DrupalCamp Colorado 2014: Day 1 Session Notes - Part 1

The following are my notes from the sessions I attended on my first day at DrupalCamp Colorado. Don't take these notes as direct quotes from either the presenter or the lecture as I paraphrased a lot of it and also my knowledge of what was being presented may have been limited (such as in the Front-End Ops session).

Session 1: Solution Architecture - designing a strategy for project success

Presenter: Diana Dupuis from Amazee Labs @dianadupuis

Diana spoke about what is solution architecture, primarily focusing on the solution aspect since the architecture is the easy and more straight forward aspect.

Strategy is… 

• Intuition
• Wisdom and experience
• Logic and common sense
• Art and poetry

When applied to application development, an engineering process (constantly reinvented) sometimes called solution architecture - attempts to define an ecosystem that will organically produce the stated, measurable goals.

Key Elements are..

• Truth and Reality - aka bullshit detector
• Measurable Goals, Prioritized 
• Overcome hidden obstacles implicitly (problems with client or project that need to be laid out and addressed immediately in order for things to go smoothly)
• A construct (the architecture itself) in the form of a document (the easy part, the solution is the large and more important aspect)

Things to look out for..

• Are we ripping the fabric of time and space - the outcomes totally beyond the reality, budget, delivery time (build me eBay for $25,000…. yah right!)
• Hidden complexity - migrating from a proprietary CMS, multilingual (not just content, but module outputs may not be wrapped in t functions)
• Too much or too little control
• No goals or features don’t match goals
• Leadership confusion (can focus on what but not how, person in charge has no technical experience)
• Truthiness: reality refuses to bow down to power (“just get it done”)

What to give back

• Truth
• Insights
• Facts
• Options

What not to give back

• Snark (“look how stupid you are”)
• False hope

“here's the car I want. Though I got this other car, for cheaper, i love it, it’s not the best car - the one i wanted. But this is the car that i can afford and it does exactly what I need it to to and I still love it.” - Diana Dupuis from Amazee Labs

Measurable Goals (examples)

• reduce calls to customer service by 25% in 3 months
• increase subscription to newsletter by 150% in 6 months
• Increase page view time from 3 seconds to 10 seconds on newly published content
• Increase traffic by 30% in 6 months (how does the application do that? Social media?)

Hidden Obstacles

• they are *always* there
• A good strategy overcomes them
• ALES
• Ask - so why are you doing this project, how is the current site, what is the problem
• Listen
• Empathize 
• Strategize

Construct

• Overview: business reasons to do the project and clear, measurable goals
• The vision with just enough detail to be flexible (the art part) - sometimes looks like a project proposal, scaled as level of project (for $ you get this - meets the baseline of your goals, if you spend another $ you can get this)
• Drupal Domains
• Core and DevOps
• Design
• Basic, responsive theme
• Data architecture (not just fields, true relations and interacts) and migration
• Art and architecture: the features and/or however your concept is described
• Testing
• Project Management (10% of project usually)

Estimation

• Include assumptions

When off track

• let go of the original plan
• prioritize against goals
• create buckets and phases
• retrospect, include everyone
• define the known and unknowns

Session 2: Building A Scalable Private VMware Based Drupal Cloud

Presenters: Al Roberts and Ron Williams of University of Colorado

Al spoke about the internal storage and devices used to support their vmware cloud based Drupal installs. Stressing the importance of having a system that can scale to your clients needs. Thinking about traffic and client data and if your systems can handle more than what you actually need.

Ron Williams speaks to the web root architecture they utilizing on their servers. It’s constructed in a way to enhance performance and redundancy. They use a combination of multiple apache servers in order to allow failover.

Though these guys clearly know their stuff, and are very knowledgeable in creating a powerful infrastructure to handle a lot of sites, I was disappointed as I thought this session would be more focused on the how to implement versus how we implemented it which instead left me feeling lost on what I had heard over the the last 45 minutes.