The SaaS Engineering class officially began on Monday. The first weeks assignment 10 video lectures, averaging 10 minutes and 2 Chapters from the textbook. Thankfully NO assignments or quizzes – YET – those start next week. First impressions: the book seems solid but I have not yet become an e-reader fan. The video lectures have suffered and only one has not required at least 2 restarts to get through. It is not surprising with 60,000 people or so accessing the site.
The review of Agile development vs. legacy or Waterfall development helped. I have only a few months experience with Agile processes. One of the points made was that using Ruby, the rich tool development tool set and Agile methods, SaaS providers can deliver changes every 2 weeks. My immediate reaction was BS. My long experience with SAP and corporate IT says that changes at that frequency will not be accepted. Given a day to reflect, I’m less definite. I can see room for minor tweaks and bug fixes being quite possible when most of the code resides in the cloud. Changes to core processes and calculations for running the business still will require a more deliberate schedule with business testing and sign off required.
I experienced another “Homer Simpson” d’oh moment. In the lecture introducing the concept of cloud services the example of Farmville resonated. Four days after Farmville was introduced there were 1 million users. In 9 months there were 75 million. If Zynga was using in-house computers, the delivery trucks would have been a non-ending line to keep up with demand. Instead the Amazon EC2 cloud scaled easily with demand.
And a final thought for this post. The book contained a profile of John McCarthy, inventor of Lisp. The profile included this quote:
If computers of the kind I have advocated become the computers of the future, then computing may someday be organized as a public utility just as the telephone system is a public utility …The computer utility could become the basis of a new and important industry. John McCarthy, at MIT centennial celebration in 1961
Patterson, David; Fox, Armando (2012-01-12). Engineering Long-Lasting Software: An Agile Approach Using SaaS and Cloud Computing, Alpha Edition (Kindle Locations 590-592). Strawberry Canyon LLC. Kindle Edition.
That my friends is a visionary statement.
If you know where I can grab a Cliff Notes version for this class, it would be awesome!
PS Also cool, when you cut and paste from the Kindle page, the reference information comes with it — keeps it legal, I like that.
Posted on February 22, 2012, in Home Page, SaaS_Daze Blogs. Bookmark the permalink. 3 Comments.
I think Facebook works on a one-week release cycle. But it’s possible to do much better than that 🙂 Start here: http://timothyfitz.wordpress.com/2009/02/10/continuous-deployment-at-imvu-doing-the-impossible-fifty-times-a-day/
Thank for reading and commenting. I do not disagree that a short development/release cycle is possible, especially for minor changes. I do have doubts about how that will work for enterprise applications, #SAP for example. Changes affecting financial or personnel data will need serious testing and business user sign-off. Then of course there may be legal requirements for change management processes.
Thanks pozorvlak for the link to IMVU. I have had chance to listen to Eric Reiss’s (co founder of IMVU) views on continuous deployment and it sounds like continuous deployment is one of the key to success for startup’s. So, yeah i look forward to embracing this philosophy, And i like what Zuckerberg had to say “Stay focused, Ship fast” and is made possible by continuous deployment. And again, Thanks for the link..