Paulo Sousa

Updated IBM's Architect eKit

Paulo Sousa — Fri, 19 Sep 2008 22:23:00 GMT

IBM has an updated version of its "Architect eKit" which includes (among other things) Grady Booch podcast series and a webcast of Grady Booch on architecture.

97 Things Every Software Architect Should Know

Paulo Sousa — Wed, 20 Aug 2008 10:41:00 GMT

There's a new wiki on the cloud about architecture. The contributors (software architects) are trying to put together a list of 97 axioms for software architecture.

Go check it at http://97-things.near-time.net/wiki

All contributions will remain freely accessible on the site, but the 97 best axioms will be included in a book published by O'Reilly Media.

Pós-graduação em Engenharia de Aplicações Empresariais (2008/2009)

Paulo Sousa — Tue, 19 Aug 2008 15:08:00 GMT

Estão abertas as candidaturas para a Pós-graduação em Engenharia de Aplicações Empresariais para o ano lectivo 2008/2009.

O curso de pós-graduação em Engenharia de Aplicações Empresariais pretende:

Colmatar uma falha na formação de engenheiros informáticos ao nível da especificação de arquitecturas aplicacionais sistemas complexos.
Oferecer um conjunto de disciplinas que visa formar um engenheiro com competências de arquitecto de aplicações informáticas.
Fornecer uma formação sólida, actual, prática e pragmática que desde logo possa ser utilizada na vida profissional.

As candidaturas estão abertas até 15 de Setembro. O edital do curso pode ser consultado aqui.

Capturing Architectural Decisions

Paulo Sousa — Wed, 06 Aug 2008 08:54:00 GMT

IBM has started a series of articles related to Capturing Architectural Decisions at its developerWorks site.This first part of the series presents a domain meta model specifically designed to capture and share architectural decisions and outlines why and how architects should consciously identify, make, and enforce architectural decisions. It is supplemented with an Eclipse-based tool to support this metamodel and also the Architectural Decision Knowledge Wiki.

The main entities in the metamodel are:

ArchitecturalDecision - capturing a decision issue (records problem statement, decision drivers, background readings, etc.), e.g., layered architecture
ADAlternative - state different alternatives (pros and cons) for each architectural decision, e.g., 3 layers versus 4 layers
ADOutcome - the decision that has been made (recording a justification)

Part 2 will look at SOA examples and the role a reusable architectural decision model can take in SOA design and deployment.

The world is Flat: audiobook for free

Paulo Sousa — Fri, 01 Aug 2008 09:35:00 GMT

Until August 11th, Thomas Friedman is giving away the audiobook version of his The World is Flat bestseller.

The Enterprise of the Future

Paulo Sousa — Fri, 01 Aug 2008 09:03:00 GMT

IBM has released its Global CEO Survey 2008. According to 1000+ CEOs the company of tomorrow will be:

Hungry for Change - capable of changing quickly and successfully; not merely responding to trends, but shaping trends.
innovative Beyond Customer Imagination - surpasses the expectations of increasingly demanding customers by deep collaborative relationships.
Globally Integrated - business is strategically designed to access the best capabilities, knowledge and assets from wherever they reside in the world and apply them wherever required in the world.
Disruptive By Nature - radically challenges its business model, reinventing itself and its entire industry.
Genuine, Not Just Generous - go beyond philanthropy and compliance and reflect genuine concern for society in all actions and decisions.

Interestingly, IBM also has a companion report with the Implications for the CIO from this survey. There you can find that among the many roles a CIO will have to play are 1) Catalyst for change; 2) Expert on the “art of the possible”; and 3) Guru of risk management.CIOs will have to ensure that their IT team enables the prevously listed 5 characteristics of the company of tomorrow (or as IBM calls it "Enterprise of the Future"). This will require effective integration of business and technology perspectives and the creation of flexible and scalable technology systems and infrastructures. The report continues with specific 'advice' for each of the characteristic.

Hungry for Change - transforming IT applications, services and infrastructures to make them flexible, extensible and secure thru Well-managed enterprise architectures, Service-oriented architectures, Virtualizationand Multisourcing and shared services frameworks.
innovative Beyond Customer Imagination - enabling collaboration across and beyond the enterprise and turning data into insight by collaboration and social networking capabilities, Web 2.0 applications, and Context-specific, self-help portals.
Globally Integrated - moving toward a shared services model and enabling integrated business operations and supply chains thru Integrated supply chains, Integrated financial organizations, enterprise architecture and global management strategy, unified communications and collaboration capabilities, social networking tools, Strategic exploitation of global enterprise resource planning.
Disruptive By Nature - preparing to support the evolving business models of the enterprise, including acquisitions, mergers and divestitures by business process modeling, enterprisewide information architecture.
Genuine, Not Just Generous - working to reduce the IT footprint and associated energy use, and to support the enterprise’s broader social responsibilities.

