Ubuntu Global Jam - Lessons I Learned

Ubuntu Cupcakes from Stacy Skaggs

The Ubuntu Global Jam was last Friday through Sunday and on Friday some Florida LoCo members got together at Chris Johnston's house in Orlando.  I've never been so I thought I would go, though it was on a Friday I could use the break.

I learned quite a few little things to make life easier working on Ubuntu Websites.  The first is a way of keeping the same development environments across project branches.  This has always been an issue for me because with bzr you can't just bzr checkout branch-name and the code changes like git.  Thats not its workflow.  Michael Hall pointed me to this post which is the method he uses for all his bzr repositories.  Basically, create a blank repository.  Inside the repository you create a trunk with the code from launchpad.  Then you branch the local copy into your feature/bug/issue branch.  Finally you lightweight checkout that branch to a working repo folder.  Sound confusing?  It does at first.  First off lightweight checkout doesn't import the code, just links to it.  Meaning when you modify code, push code, commit code its happening from the named branch not from work.  When you create your development environment around work, those files reference the work directory.  So when you change to a different feature, bug or issue the environment files still point to work and you don't have to rebuild.  Example commands:

bzr init-repo project-name
cd project-name
bzr branch lp:launchpad-bzr-name trunk
bzr branch trunk issue-specific-name
bzr checkout --lightweight issue-specific-name work

The second thing I learned was virtualenv for python.

sudo apt-get install python-virtualenv 
sudo apt-get install virtualenvwrapper

These separate the required environment for the ubuntu community websites from the installation of python you have installed and keeps everything nice and clean.  Now little did I know I was using this sort of with the make commands, however, with knowledge comes control and understanding.

I also got to see a bit of python dbugging as Michael Hall and Chris Johnston worked on that for a bit. Ate some fantastic cupcakes, and had a great time getting to dedicate some time to Ubuntu with some pretty awesome people.

Whose to blame for the misunderstanding ...

The incident I refer to is one happening more and more at work (and not just for me).  Examine the scenario:

A junior engineer has been given a feature to develop.  A senior engineer has built the basic mockup of the page, hooked it into the framework, but as the engineer attempts he runs into issues.  He asks, through email, how do I save to the database.  Its a project from a different development group, but the senior engineer has been through their code so he suggest looking into one part of the site and view specific controller and view files.  The junior engineer acknowledged and life moves on.  A week later the junior engineer comes to the senior engineer and says he is still having issues, so they open his code.  The junior engineer has created a new file (instead of following editing the files already apart of the application structure) and is attempting to drive logic that belongs in the controller in a file in the view.  He has ignored the structure of the site and is basically fighting against the natural flow of the program.  At first glance I'd say this is a lack of understanding of MVC.  However, this is not the first MVC project for this junior engineer.

Its a story playing itself over and over again in different ways.  At first glance, I'd assume the engineer never looked at the example files and project flow, instead they just looked up "PHP connect to database".  His code suggest that perhaps he looked at it, but didn't understand how the application uses the files.  Perhaps the senior engineer should have called the junior engineer over and shown them the files and explained the various methods in play to save and how they work with the view.

A friend of mine once mentioned a similar issue with contractors in the corporate landscape.  When he worked with them, over time he found himself practically programming the feature in the documentation to get the functionality he needed.  Inevitably they fought the existing application and built extraneous files.  At one point thier source had 4 different time classes that returned the time.

So how do we fix this without resorting to writing the program on a piece of paper and turning our junior engineers into a programming steno pool?  Is this a hiring issue, should we let junior engineers like this go?  A communication issue, did we not explain the solution completely should we have diagrammed it out?  A language barrier?  A training problem?  An experience issue?

At this time ... I'm lost, perhaps time will bring insight.

Passion Breeds Collaboration While Lack of it Breeds Indifference

I'll put the user in a box

Ever go to the doctor's office full of questions and ideas about a long term illness you may have?  Perhaps that's only me.

How did the doctor act?  Did he engage you in conversation?  Did he seem to ignore you?  Did he dismiss your questions?  Did he answer everything you asked and perhaps ask some of his own?

Have you ever been apart of an open source project?  On a mail list with other eager users and developers?  How did you react the last time someone sent in a question or suggestion?  Did you give them feedback?  Did you bypass the email based on subject?  Did you request more from them, perhaps code or testing?  Did you taunt them with your superior knowledge?

