When friend of the show Ward Bekker sent us a link to the recent survey write-up on the State of Developer Ecosystem in 2019 by JetBrains, we immediately set up a recording date with him to go over all the facts and figures…
Podcast: Play in new window | Download (Duration: 1:03:05 — 43.6MB)
Subscribe: Apple Podcasts | Spotify | RSS | More
DevOps appear to be quite rare
The first thing we picked up on is how many organizations are still surviving without any kind of DevOps. Even though everybody is talking about DevOpsand config management, it would appear, at least according to this survey, that these tools are still far from prevalent in the development environments. After discussing the different facts and figures contained in the webpages on the JetBrains website, we were left wondering how generic the target group was. Since this survey was conducted by JetBrains, it would definitely make sense that the respondent population was taken from their customer base and this could skew the results towards smaller, “Indy” development environments.
The sense and non-sense of Multi Cloud deployment
We then take a bit of a detour to talk about multi-cloud deployments. While it may be a good idea in theory, it is often less than ideal in practice. You will either limit yourself to the lowest common denominator of available services on the different clouds, or you will need to take on quite a bit of glue layer development… Where container based deployments will allow some extra freedom, the better way to look at multi-cloud deployment could be to land different workloads on different clouds, where they are best suited. But this no longer avoids the much feared “vendor lock-in” which is quite often the reason why one would look at multi-cloud deployment in the first place.
Lies, damned lies, and statistics?
Circling back to the main topic, we tried to take a deeper look at the raw numbers, which at the time of writing were not yet available. So we needed to try and extrapolate meaning from the available information and kind of concluded that the raw data is really required before real conclusions can be drawn… That old “Lies, damned lies, and statistics” mantra does come to mind.
And for once not missing a chance to plug our merchandise store, we’ve had a quite relevant bumper sticker available for a while now: 😀
Please use the Contact Form on this blog or our twitter feed to send us your questions, or to suggest future episode topics you would like us to cover.