And of course do all this with an ever increassing pressure to decrease budget spending.

CodeGeneration 2008

Paulo Sousa — Tue, 29 Jul 2008 10:04:00 GMT

The slides from CodeGeneration 2008 are available for download.

Open Research Scientist positions

Paulo Sousa — Thu, 24 Jul 2008 10:19:00 GMT

The CISTER/IPP-HURRAY Research Unit (www.hurray.isep.ipp.pt) invites applications for Research Scientist positions to develop research in the area of Real-Time and Embedded Computing Systems. This/These research positions will correspond to a 5-years research contract.

Area: Real-Time and Embedded Computing Systems
Application Deadline: 10, Sep, 2008
Send Application To: cister-info@isep.ipp.pt
Contact Person: Eduardo Tovar
Salary: at least 45000 EUR / year (before taxes)
Duration: 5-years research contract
Requirements: PhD in Computer Science, Electrical and Computer Engineering or related fields, with particular expertise in real-time computing systems or embedded systems. The candidates should also have had at least 3 years of post-doctoral experience (although this can be relaxed in some very specific cases) with an international-level publication record and demonstrated ability to do independent research. Fluency in written and spoken English is required.

SELECTION COMMITTEE:
Prof. Tarek Abdelzaher; University of Illinois at Urbana-Champaign, USA
Prof. Sanjoy Baruah, University of North Carolina at Chapel Hill, USA
Prof. Alan Burns, University of York, UK
Prof. Luis Miguel Pinho, Polytechnic Institute of Porto, Portugal
Prof. Raj Rajkumar, Carnegie Mellon University, USA
Prof. Eduardo Tovar, Polytechnic Institute of Porto, Portugal

Further information is available at:
http://www.hurray.isep.ipp.pt/asp/list_jobs2.asp

Product Line Engineering

Paulo Sousa — Mon, 21 Jul 2008 14:31:00 GMT

An interesting talk from Markus Volter at JAOO'07 about how to manage variability in a Software Product Line.

The Product Line Engineering blog also has an interesting post with a clarificaton on the notion of version and variant.

Software Product Lines

Paulo Sousa — Tue, 17 Jun 2008 13:57:00 GMT

Software Product Lines (SPL) are a way of managing the development process focusing on planned reuse thru a large number of products (a product family).

The Software Engineering Institute (SEI) defines a SPL as (emphasis mine):

A Software Product Line is a set of software-intensive systemssharing a common, managed set of features that satisfy the specific needs of a particular market segment or mission and that are developed from a common set of core assets in a prescribed way
[Clemens and Northrop 2001]

It's basically a concept about software industrialization, repeatable processes and strategic reuse of core assets (e.g., requirements, architecture, components, tests, ...) which try to capitalize on the commonality across scores of families of similar systems to predictably and efficiently achieve business and mission goals.

The main activities necessary to put in place a SPL are:

core asset development - the goal of the core asset development activity is to establish a production capability for products. this consist of developing reusable and configurable core assets (thru variability) as well as defining the production plan (i.e., the way an asset can be used in a product)
product development - product builders use the core assets, in accordance with the production plan, to produce products that meet their respective requirements. Product builders also have an obligation to give feedback on any problems or deficiencies encountered with the core assets, so the core asset base remains healthy and viable.
management - Activities must be given resources, coordinated, supervised both at the technical and organisational level. Technical management oversees the core asset development group and the product development group. Organizational management is the final authority for the success or failure of SPL and deals with the external interfaces.

Beneath the level of these essential activities are essential practices that fall into practice areas (i.e., a body of work or a collection of activities that an organization must master to successfully carry out the essential work of a product line). The SEI SPL Framework defines 29 practice areas that must be mastered to deploy a sucesseful SPL arranged in 3 categories: software engineering, technical management and organisational management.

