"Open" is often a feel-good word when applied to computing. An "Open Web"?

How could anyone be against it? And yet there is seldom discussion on what openness means in this context. But is openness obvious? There are many dimensions to openness in the web, and one person's openness maybe another's limitation. For instance, does javascript make the web more or less open? There are arguments both ways. This talk will try to tease apart the different dimensions of openness, and discuss the barriers that still need to be lowered.

The web exists today because of open technologies created in the past. Open protocols and formats like TCP/IP, HTTP, SSL and HTML have been elemental in ensuring the wide adoption of the internet as a standard media for information and communication today. That does however not mean that data is openly available. It also doesn’t mean that sites are more than stovepipes, only addressing their own information and services. For web-applications to truly open up, more is needed. The ‘more’ is interoperability, or rather, open interoperability. And there is a lot to be done still to come even close. My presentation will go into what’s needed to create a truly open web where data can flow freely between different services and where applications can cooperate and interoperate.

• Forms of Interoperability: Dataportability, Data-exchange and ‘Intercooperation’.
• Architectural layers of Interoperability: Business, Semantic and Technical.
• Types of interoperability that exist but are wrong (database-level data-exchange, proprietary APIs, etc)
• Web 3.0

E-government. I will cover some aspects of the Dutch situation. Interoperability is a precursor for creating an open E-government web. Subjects covered are amongst others OSB, Semantic standards, technology-choice.

With the complexity of web applications ever increasing, developers have run into problems caused by limitations in the design of HTTP. In particular, HTTP was designed for servers to transmit data only if clients request it (client pull). However, in web applications there also are scenarios where information becomes available on the server more or less in realtime, with no means of pushing these to the client.

The conventional way of collecting this information is for the client to continuously request the latest information (polling). This may saturate the server with requests, while the information received by the client is most likely outdated. Comet is a term used to describe techniques to work around this limitation, allowing the server to push data to the client at the moment it becomes available. While Comet meets the needs of application designers, it is not without its consequences for both developers and systems administrators. This presentation will illustrate how Comet works, and discuss its advantages and limitations.

GeoClue is a modular geoinformation service which aims to simplify the creation of location-aware applications. It has been designed with desktop and mobile use in mind, and works over D-Bus.

The presentation sheds light on what location awareness is all about. The design and motivation behind GeoClue will be explained and some hands on examples of using them in applications given. You'll also find out how GeoClue fits into different platforms (maemo, OpenMoko, OLPC, GNOME desktop).

As mobile and social aspects of computing have become more popular, the need for collecting, sharing and utilizing geoinformation has become more important. This need is not just about the user wanting specific geographical information ("Where can I get ice cream?", "How do I get to the hotel"), but also applications needing context to be usable: A weather applet would be much nicer if it just knew which city the user is in, and an instant messager or microblogging client might want to give the user an option to show current location to contacts.

The GeoClue project started in October 2006 at GNOME Summit with the idea of providing an easy and reliable way for applications to obtain information about current location. This is achieved by defining a few fairly simple APIs, and using a variety of data providers to implement them. Several providers are used as no single provider will be useful in all situations: GPS does not work indoors, web services require an internet connection, etc.

GeoClue participated in 2007 Google Summer of Code as part of the maemo project. Since the SoC GeoClue development has been continued by Garmin/OpenedHand, and a new API version will be released soon. Projects currently working on GeoClue support include the Telepathy instant messaging framework and the Mauku microblogging tool.

The Open Web is full of abbreviations, from SaaS to API or RIA. But the law has abbreviations of its own that present unique challenges, not to say problems, for anyone building open web applications or services. In this talk we will discuss the most important ones. We will cover copyright and database rights, review EULAs and discuss whether privacy laws can hamper social networking sites.

In recent years numerous films and books have shown representations of a 3D enabled Internet embodying Web 2.0 ideals of rich user experience and participation.

Web technologists have often tried to harness the graphical power of modern desktops to realize these visions, and while we are still a long way from a Minority Report style interface there are still some interesting things happening with 3D content and interaction online.

