pinnacle of “productivism”. The very concept of productivity itself is an abomination; it is the cause of all sorrow in modern society, that which causes us to work with a bad conscience, in order to produce more “shit” (Situationists love to use the s-word, in all sorts of variations, when talking about material objects). And on and on.
Anyway, when I started to explain fluxtream the other day, I knew what to expect.
You know, I “get” situationism. Just like when I was a teenager, I still admire Clairette’s style. Dandy intellectuals all over the world have certainly helped shape a more open society or, at the very least, a much more fun and interesting one to live in.
I just don’t think intellectual dandism scales very well. Technology on the other hand is here to stay. It is now the driver, not the servant, of radically new ways of doing business, of living together and share experiences, that involve an unprecedented amount of creativity — and style — to succeed.
Technology, pleasure, creation, freedom, personal expression - it is time that we cross the generation gap.
- Privacy
Are you worried about your privacy? You probably should. If you are somewhat into new technology, you must already be using a lot of stuff that stores your data very far away from your computer, your home, even from your country. Odds are, a good part of who you are can even be gathered from doing a google search with your first- and last name. It’s that simple. So… do you really care about your privacy? Let’s face it: unless you want to retire in a cave until you die, you are going to continue to give away lots of information about yourself - for free, willy nilly. We have entered the age of sharing.
Sharing is a good thing. But, at times, it can be very, very scary. Like, when work colleagues and recruiters know more about you than you would like. And it happens - or could happen - all the time.
As far as fluxtream is concerned, we take for granted that a lot of information about you is out there: we are, for a good part, relying on information that’s living in systems belonging to Twitter, Google, Fitbit or Zeo. We are not counting on that situation to change - on the contrary.
But
First, we want you to never be surprised again by someone, or some-thing, who knows more about you than yourself. You are the master of your own data, you know what it means, why, when and where you did all the things you did.
Then, of course, we make sure that your information is safe with us. You are our only customers and we don’t share your information with anyone. Nothing. Zilch. With no one. Not. Ever.
- Android, iOS and the Quantified Self
When the 4S launched, I decided that I would reward myself with a brand new phone with our first public beta. I own a Samsung Galaxy S and I have had a terrible experience with it. As a result, I have suffered iPhone-envy for about a year. I must say, I have endured a fair share of frustrations with my Galaxy S. First one was that the iPhone 4 came out a week or so after my doomed Android purchase. I had owned a 3G for 2 years and wouldn’t do the upgrade for a 3GS, but I would certainly have done the plunge for the iPhone 4.
However, more serious trouble started when I figured my phone had Android 2.1 preinstalled. Froyo (Android 2.2) was out since a few months already and I just had to get the upgrade. After hours and hours of trying to do that with the “official” Samsung-supported tool - it’s called “Kies” and is probably the worst piece of software I have come accross in the course of the last decade, I decided that I would try to “flash” (yes, it sounds geeky, and it is geeky, too) froyo on my brand new phone, despite all the warnings that I would void the warranty, potentially destroy my phone and loose 500€ in the process. As a result, I managed, as promised by the User Manual, to “brick” my phone and make it totally unusable. Fortunately, I was able to send it to repair in the Netherlands and got it back 3 weeks later with Froyo installed. Since then, however, it is still impossible to upgrade to more recent versions of Android through “Kies”, which invariably tells me, in its own cryptic way: “This version of your software cannot be upgraded”.
Turns out, this is a fantastically common flaw with Android phones: in most cases, you just cannot upgrade the OS, as this article attempts to expose in excruciating detail.
Then, there is my everyday Android experience. In one word: bad. Of course, it does what you expect it to do, but there’s no pleasure involved. These things are hard to explain, but they usually boil down to a combination of flaws coming from both hardware and software. The Galaxy S, while boasting a 1 Ghz processor, is still a real drag in many instances. The Froyo incarnation of Android is OK, but lacks the slick factor of iOS or, for that matter, more recent versions of Google’s OS.
Again, this is a known fact: Android users don’t use their phones so much for anything else than texting and talking. There’s a remarkable article about this, too. Let me cite the first, and most prominent facts in it:
Apple has 19% of the global smart phone market. Android has 48%. Yet…
- 2/3rds of Google’s mobile search comes from Apple iOS devices.
- Apple’s iOS accounts for 54.65% of mobile Web browsing. Android has 16.26% and Research in Motion (RIM) has 3.29%.
Can you imagine that? Proportionally, iPhone users browse the web approx. 8x more than Android users! Same pitch, totally different usage.
Finally, there is the famous old short essay by Umberto Eco about the then-raging war between the Macintosh and DOS-powered PCs. It’s even more valid nowadays in the context of the iOS/Android rivalry. Again, a short citation:
[…] I am firmly of the opinion that the Macintosh is Catholic and that DOS is Protestant. Indeed, the Macintosh is counter-reformist and has been influenced by the ratio studiorum of the Jesuits. It is cheerful, friendly, conciliatory; it tells the faithful how they must proceed step by step to reach — if not the kingdom of Heaven — the moment in which their document is printed. It is catechistic: The essence of revelation is dealt with via simple formulae and sumptuous icons. Everyone has a right to salvation.