There is a lot of costs associated with an SPL effort, such as:

Architecture - Must support variation inherent in the product line (Requires greater talent to define)
Software Components - Must be designed to be general without a loss of performance; must build in support for variation points
Test Plans, Test Cases, Test Data - Must consider variation points and multiple instances of the product line
Business Case and Market Analysis - Must address a family of software products, not just one product
Project Plans - Must be generic or be made extensible to accommodate product variations
Tools and Processes - Must be more robust
People, Skills, Training - Must involve training and expertise centered around the assets and procedures associated with the product line
Requirements - Sophisticated analysis and intense negotiations; Common req. and variation points for the whole system

Several organizations who have already implemented SPL have reported the following beneffits:

Improved productivity - by as much as 10x
Increased quality - by as much as 10x
Decreased cost - by as much as 60%
Decreased labor needs - by as much as 87%
Decreased time to market (to field, to launch...) - by as much as 98%
Ability to move into new markets - in months, not years

In essence this about management of the software bussiness and not a technological "thing". SPL epitomize the concept of strategic, planned reuse to achieve economies of scope (one in which many systems are similar but not by accident, but because they were thought to).

You can find a presentation I've made on this issue at slideshare.

| View | Upload your own

References

Paul Clements, Linda Northrop (2001) Software Product Lines: Practices and Patterns.
Addison-Wesley Professional.

Patterns of Enterprise Application Architecture by Example

Paulo Sousa — Tue, 10 Jun 2008 15:20:00 GMT

A couple of weeks ago I gave a one day seminar titled "Patterns of Enterprise Application by Example" in the context of a teacher mobility program (Erasmus) at the Glasgow Caledonian University (Scotland, UK). The presentation is available thru slideshare.

| View | Upload your own

You can also download the companion code.

Book review: Rules for revolutionaries

Paulo Sousa — Sun, 01 Jun 2008 09:49:00 GMT

Rules for Revolutionaries, Guy Kawasaki, Michele Moreno (2000) ISBN 0-88730-995-X

In this book Guy Kawasaki talks about creation and marketing of great products. This book was written around 2K and the influence of the .com bubble is quite present; Guy calls it "revolutionary times" and the central idea of the book is about creating revolutionary products not just great products. The book keeps the focused, bulleted, pragmatic way of writing Guy has used in other previous books, with lots of "exercices" and funny asides.

The book is divided in three parts: Create Like a God, Command Like a King, Work Like a Slave, which pretty much align with different phases of the development cycle. The first part, Create Like a God, addresses the basics of creating truly revolutionary products by advising to think differently (or cogita differenter as the original Latin chapter's title), do not try to achieve perfection at first ("don't worry be crappy"), market early and listen to feedback ("churn, baby, churn") to improve the product. The second part (Command Like a King) is about achieving market awareness, chrossing the chasm, make sure the product leaves the early adopters niche and gets to as much users as possible. There is a chapter on how to avoid "death magnets" that is fundamental reading to anyone trying to market a product. The third part, "Work Like a Slave", is about making sure the revolution doesn't end. By "eating like a bird and pooping like an elephant"

There is a PDF presentation for the book avaliable thru Guy's site.

My Review: 4 out of 5

SEI's Architecture Expert tool

Paulo Sousa — Thu, 29 May 2008 20:19:00 GMT

The Software Engineering Institute's is well know by its contributions to the field of Software Enginbeering in general and Software Architecture in particular. In the resource area for architects you can find lots of information but also a tool to help software architects doing their business: ArchE.

The Architecture Expert (ArchE) tool is an assistant to the architect. ArchE embodies quality attribute theories, techniques for ascertaining the appropriate quality attribute model from an architectural description, techniques for solving the quality attribute model to determine a predicted response in given situations, and the ability to use legacy designs as input. [SEI website]

You can download and try it; Seams very interesting. There is also a presentation available.

IntentionalSoftware @ JAOO

Paulo Sousa — Wed, 28 May 2008 15:14:00 GMT

InfoQ has published a video by Charles Simonyi at JAOO.

Ads in our dreams !?

Paulo Sousa — Tue, 01 Apr 2008 17:03:00 GMT

If this is true, it is kind of scary...

Google DreamAds

[...] Essentially DreamAds is Google's latest plan to extend AdSense - this time into our dreams. It may sound far-fetched, but [...] it is possible thanks to the latest in cheap magnetic resonance scanning and mind-reading techniques developed at Stanford University. [...] Google is aiming to become the first company to monetize a totally wasted period of time in human life: sleep, which takes up almost 1/3 of our lives.

From: http://www.readwriteweb.com/archives/google_dreamads.php