I have found this picture in a comment to this Reddit thread and would like to better understand its message. I do not know the image's original source, maybe it will come up later.
1 Answer
The Reddit thread asks:
Why is the Docker Mind Shift so hard?
The OP then links to The Docker Mind Shift and states that
The Docker Mind Shift helps explain the challenges many people struggle with, in DevOps and containers. Thoughts?
(Currently,) the highest upvoted comment states
to be honest my biggest challenges have been with the speed of tooling changes rather than the concepts. In theory, i get the workflow and ideas behind docker fairly well, but in practice sometimes it does feel like you're throwing a whole load of plates up in the air and then walking off before they hit the ground again. I get that as a new tech idea it moves fast, but just trying to keep up with the ever moving goalposts can be exhausting. It does seem to have gotten better in the last 2 years though to be fair.
In response to this comment, sirex007 responds
after finishing any reasonably sized project in docker, i'm always reminded of this image
and links to the above image. This image is attempting to comically illustrate the confusion many feel with the world of DevOps. There is a smorgasboard of toolkits and frameworks out there, and the bullets in the image are a commentary of the nonsense of the diagram to the right.
This is seen in the first bullet:
- Did you just pick things at random?
The subsequent bullets then expand on this question. For example, Redis is a caching system and stores key-value pairs. Similarly, MongoDB is a type of NoSQL database which also supports key/value pairs and caching, so it is a little silly to have both Redis and MongoDB.
It seems that sirex007 feels that this graphic illustrates the point that I get that as a new tech idea it moves fast, but just trying to keep up with the ever moving goalposts can be exhausting.
and my biggest challenges have been with the speed of tooling changes
as there are a lot of trendy tools appearing in the graphic, which is making the point that the implementer doesn't seem to understand the toolkits in use in his or her stack well.
This is a fair point: with the DevOps goal of rapid iteration and deployments, the appearance of new tools on a frequent basis and the changes within those tools could certainly seem like they are moving at a breakneck pace. I would argue that this is evidence of the success of the DevOps model however. If these projects are able to successfully iterate rapidly using DevOps, it is proof of the model's successes - though perhaps the cost is trying to keep up with that rapid pace.