A friend of mine and I were talking about doctors and we came to the conclusion that the best doctors were the ones that wanted your input on your care.  It felt like it was because they were passionate about their jobs.  It seemed to us that the doctors who weren't either had lost the spark, or were only interested in medicine for the prestige, money, or perhaps family history (my father and his father and his father's father, etc).  That somehow led us to talking about open source projects and our time on one.  I came to realize that my passion for the project and for my career affected my involvement with the project.

When I was a developer I loved to hear from everyone.  I read every email, I was always on IRC.  As I moved into a manager role I became more remote.  I was hardly on IRC, I usually ignored listserv email.  Tickets from users were no longer wonderful feedback but annoying things that they could have fixed themselves.  My attitude dropped as I moved from that which I'm passionate about to something that I'm not.  So, I've stopped managing that project and tried to go back to engaging and coding.  The passion, its coming back, a bit slowly but only time will tell.

So the moral of this story?  When you are dealing with some a-hole professional don't be quick to judge.  Perhaps he made a wrong turn and is in a position that he has no passion for.  Instead of cursing him, say a little pray hoping he'll find his way to a profession he is passionate about. 

Engineer, Developer, Programmer, Analyst ... pick one?

I remember a while back my old boss mentioning he had chosen the working title of "Software Engineer" for positions at work because Engineers are not "web masters".  When I started this article I had been looking at positions on-line and I noticed a mix of definitions for the same work: Programmers, Software Developer, Software Engineer, Analyst.

From Dictonary.com

• Software Engineer - a person who designs and writes and tests computer programs 
• Software Developer - No Definition
• Analyst - a person who analyzes 
• Programmer - a person who writes computer programs.

Researching this post has lead me to the following conclusion.  A Software Engineer and a Software Developer are modern terminology for Analysts and a Programmers as Software as a field gets older and understands itself better.

Long ago we compartmentalized the field.  Analysts analyzed the needs of the customer and came up with the design then handed it off to programmers who actually created the application.  As the field has matured we learned a little bit about ourselves, we all like to design but secretly need to create too.  Software Engineers and Developers have come to embody a unified process of design, development, and implementation.  As you go up the technical ladder you design and mentor more than you write and test, but those things do not go away absolutely.

What is the difference between engineers and developers?  At first glance an engineer is a highly exact individual from a school of engineering.  However, in reality the difference is field of work.  Web companies, business groups, and application centered hardware agnostic positions tend to be described as Software Developers.  Technology companies and hardware centered positions tend to be described as Software Engineers.  As members of the software field both have to design the application to the users specification, write the application, and develop tests of that system.  Its the nature of the field.

Continuing Education for a Programmer

CSE at UF

One of my junior engineers asked the other day for resources to continue building up his experience with web development over the Christmas break.  So I put together a length email detailing items that I thought would be good for my team to read and do.  After re-reading the email, I realized that this was perfect blog material, so here it is un-cut from my original email.

Web Programming
The piece that makes web programming different from say standard gui applications is the web itself.  You have HTML and CSS and javascript on top of frameworks and server languages and databases.  Its a lot of different angles to become proficient in.  Start small and work your way in.  Some resources for learning more about the web.

• CSS Zen Garden - Why do we use CSS?  This is why, a perfectly structured site that can look completely different with the change in a style sheet.
• w3 schools - The W3 decides web standards, though browsers are sometimes lacking in following them.  They have a ton of tutorials on their site, check it out.

Computer Science Reading
Becoming better programmers doesn't always involve reading books about programming.  It could be a book about a developer, or a book about a methodology.  Some books you all should consider reading over your break

• The Cathedral and the Baazar - Ever heard of Eric S Raymond?  This essay talks about the differences between Open Source and traditional development.  The book contains his original essay and other articles he has written, you can find all of his articles on his site.
  
• The Art of Unix Programming - Another ESR Book about the philosophies of decisions made in building unix.  It can get preachy at times, but push through you'll be glad you did.
• The Mythical Man Month - Want to be a project or technical manager some day?  Read this book.  Its probably been suggested in class and if you were like me, you ignored it as you had three other classes suggesting 6 books a piece just like it.
  
• Just for Fun - The autobiography of Linus Torvolds.  It is not a technical book at all, but a very good read.
  
• The Design of Design - another Fred Brooks book, this one I have yet to start but it has excellent reviews and reading the overview it sounded like a good read for anyone in Software Development.
  
• Design Patterns - Very interesting book for those who love algorithm and technical design.  If you've taken the Oo class its required reading.

