Formed in 2009, the Archive Team (not to be confused with the archive.org Archive-It Team) is a rogue archivist collective dedicated to saving copies of rapidly dying or deleted websites for the sake of history and digital heritage. The group is 100% composed of volunteers and interested parties, and has expanded into a large amount of related projects for saving online and digital history.
Velocity was great this year. It was only my 2nd time out to Santa Clara, but from what I understand every Velocity conference is great. Few conferences can provide such rapid payback. Velocity does a good job setting the right vibe. The talks are often as entertaining as they are informative. This video of Scott Hanselman is a great example of that. He did an artful job of telling an open source audience they can run their code in the Azure cloud too. I happen to know a few companies that are doing just that. It’s good to see Microsoft continue to embrace open source, as I am a fan of running C# on Linux myself.
Getting a room of passionate experts together and making them laugh over plenty of food and drinks is a setup for awesome collaboration. When this crowd is having a good time the conversation drifts to tech talk naturally. If you’re not working the hallway track at the conferences your attending you’re missing out. I always enjoy finding others dealing with problems similar to the ones I’m currently facing and the two way peer review that comes out of it. On occasion someone has a silver bullet too, where they killed a problem dead in their organization and the solution ends up being easy to implement and highly portable.
As Velocity wrapped up I found myself with a 3:30pm flight back to Atlanta but a hotel with a strict 12pm checkout. I knew devOpsDays was coming up, google.io was going on, and a friend at the conference reminded me webperfdays was about to happen as well. Webperfdays was at Google’s Mountain View campus just up the street. When I looked into it the ticket was only $15.00, so why not spend half a day on Google’s campus talking about web performance if I didn’t have anywhere else to be?
The session was awesome. Webperfdays was a much smaller group than at Velocity. The day included open spaces discussions in addition to main talks and an ignite session. I was learning so much I actually called at lunch to move to a red-eye back to Atlanta so I could stay for the full day.
There were a couple of things that factored into staying after lunch. The first was I didn’t really know anyone there. I recognized a couple of people but it wasn’t anyone I had spoken with much. Looking at the line up for DevOpsDays, I didn’t see too many new names and had talked to most of the main presenters within the past couple of months.
This was all enough to lure me in the door but I stayed because they blew my mind. Andrea Giammarchi’s talk on mobile web performance did real time demos showing how slowing your app can provide a more consistent experience across devices, but assuming only state of the art equipment can lead to performance problems for the majority of your audience. This blog post walks through the talk and provides links to benchmark tools used during the demonstrations.
I also enjoyed Tony Parisi’s talk on webGL. In contrast, he made a lot of good points on how awesome our hardware can be and the crazy things it can do for us. This video covers a lot of the same points Tony made at webPerfDays. I was not aware of the 3D capacity our browsers have or how accessible it is.
My mind sufficiently blown, I came to the realization I hadn’t really been thinking of web performance at all. I’ve been very focused on the CAMS model of devOps, which for the most part meant getting changes to production as safely and quickly as possible with lots of monitoring and reporting. The focus was really on not breaking anything. A degradation in performance does qualify a change as breaking if it’s severe, but making sure our numbers don’t slip isn’t the same as actively trying to make them better. This is a cultural perspective within a shop, and to promote that from within I have to become a web performance advocate myself.
I don’t think I’m the only one guilty of adopting this devOps anti-pattern. It is human nature to pick teams and eventually get competitive. DevOps wants to de-silo our organizations, but I have witnessed industry experts self-segregate at the conferences where we’re working on correcting such problems. This is why I hadn’t really spoken with anyone at webPerfDays while at Velocity: We had all stuck to our respective tracks. The folks I knew were either at devOpsDays or had gone home.
I didn’t have a lot of session overlap with the folks I work with because they were mostly on the web performance track. Since I’m a devOps Engineer it only seemed natural that we stuck to our areas of interest. I’ll even admit I’m guilty of actively turning off my attention on some deep dive performance discussions, because, “it’s not my area and I need to stay focused”.
That thought process stopped as soon as I got back to Atlanta, and I hope you avoid viewing our industry so myopically as well. We have the capacity to study devOps, web performance, and more all at the same time. While we may not always be the experts in the room we can be fluent enough in other areas to be mindful of our impact and appreciate the wins and losses our teams face daily in their respective arenas. We can create a culture of sharing in our organizations, but if we can’t empathize with the information we’re receiving we cannot respond proportionally.
