Tag : development and web operations

Scaling a Web Application is a very hard problem, especially for small project and teams who do not have sufficient manpower, money, and time to solve this problem. Luckily Amazon already had to solve this problem in their datacenters and offers their services to other developers.

This talk will introduce the two most important Amazon Web Services, the Elastic Compute Cloud (EC2) and the Simple Storage Service (S3), and will present different ways to leverage them in your own web application. Ruby on Rails will be used for the examples but they will apply to any web framework.

Several use-cases will be covered that show how S3 and EC2 can be used to move load from your servers to Amazon’s or even to completely host your application at Amazon.

After the talk developers should know when and how to use Amazon’s services to scale their own application at low costs. We will also hear the CEO of Web OS platform Ghost and how they use more Amazon Webservices than anyone else.

Speakers :
-- Jonathan Weiss
-- Zvi Schreiber

The Facebook Platform has made accesible one of the richest demographic subsets available, yet most developers are content to create applications that rely on old world interaction.

The data available radicalises the way businesses, marketeers and developers can exploit the social graph. This enables application architects to visualise the intuitive and hence make applications that are responsive, and thus subsequently immerse the user in his own social network rather than that of the product or service being conveyed; an inherently more intimate interaction and hence more persuasive.

This presentation explores the various possibilities, pitfalls and future hopes surrounding the Facebook API in general, but also in relation to the way developers can make more socially aware applications.

Speakers :
-- Ankur Shah
-- Gi Fernando

The geonames.org geographical database is available for download free of charge under a creative commons attribution license. It contains over eight million geographical names and consists of 6.5 million unique features whereof 2.2 million populated places and 1.8 million alternate names.

It's developed mainly by speaker Marc Wick, Geonames is integrating geographical data such as names of places in various languages, elevation, population and others from over <a href="www.geonames.org—data-sources.html"> thirty-five sources</a>.

The data comes in different data formats : flat csv files, gml files, proprietary xml files, excel, esri shape files, binary files, even ms access dumps.

Learn how he handles this data with his LAMP software stack: linux, apache, tomcat, java, lucene, postgres/postgis. He'll also touch on his liberal Creative Commons licensing.

As Sir Tim Berners-Lee said, "This is a tremendous set of data you have there." Come learn how Marc did it.

For some time now browser evolution has been slow and fraught with politics.

In recent years the only innovation there has been has come from the Ajax community, in the form of exciting libraries and frameworks that allow programmers to push the browser to the limit.

But this has in turn created two further problems; first, these initiatives are not standardised, and consequently they don't take advantage of the native platform in a way that compiled code can.

In this session Mark Birbeck explains a two-pronged approach to addressing this problem. On the one hand small and focused standards need to be created for new APIs that will be useful in many different kinds of application.

But in parallel, a new kind of library needs to be developed that will take advantage of native implementations of these standard APIs when they are available, but will fall back to Ajax implementations as necessary.

This opens up many exciting possibilities for internet-facing that run on both the web and the desktop.

Apache Solr is an open source enterprise search server based on Lucene, with XML/HTTP and JSON APIs, hit highlighting, faceted search, advanced caching, index replication, and a web administration interface. Solr is utilized by organizations such as Netflix, United Way, Smithsonian, News.com, Gamespot, Internet Archive, and Digg.

This session will provide an introduction to Solr, showing how these powerful features can be used to increase website findability.

Ajax is becoming a part of almost every web application. Users expect this and are increasingly disappointed when page refreshes occur for trivial updates. An Ajax framework can help you create or update your application using the best techniques available.

Simon Willison, co-creator of Django and OpenID advocate, will take you through the scenarios where you'll want to use Ajax. He'll also explore the frameworks you can use to make improving your web app that much easier and faster.

Simon will walk you through Dojo, Prototype/Scriptaculous, jQuery and YUI.

AJAX has brought real applications to the browser. Database-backed applications, that used to be programmed for the desktop are now developed for Firefox, IE and Co. But working in the browser was usually over when you lost your network connection.

With Dojo Offline and Google Gears you can develop browser based applications that continue to work even when your network connection has said goodbye.

The ability to work offline, using your data on the road or with your customer, closes the last gap between desktop and browser based applications. "You're not connected to the internet"...doesn't matter anymore.

Realtime collaboration is also a key component to new apps. Comet, coined by Dojo creator Alex Russell, describes the set of technologies used to achieve it.

The session will teach the basics on how to create web applications that are able to run offline and how to add realtime collaboration via Comet.

New websites don't exist in a vacuum any more. Users expect integration with the broad platform of Web 2.0.

Taking examples from the Ruby on Rails implementation of Dopplr (the social network for frequent travellers) this talk will show developers how to:
* Import social networks from popular sites like GMail, Twitter and Livejournal
* Integrate with Facebook
* Create and consume people and event data using Microformats
* Use OpenID not just for login but also to aggregate user information from OpenID providers like AIM and Livejournal
* Work with OAuth, the emerging open standard for API authentication

On November 6th, Google launched the OpenSocial API. This API is designed to let application developers create their apps once and have it applied to all social networks.

In this session learn about the new APIs, what they provide access to (profile, storage, and activity), their GData counterparts, and what sites will be accepting them.

The security landscape has changed dramatically in the past 12 months.

Unless you are aware of Intranet Hacking, CSRF, Javascript Highjacking, and the many ways to fool an XSS filter, it's likely that your web application will not be secure.

Attackers used to concentrate on ActiveX, but now Javascript, CSS and even simple HTML elements have are used against websites.

This session will outline the challenges facing the inhabitants of this strange world called 'Web 2.0' and the options for protection, both from the point of view of site owners, and web users.

Ajax is becoming a part of almost every web application. Users expect this and are increasingly disappointed when page refreshes occur for trivial updates. An Ajax framework can help you create or update your application.

During this three-hour session Simon Willison, co-creator of Django and OpenID advocate, will take you through the scenarios where you'll want to use Ajax. He'll teach basic the fundamentals of Ajax as well as advanced techniques. Simon will also explore the frameworks such as Dojo, Prototype/Scriptaculous, jQuery and YUI you can use to make improving your web app that much easier and faster. Along the way you'll be writing code to test out the items you're hearing about.

Everyone's doing it - the poster children for "Web 2.0" are built on top of the LAMP stack.

The next generation of web-based applications are built with free tools, with few people understanding the best way to scale these applications out. But patterns emerged very early on - all of these applications share some common architectural principles that seem to be working.

In this brief state-of-the-world, we'll look at the various approaches to scalable internet application architectures and what we can learn from them.