This talk will present some currently available technologies (e.g. ActionScript 3) and discuss not only the impact they could have on the way you present content to your users, but also exciting new possibilities they open up for user interaction. Emphasis will be given to open source tools and libraries (especially Papervision3D) that facilitate the creation of 3D interactive web content.

Tegenwoordig kan bijna iedereen een website in elkaar zetten. HTML en CSS zijn bekende technieken en met een beetje inspanning is de basis ervan gemakkelijk te begrijpen. Ook zijn er veel open source systemen te verkrijgen, waardoor u bijvoorbeeld in een handomdraai een webshop creëert.

Helaas leidt het beheersen van de techniek, of het gebruiken van een meegeleverde standaard template, niet automatisch tot de ontwikkeling van een goede website.

Het gebruik van de nieuwste technieken is natuurlijk uitdagend, maar mag niet leidend zijn bij de ontwikkeling van een website. Uitgangspunt moeten de doelen en vaardigheden van de bezoeker zijn.

Bij de ontwikkeling van een website is aandacht voor Website Usability dan ook minstens zo belangrijk als de techniek erachter. Een website moet namelijk bruikbaar en gebruiksvriendelijk zijn. Heeft u er bijvoorbeeld al eens over nagedacht hoe uw broncode in elkaar moet zitten zodat bezoekers met een visuele of motorische beperking uw website kunnen gebruiken? En ondersteunt het gebruik van AJAX uw webformulier, of maakt het deze juist onhanteerbaar?

Website Usability is een belangrijke voorwaarde voor het succes van uw website, omdat bezoekers anders voortijdig afhaken. U ziet dan wellicht hoge bezoekersaantallen terug in de statistieken, maar wanneer u gaat kijken naar hoe lang men op de site blijft, hoeveel pagina’s men bekijkt en hoeveel aankopen men doet, dan blijkt dit flink tegen te vallen!

In deze presentatie wordt dieper ingegaan op veel gemaakte fouten op het gebied van onder andere website vormgeving, de werking van formulieren, foutmeldingen, Javascript menu’s en accessibility. Met behulp van screenshots worden diverse praktijkvoorbeelden toegelicht.

Most of the user's interaction with the Internet, or rather "The Web" is currently going through the web browser. As a development platform, this appears to be rather limiting. The diversity of input devices, needs for caching and offline usage are all problems which are not consistently solved for most applications.

Deep integration of data from the web while using local UIs provide one way of getting around such limitation. The KDE team is working towards deep integration of online data and web services into applications and the primary UI of the device. This allows for streamlined user experience combined with merits, modern web applications offer the user.

In this presentation, Sebastian demonstrates and explains some of the technologies used in Akonadi, Plasma and Qt that integrate user's data such as email, contacts and bookmarks as well as the Social Web and other web services deeply into the desktop, taking advantage of locally available infrastructure on the client device. Akonadi is a local cache and one-stop-shop for applications that want to integrate PIM data. Plasma is the primary user interface of the KDE desktop, but also a library to build canvas-based user interfaces for wide range of devices, from mobile to media centers.

Qt itself offers numerous techniques that allow asynchronous fetching of data, XML parsing, XSLT transformation of XML data sets. Qt's webkit integration makes it easy to directly integrate content from the web into the applications, or give applications this "webby" feeling.

De IP adressen zijn bijna op - het gaat harder dan de aardolie. Het aantal internetverslaafden groeit nog steeds, het nieuwe Google Chrome OS kan ook niet zonder en alle gemeentes moeten helemaal digitaal. Als een normaal meisje van een jaar of 20 wakker wordt, kijkt ze tegenwoordig eerst op Hyves en dan pas in de spiegel. Ontbijten doet ze al helemaal niet meer - of ze bestelt 't on-line. Ondertussen denkt de hoofdcomputer van Google dat 'ie wordt aangevallen als Michael Jackson dood gaat - en wat gebeurde er ook alweer met Skynet in "The Terminator"?

De vervloeking van onze oude levenswijze, de opkomst van alle informatie die je niet wilt weten op momenten dat je even niks wil. Porno beschikbaar op elke straathoek en daartussen.

