Skip to main content

The importance of continuous integration

Leading a team of developers in the effort of building a robust, quality software product should involve the establishment of some process and tools to assist the team effort and serve as a safety net for the errors of getting people to work together. Continuous integration is, I believe, a crucial element of that process. Introduced by Martin Fowler and Matt Foemmel (see article Continuous Integration), continuous integration establishes the practice of frequent integration of work developed by the several team members verified by automated build and testing of integrated code within a clean sandbox. This practice is valuable for several reasons:
  • It promotes the development of a clear process of building/deployment independent of any specificity of developer's platforms. Code that exists on a single platform only is bound to become dependent on specific aspects of that platform without anyone really noticing the dependencies until trying to port to other platforms. The existence of the integration clean sandbox allows these specific dependencies issues to not go unnoticed.
  • It promotes the development of testing. Being based on the premise of "test often" it makes the testing development part of the team's process. The fact that there is a platform built specifically for build and testing verification transforms, from the developer's perspective, testing efforts into an even more useful and justifiable effort.
  • It allows for quick detection of code integration issues by providing the clean slate for bringing all the code together. The little quirks of code combining can be detected by effective smoke/regression testing.
How hard is it to put in place continuous integration ? It depends a lot on where you are in your development process when you decide to take it up. If your team already makes use of build tools (e.g. Ant, Maven, even Make), makes use of a version control system (e.g. CVS, Subversion) and already does some kind of automated testing (e.g. xUnit) it can become pretty straightforward (You do use these don't you?!?). Continuous integration becomes simply a matter of setting up the integration sandbox and establishing the automation to detect changes in the version control system, building the changed system and testing the changes. You do not even need to go very far to accomplish that. You can find already systems that give you the continuous integration functionality that you need. At Tizra we have chosen to use the Cruise Control open-source, free framework for our continuous integration process (BSD-style license). It does pretty much what we need to do and keeps us on top of any integration issue that might arise from our development effort. It provides us unit tests results gathering, reporting and historical stats gathering for all checks. Continuous integration should be, in my opinion, an important practice for any serious software development company and I, for one, will make every effort to ensure it is put in place in every project I work with.

Comments

Popular posts from this blog

What Einstein Taught Us About Searching Inside Publications

When the Collected Papers of Albert Einstein went live on Tizra a few years ago, it was a huge step forward. Suddenly, anyone anywhere could search and access the output of one of the 20th Century’s great minds…from love letters to breakthrough articles that changed how we think about the nature of time and space. But the project also showed the limits of traditional tools for searching within large, complex publications. These limits sparked a collaboration with Princeton University Press and Einstein Papers Project editors, which this year resulted in a dynamic new search interface, which we’ll be demonstrating in a  Webcast Friday, December 15 at 1pm ET . The interface not only makes it easier for Einstein researchers to home in on relevant content on both mobile devices and desktops, it points the way toward faster, better searching within a wide range of publication types, from reference books to periodicals, technical documentation and standards to textbooks. Click To Re

Using XML to Create a Better Online Reading Experience for the American Payroll Association

Congrats to the American Payroll Association on their recent launch of XML-based publications on Tizra!  Thanks to this collaboration, APA's authoritative books for payroll professionals are now available in crisp, reflowable HTML, creating a user experience that feels like a truly digital native product, rather than a conversion from print. XML-based publishing also creates a better mobile reading experience, supports more precise search and navigation, and opens the door to better accessibility for users with low vision and other disabilities. Our partners at  Scribe  did a great job supporting APA through the process of producing the XML for loading into Tizra, and we’d definitely recommend them to anyone interested in such a transition. It’s hard to overstate what a big step forward this is for Tizra as a platform and a company. XML has long been planned for in the product's architecture, but now for the first time, we have a working example that demonstrates the pow

Behind the Screens, Pt. 1--Creating a Site With the New Tizra Publisher Control Panel

Now that it's public , we're excited to show the new Tizra Publisher web control panel in a bit more detail. To provide a real-world example, we'll show it in use building the eat.shop guides site for Cabazon Books, which went live a few weeks ago. While in practice the process is quick—with initial online selling capability available in a matter of minutes—there's a lot to the software, so we'll break it up over a few posts. 1. Upload a Document When you open your Tizra Publisher account, you're presented with the control panel homepage in your web browser. The cog dropdown provides quick access to key tasks from anywhere in the system. Start by using it to upload a PDF. In this case, it's the full 132 illustrated spreads for eat.shop nyc. Note how the progress bar informs you as the system imports the file, extracts metadata, breaks the PDF into individual pages, and indexes it for full-text searchability. Apart from the upload, which of cours