GravityZoo: Bringing Your Desktop Applications To The Internet As A Service
Category: | Business |
---|---|
Keywords: | cpython, ubiquitous, distributed |
Title: | GravityZoo: Bringing Your Desktop Applications To The Internet As A Service |
Authors: | Mahdi Abdulrazak and Marcel von Birgelen |
Contact: | mahdi.a@gravityzoo.com |
Date: | 2007/05/01 23:08:00 |
Web Site: | GravityZoo |
Summary: | Python is the language of choice for the development of the GravityZoo ubiquitous computing framework. |
Logo: |
Introduction
The GravityZoo Company, founded in late 2005 is a pioneering innovator of Software as a Service (SaaS) enabling technology, Ubiquitous Computing and Ambient Intelligence Systems. The core technology of the Company is its framework. This GravityZoo Framework is designed to develop and host any kind of application while simultaneously making these applications available at anytime, anyplace. The latter on any type of client device, be it a PC, a mobile client like a smart phone or an embedded device like the microwave in your kitchen, whereby the application has become device aware.
An essential part of the GravityZoo Framework is its sophisticated object sharing infrastructure. It is within this infrastructure that object-updates "generated" by the applications are shared with the devices connected to The GravityZoo Framework. (each device running a GZF Client). Regarding these Shared objects one can think of both simple User Interface objects or complex objects related to for example the operations of a nuclear reactor. In both instances these objects are handled identically by the core of the GravityZoo Framework. Identically, because The GravityZoo Framework is designed to replicate the state changes of all those "shared objects" between client and server.
GravityZoo Framework
The GravityZoo Framework consists of several layers with each a distinctive function.
Client Side
The client runs a small software stack. This stack consists of a core and one or more toolkits. The toolkits form the interface between the Operating System of the client and the Client core.
Backend
The backend does the actual application processing and stores most of the application data and the entire state of the applications. The backend therefore is the place where the real action takes place.
The REP routing infrastructure enables the messaging between the client(s) and the backend and the communication between clients. The REP protocol used is the workhorse of the GravityZoo Framework. It is designed to cope with the limitations of the current common Internet connections.
The application Processing infrastructure is responsible for the execution of the applications itself and consists of at least one server with an Application Server role.
The Storage infrastructure stores the user data and the application state. The Storage infrastructure is known as the Central Object Store (COS) which is intended as a huge intelligent "object database".
The Development As A Service (DAAS)
DaaS is a newly developed GravityZoo concept allowing software developers to develop, maintain and test their applications at Anytime, Anyplace and on Anydevice. The DaaS environment at this stage only supports development in Python language. This can be done directly in the GravityZoo Development Environment, allowing you to easily test and develop small applications, online. In later stages, this environment will be expanded to a fully featured IDE. Also, we will offer easy conversion of existing projects developed in common IDEs like Visual Studio .Net. The GravityZoo Client at this stage allows you to access the DaaS environment and several demo application. Some of these applications will be release to public under an Open Source license.
Why Python?
The engineering team responsible for the initial prototyping was mainly Artificial Intelligence oriented. It consisted of professionals, graduated from the Maastricht University, most of them quite theoretical and with little to no actual programming experience. The project itself required a highly object oriented Programming Language able to deliver the features we required for our prototyping but also was easy to learn because of the lack of coding experience. Python was chosen because it matched the criteria set by the project while still easy to learn.
Although we considered Python perfect for the prototype phase, we anticipated that ultimately most of the prototypes would have to be rewritten into C and C++. However during the prototyping and alpha phase Python turned out to be more qualified for the job than initially expected i.e. Python turned out to be a true problem solver. At present most of the original Python code is still untouched or replaced by new Python code.
The Python Successes
Speed of development
For us, development in Python has been proven to be much more efficient. Our developers generally finish Python based projects much quicker than comparable projects in other languages like C, C++, C# or Java. Apart from the latter, Python turned out to be easy to learn for our developers, so easy that even new team members can fairly quickly start focusing on the project rather than learning Python.
Cross platform capabilities
Python is ported to most platforms that are relevant for our Python code base. This allows large pieces of the GravityZoo Framework to be ported to other platforms, with only small adaptations to the Python code base.
Object oriented architecture
The highly object oriented architecture of Python integrates easily with the core of the GravityZoo Framework, which is entirely object based. Python was quite capable of supporting all data structures needed in the development of the GravityZoo Framework.
The Open Source advantage
The fact that Python is open sourced works like an insurance policy for us. Although so far we have not yet run into serious problems, we know that we have the possibility to fix it.
Interfacing with other languages
Python allows us to easily interface to most other languages. This makes it easy to hook other code into our Python code base. Interfacing with other languages is a core aspect of the GravityZoo Framework.
Stability
One major concern in the beginning was the stability of Python. Until now, Python has been a very robust platform for much of our framework. Most bugs and problems were easily resolved.
Performance
The performance of Python was also a huge concern for us. During the process we learned a lot of the capabilities of Python. Fortunately, our concerns were largely proven false, the performance of Python has proven to be much better than we expected. For the remaining issues, we have learned to cope with some of Pythons implementation limitations and "design decisions". Also, the easy interface to other languages allows us to offload certain operations to other languages quite easily.
The Benefit For The Python Community
The GravityZoo Framework allows the Python developer to easily bring his/her existing and new Desktop applications to the Internet as a service. The only action required is to integrate the GUI layer without the need to compromise the object oriented method of development.
We firmly believe that this technological development will allow the Python language and its community to deliver an even greater contribution to the creation of tomorrows Information Society.
For more information, please visit our web site
About the Authors
Mahdi Abdulrazak is Chief GravityZoo Evangelist and also responsible for GravityZoo's Open Source strategy. Marcel von Birgelen is the CTO, lead architect and engineer of GravityZoo.