Re: [DISCUSS] Web20Kit: A Web 2.0 technology evaluation kit

[Craig L Russell]

Hi Martin,

On Aug 26, 2008, at 7:51 PM, Martin Cooper wrote:

Yeah, an association with WebKit was my first assumption as well.

Agreed. New name.initiate().run().

technology demonstration such as this may lead them in non-optimal
directions.

My greater concern, though, is purporting to compare technologies using one application with a very specific purpose. Certain technologies may shine in one application scenario and suck in others, while other technologies may show quite the opposite characteristics. Picking one application with which to demonstrate and, especially, compare a set of technologies doesn't tell a particularly useful story unless you happen to be building just that type of application. Sadly, many people don't take that into account, so that a

There are at least two different objectives for the project. First is to show how to build applications in different frameworks that would give an idea of how the pieces are put together using the different frameworks. Second is how the implementations actually work under load. What we would like to do here is to put an application together that has a number of web 2.0 features and then mix and match the presented load based on users' configurations. So rather than a fixed set of operations that yields a single number result, varying the load would allow you to compare operations that better match your application.

certain applications, a content repository might be more appropriate.

web services. There are many, many more options to consider.

Incidentally, there also appear to be assumptions about the architecture. For example, it looks like the use of a database is assumed, whereas in Similarly, the use of a server-side web framework versus directly calling

Craig

These are excellent points where it would make sense to add a replacement point for technology. With some skilled architects on the project, defining the replacement points (architected interfaces) would be a strong plus.

--
Martin Cooper

On Tue, Aug 26, 2008 at 7:13 PM, Brett Porter <[EMAIL PROTECTED]>wrote:

Without too much thought into the rest of it just now, the first thing

I thought was that this would have something to do with WebKit, which
it doesn't and would probably be very confusing?

- Brett

2008/8/27 Craig L Russell <[EMAIL PROTECTED]>:

This is a proposal to incubate
http://wiki.apache.org/incubator/Web20KitProposal

We're looking for a couple more mentors.

Web20Kit

Abstract
Web20Kit is a web 2.0 toolkit to help developers evaluate the

suitability,

functionality and performance of various web technologies by implementing

a

reasonably complex application in several different technologies.

Proposal

performance, and scalability of popular web technologies. Multiple

functionality but staying true to the philosophy of its base
language/framework.

Background

PHP, Ruby on Rails, and Java EE to develop their applications.

Web20Kit will develop an example application to understand the benefits, implementations of the application are planned - each providing the same Most web 2.0 sites today use open source languages and frameworks such as

Deployments

of these applications also use popular open source servers such as Apache

httpd, Tomcat, MySQL, Memcache, and Glassfish. Many other

gaining popularity.

servers/technologies such as lighttpd, mogileFS, mongrels, JRuby are also With the myriad technologies available, it is not easy to understand how they differ, especially in terms of performance and scalability. With

varied

levels of documentation available for some open source applications, it

is

also quite difficult for a web 2.0 startup to understand the correct

usage

of these technologies so that they don't become a bottleneck as their

site

grows.

Rationale


What it does

Web20Kit defines an example web 2.0 application (the initial

Web2.0kit is a toolkit that will attempt to address the above issues.

implementation

uses an events site somewhat like yahoo.com/upcoming) and provides three implementations: PHP, Java EE, and Ruby on Rails. The toolkit will also define ways to drive load against the application in order to measure

performance.


others.

What you can learn from it

a) Understand how to use various web 2.0 technologies such as AJAX,

As developers join the project, they can implement the same application using their favorite web frameworks and compare their implementations to memcached, mogileFS etc. in the creation of your own application. Use the code in the application to understand the subtle complexities involved

and

how to get around issues with these technologies.

b) Evaluate the differences in the implementations: PHP, Ruby on Rails,

Java

EE, and other contributed implementations to understand which might best

work for your situation.

c) Within each language implementation, evaluate different infrastructure technologies by changing the servers used (e.g: apache vs lighttpd, MySQL

vs

PostgreSQL, Ruby vs Jruby etc.)


scalability of the chosen platform.

e) Experiment with different algorithms (e.g. memcache locking, a

d) Drive load against the application to evaluate the performance and

different

DB access API) by replacing portions of code in the application.

A robust, community-developed standard implementations of a web 2.0
application using different technologies will enable developers to

compare

and contrast these technologies in a manner that does not exist today. By providing excellent sample implementations of a concrete application that

is

available to everyone, we will enable faster and easier application

proposal, we encourage others to come up with many more using other

development for users. Although we list three implementations in this

language

stacks and/or frameworks e.g. Spring framework, Python etc.

Current Status


Meritocracy

because the initial developers want to encourage this style of

This is a new project with some sample not-ready-for-prime-time code. The initial developers are very familiar with meritocratic open source development, both at Apache and elsewhere. Apache was chosen specifically

development

for the project.

Community
Web20Kit seeks to create developer and user communities during

incubation.

Core Developers


Alignment
The developers of the Web20Kit want to work with the Apache Software

foundation and set of practices for community-based development.

Known RisksOrphaned products
This project has a lot of enthusiasm among the core developers, has

The initial core developers are Sun Microsystems, Inc. employees, and faculty and students at UC Berkeley. We hope to expand this very quickly. Foundation specifically because Apache has proven to provide a strong

ongoing

development, and is not orphaned.

Inexperience with Open Source

practices.

Homogenous Developers

The initial developers are well-versed in open source methodologies and The initial group of developers is from two organizations. We would like

to

expand this and that is a primary reason for bringing this project to

Apache.

Reliance on Salaried Developers
Although part of the initial development team are students, the core
developers are employed by Sun Microsystems.

Relationships with Other Apache Products

None in particular, except that Apache HTTPD is the most common place to

run

PHP, and which the initial PHP implementation uses.

A Excessive Fascination with the Apache Brand

We believe in the processes, systems, and framework Apache has put in

place.

The brand is nice, but is not why we wish to come to Apache.

DocumentationInitial Source

Sun Microsystems Inc. intends to donate code for their PHP implementation

of

the sample events application as well as code to drive load against the application. UC Berkeley intends to donate code for the Ruby on Rails

implementation.


starting place for a much more robust, community- developed

This code is still a work in progress and will be provided primarily as a

implementation.

External DependenciesRequired Resources
Developer mailing lists
[EMAIL PROTECTED]
[EMAIL PROTECTED]
[EMAIL PROTECTED]

A subversion repository

A JIRA issue tracker

Initial Committers
      •
Akara Sucharitakul <
[EMAIL PROTECTED]> Shanti Subramanyam <
[EMAIL PROTECTED]> Binu John <
[EMAIL PROTECTED]> Kim Lichong <
[EMAIL PROTECTED]> William Sobel <
[EMAIL PROTECTED]> Arthur Klepchukov <
[EMAIL PROTECTED]> Craig Russell <
[EMAIL PROTECTED]>
SponsorsChampion
      •
Craig Russell <
[EMAIL PROTECTED]>
Nominated Mentors
      •
Craig Russell <
[EMAIL PROTECTED]>
Sponsoring Entity
The Apache Incubator.

Craig L Russell
Architect, Sun Java Enterprise System http://java.sun.com/products/jdo
408 276-5638 mailto:[EMAIL PROTECTED]
P.S. A good JDO? O, Gasp!

--
Brett Porter
Blog: http://blogs.exist.com/bporter/

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Craig L Russell
Architect, Sun Java Enterprise System http://java.sun.com/products/jdo
408 276-5638 mailto:[EMAIL PROTECTED]
P.S. A good JDO? O, Gasp!

smime.p7s
Description: S/MIME cryptographic signature