Kortom: de zin en onzin van de snelle vervlechting van onze cultuur met het internet. Wordt 't nog ooit wat?

This talk will give a real world example of the benefits of creating published and open API’s using Ampache as a case study. In 2002 Ampache added an XML-RPC API that was never documented or published. I will detail many of the mistakes made, and lessons learned from both a community / documentation standpoint as well as technical mistakes made in not maintaining backwards compatibility. Additionally, I will address limitations encountered when end users attempted to use it on non-customized web servers. I will detail the creation of the Restful API in both its first and second incarnation. Using demos of a few of the applications that have implemented the new Restful API as proof that Open, and documented API’s a nothing but good for a web applications adoption and growth. I will talk about lessons learned and the advantages of a published, and open API. Touching briefly on when it’s ok to break backwards compatibility, and the added server load experienced with exposing an API and ways to optimize it through caching and dividing of result data in the application and on the server. I will also discuss XML responses versus JSON and the advantages of providing both.

Content repositories allow you to separate the actual front-end of your application from background processing tools. More than just their underlying databases, they impose common rules for data access, and keep multiple applications up-to-date on data changes through signaling. Midgard2 provides a flexible content repository that avoids the restrictions of the traditional ORM approach. And not only your PHP web application, but also to possible Python, Objective-C and C# tools you use.

This enables you to split applications into smaller, easily maintainable and scalable pieces that can be run on different systems and platforms as needed. In addition to web, the Midgard2 library can be used for desktop and mobile application development, building software that synchronizes with web services. It is based and engineered fully on the top of the desktop (GNOME) software stack. Being highly modular and having very little dependencies it scales from a note taking application to a full-blown CMS system. Combined with advanced replication capabilities it allows you to synchronize data between offline and online instances of your service.

Midgard2 is an Open Source Content Repository and it provides an objectified view to the data and services surrounding it. At the basic level it abstracts the database access (SQLite, MySql, PostgreSQL) but this is only where it all starts. Serialization & replication, managing own storage objects, multi-process access to data are all covered. The fully object-oriented (GObject-oriented) API allows you to focus on the data, not the database syntax.

The Web is going Social and the Social web is going open! The social web for many is as big as a paradigm shift as the very invention of the internet it self was, the way we use the web and interact has been changed for ever; In this session we'll cover the core open technologies (often referred to as "The Open Stack") that drive this rapid innovation like OpenSocial, XRDS, OpenID and OAuth, how these can be used to solve the registration-hell problem, and finally how to build OpenSocial applications or become a social platform your self.

The past decade has witnessed an almost explosive growth of multimedia presentations. A typical multimedia presentation today consists of a set of video/audio clips, which are relative to each other in terms of temporal and spatial relationships. These video/audio clips can come either from the same media source, or from various media sources, or even from different servers. To avoid service interruption and eliminate switch delay among these clips as if they are continuously rendered from one media source, proper cache and prefetch strategies must be employed.

SMIL is the W3C standard used for rich media presentations that integrates streaming audio and video with images, text or any other media type.

In this paper, we present the design and implementation of seamless playback for video/audio in Ambulant, which is an open-source media player with support for SMIL 3.0. To optimize the playback experience, various facilities, both automatic and author-controlled, are designed and implemented with the intention of reusing renderers and prefetching the media clips in advance. We carried out experiments in order to validate that our techniques can lower the initial start delay of media rendering, decrease streaming disruptions, increase cache utilization and improve service response time. In addition, we are planning to study the applicability of our techniques to other languages than SMIL.

Deze lezing gaat over het ontwerpen, de installatie en configuratie van een open source web "portal", dat veilig toegang biedt naar één of meerdere 'interne' (of op een andere manier niet volledig openbare) websites via authenticatie met een token.

Het webportaal zelf is volledig gebaseerd op open source software, namelijk FreeBSD and Apache 2.2, met enkele Apache specifieke modules (mod_proxy_html en mod_auth_xradius). De implementatie die ik zal beschrijven gebruikt RADIUS en CryptoCard tokens, wat niet open source is, maar zou wel gebruiktworden met andere RADIUS backends.