DOS is Protestant, or even Calvinistic. It allows free interpretation of scripture, demands difficult personal decisions, imposes a subtle hermeneutics upon the user, and takes for granted the idea that not all can achieve salvation. To make the system work you need to interpret the program yourself: Far away from the baroque community of revelers, the user is closed within the loneliness of his own inner torment. […]
I find it a strikingly interesting analogy. While I have a lot of sympathy for the catholic worldview and the concept of universal salvation, it is a fact that the rapidly-evolving world or technology suffers from a predominantly protestant mindset: it is highly competitive, highly speculative: a few prosper, everyone else struggles.
There is no doubt that fluxtream belongs to this world. We certainly hope we are among the lucky few, but we are facing a hostile, if fascinating, world with an abundance of promising ideas and very little capital. Apple’s walled garden is a broken promise in this world and, indeed, my android phone works better for the stuff we are betting on. It logs texts and phone calls for instance - which is a feature I’m afraid I cannot do without now, does geolocation through google latitude better and applications like Funf Journal, which are invaluable for us, are impossible to do on iOS.
This poses an interesting question for Apple. End of the nineties, Apple had to open itself to the world. It had created too many closed technologies: network protocols, file systems, programming languages,… To get it back on track, Jobs chose not to continue to do things differently, but to do things better, and succeeded at it. But he did it by embracing open standards. With the advent of cloud technologies, it looks like Apple is still doing things better, in some way, than its competitors, but it forgot to do it in an open way. I bet it is not going to bring them only good things.
Will I buy an iPhone 4S? Or is it going to be the next top-of-the-line Android? Shall I keep my money and continue using my crappy Galaxy S? One thing is for sure, Fluxtream is launching in 2 days, and it’s the best lifelogging experience around no matter what :-)
This video made me think a bit about what we do: Fluxtream is a Quantified Self inspired project and QS is almost certainly often perceived as your typical left-brainer nerd initiative. I think it is just the opposite: reflecting on your life through numbers and sharing this information with others will have you exercise your intuition at least as much as your analytical capacities.
This world needs more human sharing and interpretation of the raw statistics that keep pouring on us!
- Export API
We have brainstormed quite a bit about how to provide developers with a nice API for them to play with. We think it is quite compelling to be able to access all your personal data, as they are visible in fluxtream, using just ONE API key.
After all, a Fluxtream API would allow you to get information (though not always complete) for a great (and growing) number of personal data services:
- dropbox
- fitbit
- flickr
- foursquare
- freshbooks
- github
- google calendar
- google latitude
- gowalla
- khanacademy
- lastfm
- nike plus
- picasa
- sms backup
- toodledo
- withings
- zeo
Disclaimer: you will not find all of these connectors just yet, as a lot of them are work-in-progress.
Surely, the Locker Project has a clearer agenda on this. But we are thinking something quick-and-dirty for simple hacks were you just want to get the job done with a minimum of hassle: BASIC auth and a few methods with a start and end date and the type of object you want to retrieve.
As for the format, we would probably just serve you the stuff we use internally. It would make sense, as it is the information you would most likely expect to lay your hand on. And it’s simpler for us to implement, too.
- Why the name?
I hear this question pretty often. Since you are reading it, you are probably wondering as well.
Life is a flux. We would like to think we are in control, but in reality, things are in a constant state of evolution: the tools we use, our personal goals, our interests, our friends… everything changes.
One thing remains constant: information.
In today’s digitized world, our social footprint, our media consumption, the productivity tools we use every day, even our bodies,… all of that leaves a permanent trace in bits and bytes,… a never-ending stream of tiny events that, increasingly, defines our very self.
Fluxtream is there to provide you with a way to control, dissect, repurpose and otherwise analyze this non-stop flow of information that tells so much about your behavior, about your personality and daily choices. It transforms the flux of life, this disorganized, mixed-up conglomerate of stimulus/response automatisms into discrete, well-defined streams of processed information that you can act upon.
-> Flux: real life
-> Stream: digital lifeGot it?
- Our winning technology stack
Since we’re now approaching beta, we are setting a few things in stone really, technologically speaking. We’d like to share some of our choices with you. It’s nothing fancy (yet). We use proven stuff that millions of developers are familiar with. At this stage we aim for an easily maintainable architecture that we can scale out using a simple server sharding strategy.
