I am not going to dwell on the downsides of this event, but I have never in my life felt better about going to WordCamps and DrupalCamps than I did after this event. I was there for the ‘CMS Summit’ part, which had some good talks and speakers. I knew several of the attendees and met a few new folks too. It was actually really nice to talk to Joomla devs about the advancement and use cases for different CMSes from their perspective. Most of the people there were for the ‘Domain Event’. This is the conference for people who buy Top Level Domains in bulk in hopes of selling them to the highest bidder. I gotta say I really feel these were my people, vertical-wise or otherwise.
So why do I feel lucky to go to WordCamps and DrupalCamps? To sum it up in a few points. I get to learn about the awesome tech underneath a thing and not just the top level benefits from marketing folks. The Camp system believes it should be affordable for all and provide lunch, snacks and coffee/tea, or are at least clear on options. And maybe the largest thing is Camp folks are my people, in spirit and in belief in FOSS. The TLD folks are not against FOSS but they sure don’t speak the same language as those who embrace it.
I did go to a few sessions and took some notes. But honestly, the only two I attended that I think are worthy of talking about are discussed below. This is not to say there were not great other talks, just I didn’t see them. For instance, I know that the awesome Steve Burge, author of Drupal 8 Explained and many other great titles was there but I missed his talk.
CONVERSION RATE OPTIMIZATION 101
Chris Edwards
I got into this session, the first one I saw after my first trip to the registration desk, a little late. I love SEO talks for a couple reasons, neither of which are because I actually do anything with it. First, it is a glimpse into applying data science as applied to marketing, which is just fascinating to me. Second, as internet development moves towards more and more services connected together, it is important to stay up on trends around automation points and look for new and interesting amalgamations. It was a great talk, as Chris is a real wealth of info on this subject and a great presenter.
Raw Notes:
Quantitive data
Google Analytics
Not Free
AB Testing tools integrate
Mixpanel
Kissmetrics
Want to be building a funnel
HotJar not free but easy to buld funnels
Collect Qualitative data
On page surveys jumping in and asking how to improve the product
Use heat maps
(have models looking at product, not you!)
Watch your visitors
!record them!
I agree to terms moved RIGHT above checkout button improved sales
made it more accessible based on heat maps of mouse movements
Saved “Lost Visitors”
Optimizely
Test everything
Obama was the master of this with his data scientists
so did Trumps
Test your money pages
at least 5000 users per variation on testing,
might need more
A/B Significance test
Double check amazing! (larger than expected, rethink it)
Usertesting.com if you want to pay for it
Hotjar has a way to ask for test users
James added A/B testing with Hotjar for free, worth the money
upped conversions!
TECHNICAL DEBT IS KILLING YOUR SOFTWARE TEAM
Chad Windnagle
Chad comes to CMSSummit from the Joomla world, though his knowledge base is far wider and seems pretty deep as well. This talk was about a subject I get to talk to a lot of agencies about, though I rarely mention it by it’s proper name. Mostly we are talking about process improvements rather than application code improvements but the same underlying principals apply. Chad’s great quadrant diagram captures the overall idea in a way I had not heard discussed before. There are times you can be Reckless and Deliberate on a project, making an agreement that you will meet a temporary deadline but ultimately is the wrong way to do it, and set plans to fix it later. If that does not meet the definition of reckless for a production website workflow, then I don’t know what it means.
Raw Notes:
Was writing new code, then had to take over a Perl project
that was released in 1987
Changes over time
legacy application cleanup
@WardCunningham invented the word Techinical Debt to cover what devs have to do to change a running application
Tech debt is complex
DUplicate code
Inconsistent code
Parsing Errors
Slow code
Inflexible
Dead Code – Commented out code that know one knows why it is there
Careless debt and careless management (need to ask for reasonable mothodology)
participate in the team, don’t just inject
Debt Quadrant – Martin Fowler
Reckless | Prudent
Deliberate |
______________|_________________
Inadvertent |
|
Fixing?
Strict Logging
Code Analysis
IDE code quality
Code Quality Gatekeepers – SonorCube
Talk to developers (they know where bad code is)
Rebuild or Refactor?
Early on you always want to rebuild
refactoring though is needed
refactoring effectively
If you have to rebuilding successfully
Stranger system approach (like strangler figs)
wrap the new code around the old code to replace the old system over time
Got to increase velocity
I gave two talks
Every project is a story: Applying storytelling to your client interactions
WP-CLI: Don’t Fear The Command Line
All in all, I am very glad to be going back to Orlando in a few weeks for WordCamp. As for MERGE! well, I learned a lot from it.