De basisconfiguratie is redelijk eenvoudig (voor een webserver met statisch pagina’s bijvoorbeeld), maar kan uitgebreid worden tot meerdere webservers en tot op zeker hoogte zelfs web services.

At last year’s KDE conference "Akademy" the vision of the Social Desktop was born. The idea is to bring the power of online communities and group collaboration to desktop applications. One of the strongest assets of the free software community is our worldwide community of developers and users who believe in free software and who work hard to bring our software and solution to the mainstream. The idea is that the free software world can create a huge "Unique Selling Proposition" by combining our community with our software.

The idea is to create an open standard for exchanging data between different online services and desktop applications. This standard is call open collaboration services and is hosted on freedesktop.org.

Ideas for the Social Desktop are:

• Show other KDE users near me.
• Show what my KDE friends are doing at the moment.
• Become a fan of an application or a developer
• Search an online knowledge base directly from the application or your desktop without a browser.
• and more.

This talk gives an overview of the current state of the Social Desktop. Which features are already implemented, which features are a work in progress. I will also give hints how developers and users can get involved and would like to start a discussion to develop the social desktop concept even further.

Whether you're viewing a video on YouTube, a slideshow on Flickr or an advertisement not made by Google; chances are you're looking at a Flash movie. Flash has become so ubiquitous in recent years that it is difficult to imagine the Web without it. Yet Flash is proprietary technology and the Web will never be truly Open so long as Flash remains in its current state.

The lack of a free Flash player has always been a sore spot for the free software desktop. Almost every GNU-based desktop is complemented with an installation of the proprietary Adobe Flash plugin. The Gnash project was founded in order to fill this perceived gap.

Flash and "openness" have always been at odds, because the creators of Flash have been careful not to reveal much information about their technology to the public. The community at large over time reverse engineered and documented the workings of the Flash player and associated protocols. However, there are problems facing Gnash that no amount of technical expertise will ever solve.

This talk will focus on the challenges facing the Gnash developers and their connection to the Open Web. Substitute technologies for Flash, such as HTML5, will also be discussed.

In de lezing besteed ik aandacht aan de functionaliteit van de Yubikey, de voordelen en nadelen van de key en de diverse werkwijzen die de key ondersteund. Ik ga in op gebruik van de key in combinatie met een LAMP server. Vervolgens vertel ik wat CAcert is en doet. Aan de orde komen met name client certificates, hoe deze te genereren en te laten tekenen en hoe deze in combinatie met Apache en PHP te implementeren.

Tenslotte laat ik zien hoe je een binding tussen de Yubikey en het certificaat aan kunt brengen.

Afgesloten wordt met een demonstratie van 3 factor authenticatie met gebruikmaking van deze technieken. Een en ander zal worden gelardeerd met programmeervoorbeelden en live voorbeelden van het gebruik.

Nearly every software project relies on its community for feedback, bug reports and word-of-mouth advertising. Unfortunately, having a strong user base does not automatically lead to a strong and successful community.

How can you turn users into contributors? How can you avoid negative publicity? How can you deal with controversial opinions? How can you build a strong team from people spread over various timezones? How can you ensure that diversity is beneficial instead of harmful?

How can you turn your project into a welcoming place?

These are the questions nearly every project runs into when it starts growing. It becomes more complex when smaller projects integrate into a larger community like Gnome and KDE or those around Linux distributions: they will have to compete for contributions against the bigger and maybe more appealing sub-projects and tie new members to their base.

This presentation will give an introduction to the basic rules of community building, look at internal processes of communities, show you tools that help you manage your project's perception and provide you with resources on the subject.

Do you know that the web is evolving?

First, the web was a document repository. Then, a world of applications. The technologies have to move the same way. That's why HTML5 exists.

HTML5 brings more adapted features for a modern web. HTML5 doesn't come alone. CSS3, JavaScript 1.8, SVG, ... aim to make the web experience better.

This includes offline storage, native video & audio, thread mechanism, downloadable fonts, more interaction with users, geolocation, and more. Firefox 3.5 is a new big step in this way.

Let's see how far we can go with the new standards and Firefox 3.5.