- Java 6: a fast vm and efficient programming techniques = outstanding performance
- Spring and Spring Security 3: straightforward action-based web framework; a very handy set of annotations make it extremely simple to expose web methods
- JRebel: instant class reloading; with that we get the same development comfort as with interpreted languages like ruby and php. An invaluable tool
- Hibernate/Mysql: we use a very denormalized schema and some data redundancy in provision for some future, alternative nosql storage; for now, it’s still very, very fast
- Lucene: in combination with hibernate search: seamless mysql/lucene synchronization; almost infinite possibilities with regard to stemming and lexing…
- StandingCloud: these guys provide us with fantastic support and the guarantee to be able to use cloud infrastructure from amazon, linode, rackspace, slicehost and others at no additional cost
- GitHub: THE web-based Git repository; provides easy rights management and code browsing. We also use their (basic) Issue Mgmt System
- Twitter’s Bootstrap css framework - we use it for our admin GUI - get a rockstar GUI designer for free
- New Relic: performance-wise, things are looking good right now, but bottlenecks are difficult to identify in a live system; so when shit hits the fan we think this one can help us keep our asses covered
- Eclipse: I sincerely cannot imagine how we could live before this IDE’s incredible refactoring support; allows us to re-engineer the code inside and out in no time; plus, it is very popular and it has a great community
- Loggly: logs are your best friend for after-the-fact debugging; loggly will let you create powerful dashboards and identify errors and attacks FAST
On our short-list of cool technologies to integrate to the platform:
- Redis: we would like to improve caching and get instant access to bookkeeping data that we are currently storing in the database
As I said, nothing fancy: just the best available tools of the trade for the great majority of us java j2ee developers. It means readily availability of a core set of best practices around agility, quality assurance and sound architectural patterns.
- Singly: a web-based, decentralized, open source, universal personal data platform... and funded
Singly just launched. What is it? Here’s what they have to say:
- All of the web’s relevant links curated by your social streams
- All of the people you’re connected to across the web
- All of your photos in one place
It’s based on the LockerProject which is a fascinating open source project with basically all the recent buzzwords in it: decentralized, nosql, using nodejs…
Singly/LP is interesting on many levels: first, it’s a major and technologically ambitious open source project for personal data, providing a platform for other developers to build real products on it. Second, it is a clear sign that the personal data space is a hot one. It is not the only one, though: ManyBots, for instance, is less ambitious, but may have a clearer path to developer adoption.
Are we using it? Not yet. Why? We are concentrating on meaning, the real meat behind the data. We have our own home-brewed data import/sync/access layer that we like to think is simple enough that we know exactly how it works. But no, it isn’t decentralized and it’s not open source. We don’t think it matters, however: we think it’s not necessarily the data itself people want to own, it is what it means that’s really important.
- Minimum Viable Product
We have drastically reduced the scope for fluxtream recently.
The result: we are focusing on a simple - yet terrific, user experience. We are making it fun for you to remember your state of mind any given day, by letting you visualize everything you have done and easily comment on every event, big or small. We want you to dive into your recent and distant past and get a vivid picture of what was on your mind back then.
It’s an addictive experience: at the beginning, you will think “What the hell was I thinking of?…”, then you will start connecting the dots. In the end, you will be glad to be the main character of a developing and never-ending story, your own.
And this is only the beginning.
-
Tweets
- btw, I'm now working full-time on fluxtream/bodytrack integration. https://t.co/ZVYhEEQr #yay
- I'm Speaking at OScon 2012, in Portland: "BodyTrack: Open Source Tools for Health Empowerment through Self-Tracking" http://t.co/mq9s1YEw
- I'm attending Community Leadership Summit 2012 -- http://t.co/1EnDyz8i
- I'm organizing a conference on Complementary Currencies. Great Speakers. June 19th. http://t.co/Z961Kn4a #SmartCity
- @quanders sure!
- Instagram for Android Test #test http://t.co/HpGAME5q
- Just finished my last mission at the EC. I'm for hire, get to know me: http://t.co/Qhm2t4ZG
- Two Stanford profs put an AI course online for free. Then 160,000 students signed up. Read about it on the WIRED app: http://t.co/aKNELAr8
- Had a great time with you guys. Hope to see you again very soon. @annerwright @jeztek http://t.co/YHveh0ut
- J'ai eu 40 ans hier. Mazette. Pour l'occasion, je vous invite à me rejoindre à la Maison du Peuple (Parvis St Gilles) ce soir dès 9h!Mar 16, 12 from web Reply, Retweet, Favorite
- I just created a Meetup group on alternative currencies. Currently looking out for speakers and presentations: http://t.co/DxqqN4IHMar 16, 12 from web Reply, Retweet, Favorite
- bloody hell: I'm 40!
- @WordGhost thanks!
- A must-see video about #growth, #debt, resource depletion and what you can do about it (not much): http://t.co/5ggt5iuX #transitionMar 06, 12 from web Reply, Retweet, Favorite
- The final program for tomorrow's presentations at the Brussels QS Meetup: http://t.co/ESOrYiHr
- Just finished http://t.co/wuoYLpER - By a world-class currency expert: how to change the world NOW with alternative currencies. A must read!
-
Available April 1st 2012!
- I am available for contract work starting April 1st.I usually live and work in Brussels.My rates vary depending on the role and the technologies at stakes (Developer/Architect, Java/Javascript) and the fun factor involved. Since I usually have fun when I am very productive and vice-versa, you will find this statement to be a total bargain.If you would like to discuss a project or position that could be a good fit for my specific profile, please do not hesitate to drop me an email: candide [at] fluxtream [dot] com
