Splitting light

Season 2 Episode 40

The cost of speed

If you are no longer interested in the newsletter, please unsubscribe

Around early June 2019

From November 2017 to May 2019, a half since the pivot, while maintaining half a dozen existing storage products we had launched one in general availability and were in private beta for another one. Object Storage was now in two regions. Block Storage was getting multiple AZs. At peak we were 14 in the storage team but on average it was more around 7 or 8 members.

Image for the Cyberduck tutorial post (1)

But the speed had a cost. The cost had three dimensions. Organisational, team-wise, and personal.

We had constantly pushed forward. Sometimes encroaching on the scope of other teams. Sometimes we had short circuited people. Our only goal was moving forward. We didn’t do it for politics. But the way we did made some people uneasy. It’s not that we made some enemies, it’s more that some people felt threatened perhaps? They couldn’t tell us what to do and when. This friction grew.

Between the lightning strike we had triggered on the Block Storage and compute integration, the fr-par region we launched without formal approval, Théo (a) searching new locations to deploy Object Storage, the friction kept increasing. Théo started to have more and more constraints. The static electricity kept building up. The tension was palpable.

Image for the migration data with MinIO client tutorial (3)

Inside the team, we had been relentless. That translated into two ways. The Block Storage lightning strike had broadened the rift in the team. The source of that rift had specifically been because we pursued speed over other bits. That resulted in Florent (b) having to change teams and leave Scaleway. It led Loic (c) to not come back after his international experience. The other facet was Marian (d) being thrown at a very hard problem and did not succeed. I failed to guide him... It was a failure precisely because I did not want to be slowed down, because I pursued speed at all costs. I would later learn that sometimes it is better to take more time to do some things.

On the personal level, I could not understand it, but I was exhausted. The work had drained me. Between the intensity of the work, the team and the organisation, I was empty. The on-call inside the team was not fully organized yet in the team and it was straining. The loss of the hardware lab had shaken me to the core. I felt bitter.

This is the cost of speed. We wanted to go fast. We were extremely motivated. But the speed had a cost. I can’t tell you if the rest of the team felt it. I only know how it was for me.

For the first time since I had joined Scaleway, I wanted to leave. Not because of the job. Or perhaps more precisely because of the whole experience itself. It was everything. I both loved it and hated it. It was too intense. The light was blinding.

Break next week. Season 3 starts on June 16th, stay tuned.

All visuals done by the design team led by Emmanuel Mourgue at the time. Check him out. An artifact of the space oriented marketing before the pivot to elements. I wanted to honor these beautiful graphics.

(a) Théotime Rivière, Storage Product Manager then, now Founder of Freedom From Scratch

(b) Florent Rivoire, DevOps Engineer at Scaleway at the time, now “very good in his life”

(d) Marian Gappa: then part time intern, now DevOps consultant at Zenika

If you have missed it, you can read the previous episode here

To pair with :