Tuesday, April 29, 2008
30th International Conference on Software Engineering (ICSE) - 40 years of Software Engineering
Saturday, April 26, 2008
RiDE: The RiSE Process for Domain Engineering
Here is the abstract of the work: Software reuse the process of creating software systems from existing software rather than building them from scratch - is a key aspect for improving quality and productivity in the software development. Quality can be improved by reusing all forms of proven experience, including products and processes, as well as quality and productivity models. Productivity can increase by using existing experience, rather than creating everything from scratch. However, this process is more effective when systematically planned and managed in the context of a specific domain, where application families share some functionality.
In this scenario, Domain Engineering (DE) - the activity of collecting, organizing, and storing past experience in building systems or parts of systems from a particular domain in the form of reusable assets has been seen as a facilitator to obtain the desired benefits. Nevertheless, the existing domain engineering processes present crucial problems, such as: they do not cover the three steps of domain engineering, for instance, domain analysis, domain design, and domain implementation; besides, they do not define activities, sub-activities, roles, inputs, outputs of each step in a systematic way.
The full text you can get here.
Thursday, April 24, 2008
RiSE’s Interviews: Episode 4 – Software Reuse with Dr. Charles Krueger
Wednesday, April 23, 2008
RiSE at Forum Internacional de Software Livre
Between April 17 and April 19 it was held in Porto Alegre/Brazil, the Forum Internacional de Software Livre (FISL). FISL happens every year, since 2000, and this year it achieved new record of attendants: 7,4 thousands. The event counted with important names of free/open source software, as John Maddog Hall, and big companies that invest and use free software as a differential: Intel, Google, Sun, Yahoo Brasil, among many others.
21 countries and all Brazil's states were represented there. The RiSE company , in cooperation with CESAR and CIn/UFPE, was represented too with Yguaratã Cavalcanti. Yguaratã presented the initial solution for bug reports duplication problem, which is part of his master thesis. The complete paper can be found in FISL proceedings.
The event was characterized by the diversity of themes presented in the talks, such us: governance, adaptation of free software, digital inclusion, mini courses, innovation guided by free/open souce software etc. Furthermore, 24% o the attendants were composed by students and 63% by professionals or IT employers. These numbers also demonstrate the importance that free/open source softwares are holding by companies: companies that provide IT solutions based on this market, or companies that use or intend to use free software based solutions.
The event leave us the conviction that business based on this market is here to stay for long, and if IT companies intend to keep playing they need to learn how to copy with free/open source software market/business. And it is also a good choice for startups.
Saturday, April 19, 2008
21st IEEE Conference on Software Engineering Education and Training (CSEE&T)
About CSEE&T, it was very good also. It is a good conference to discuss about software engineering education and training in general. There, we had discussions about games in software engineering, agiles and formal methods, etc. The keynote speakers included Bertrand Meyer and Watts Humphrey.
I was there presenting part of our experience teaching software reuse with the paper: A Case Study in Software Product Lines: An Educational Experience. It, in general, was well received by the software engineering education community with some questions. Next year, the conference will be in
Monday, April 14, 2008
Microsoft Research Fellows - RiSE's member is there
One more time, it is a day to celebrate in RiSE. One member of our staff, Daniel Lucredio, was awarded to Microsoft Research. Daniel is going to
Good luck there, Daniel.
Monday, April 7, 2008
6th RiSE Day
The first section was devoted to the concluding work. We discussed some aspects related to reuse adoption {motivation, problems, goals and the framework proposed as solution}; new research trends in the LiFT - Legacy Information retrieval Tool (link in portuguese); the current state of the ToolDay, a tool for Domain Analysis (link in portuguese); set of work related to search and retrieval reusable assets {such as, Data Mining, Semantic, Ontology, Context-aware}; and, a systematic architecture method for SOA-based enterprise applications.
The second section covered work related to Software Product Lines such as: requirement engineering for SPL; domain design methods; core assets development in SPL (using the Mobile domain); and, aspects related to evolution and maintenance in SPL. This section finished with a discussion about an effective component testing approach.
The last section was a mix of other research trends, such as: new approaches for improve the reusable assets search and retrieval; duplicate bug and change requests detection; and a cost model for SPL.
This RiSE Day showed that the group is covering different aspects {technological, organizational, business and financial} related to software reuse {introduction, evolution and maintenance}.
In the 6th edition of the RISE Day, we had about 30 participants, from other groups of the CIn-UFPE, UPE and others universities; software companies like SERPRO, Pitang, C.E.S.A.R, Porto Digital among others.
We would like to thank all RiSE members, to dedicate your time to discuss a lot of work and help in disseminate the reuse practices in the Recife software environment.
Thank you guys! See you in the 7th RiSE Day.
* UPDATED *
See all pictures here !
Sunday, April 6, 2008
Proactive Bug Reports Discover
The developers behind PyGame have introduced a proactive way to capture bugs discussion that have not reported to the PyGame mailing list [yeah, they use email list to report bugs]. The motivation is the fact that users, sometimes, associate the PyGame's problems to the operational system is running, which lead them to report the problem to the OS's bug tracker, or simply discuss about PyGame's problems in personal blogs or whatever.
The method proposed in PyGame's blog is interesting and consists in searching into specific places [sites, mailing lists etc] to find bug reports related to PyGame. However, some questions about viability and scalability rise, such as: what about the cost to maintain crawlers searching for bug discussion in the web? How to decide if a bug discussion is really relevant or if it is really a bug? etc.
For PyGame's developers, this solution is reasonable because they have specifics web sites, mailing lists, and CR tracking systems, where the searches must be performed. But, for widely used softwares, such as Firefox web browser, the technique might be very costly, or even impracticable.