You should also consider adding to your weekly "programming/developer fitness program" reading blogs by programmers.  A few on my reading list are

• Joel on Software - Joel Spolsky has worked all over the map in software development, from programmer to manager to tech writer.  He isn't as prolific anymore, but there are some really great articles in his archive.  He is one half of the minds behind Stack Overflow and created Trello
  
• Coding Horror - Jeff Atwood another almost former blogger he has been picking up momentum again recently.  He is the other half behind Stack Overflow.
• Mark Shuttleworth  - Started following his blog after last years UDS, he has some interesting posts on a variety of subjects.
  

Those are the individuals I follow currently, though my actual list is quite larger I've stopped reading some of my .Net blogs since joining CTRIP.  Another reading strategy is to follow planets that aggregate blogs.  A few planets to think about

• Ubuntu
• Gnome
• Java.org
• Python
• Perl
  
• Apache
• Linux Kernal

If you have been around technology for a while you may already have a couple of technology sites that you read ever day.  For me its

• The Linux Journal  - the original linux magazine
  
• LWN - from NickS its another source of linux news
  
• Ars Technica - A great aggregator of technical and science news
  
• Wired - Its wired the magazine on the web
  
• Slashdot - another technology news website

"Google" New Ideas in Software Development
I keep throwing out things like Code Review, Gerrit, and Jenkins during our standups.  These are things that are coming down the pipeline at CTRIP, but why are they coming at all?  Being a better developer involves immersing yourself in new technologies and methodologies and being able to distinguish the fads from the game changers.  Use Google and Wikipedia to your advantage and look up things you are hearing in class and around the office.  Some key terms to "google" that have been floating around lately

• Agile Development
• KanBan
• Bazaar
• Gerrit
• Code Review
• Testing (there are different kinds than just Unit, Integration, and Acceptance)
• Jenkins
• HTML 5
• SPARQL
• Semantic Web
• Linked Open Data
• Bio-Medical Informatics vs BioInformatics
• Clinical Trial
• Personalized Medicine
• Translational Science

Try other frameworks!
You may notice that I often give you all(my team as this was originally written for just them) a vague idea of how it should work and then expect you to go find the actual solution.  Why is that?  Its because I have experience in other frameworks and know how to do what we are talking about in them, but perhaps not exactly in grails.  Learning additional web frameworks and in other languages will often connect concepts in your primary framework that you haven't already grasped.  Think of it as looking at web programming from a different angle.  The perspective is different so there are differences, but their are also similarities.  A couple of good frameworks to get started in:

• CakePHP (this is our other framework at CTRIP/ESE)
• Django - a python framework similar to cake and grails
• ASP.NET MVC 3 - a complete 180 from our usual style but not a bad idea to understand a completely different viewpoint.  As students you should be able to get VS 2008 for free or use Visual Web Developer Express.  Use C# as its closest to Java.  Just be careful of the differences like toString() which in C# is ToString() ... very annoying.
  
• Spring - one of the layers that Grails is built upon.  Its a good idea to learn Spring without Grails to better understand Grails itself
  

Learning is often doing, you can't just read it from a book and expect to be able to do it.  Find tutorials in these frameworks and take them for a test drive.  You'll learn a lot about web development in the process.

IDE's: is there a tool for every occasion?

IDE, Integrated Development Environment, a programmer's best friend.  The best ones run unit tests, pre-compile code, do code completion, and run the code in a demo environment.

The value of a second pair of eyes

A long time a go in an office far far away I had to sit for four hours a day watching someone else write code.  It was slow, it was boring, and it drove me nuts.  It was pair programming and in retrospect I've noticed that it did make both of us better programmers and the code itself better.  I often get little reminders of one of the important reasons for pair programming.

A second set of eyes

This weeks example, I received an IRC message asking if I could help out with a CSS issue.  I'm not a "CSS guy" I'm more a "CSS thief" but I was game.  So, we looked over his Django template and a solution eventually presented itself.  I mocked it up, uploaded a patch and he merged it in.  Great!  Now he states if he could only figure out one other little issue.  

I couldn't help myself, I was in programming mode.  I asked "what was wrong?"  A ManyToMany object wasn't printing out on his page and had him stumped.  It stumped me to, I barely know Django and after going through all the documentation everything looked correct.  I threw in a "{{ forloop.counter }}" and nothing showed up.  Found sites that were doing exactly what he was trying to do and they matched.  Finally I saw he had one little element "related_name" in the foriegnkey attribute.  I had read about it, and it wasn't necessary so I thought "why not try removing it and seeing what happens."  His code worked.  Code he and then I had poured over for hours that was technically correct and one little "un-necessary" element was the cause.