Moving workloads in and out of the cloud isn’t really possible—or a good idea.
Stories by Matt Asay
If the ubiquitous spreadsheet program is the gateway to data science, Python aims to be the next step.
Making great efforts to tout its closed source software, Snowflake may be trying to convince an audience that doesn't really care.
Try using Kubernetes like an app server for smaller teams instead of treating it like a centralised cloud.
Companies that jumped in early are reaping the rewards, but it’s not too late. Businesses must now find the people and partners to help move ahead.
Both the single-stack architecture and the best-of-breed approaches can limit you. Open source and building in the capability for change are key.
Someone once said that Python’s data science training wheels would increasingly lead to the R language. Boy, was he wrong.
Few of the Supreme Court Justices seemed to understand what an API is or does, but their decision was a victory shout for software developers.
Commodity services that would allow workloads to run across multiple clouds don’t exist. But that hasn’t stopped the multi-clouders from trying.
For Artillery’s creator, the key to the popularity of the open source load testing tool is a focus on the needs of both developers and operations.
According to Graham Neray, CEO of Oso, authorisation will be the next layer of software to be abstracted and made less onerous for developers.
We need one platform to ‘process, store, secure, and analyse data in real-time, across all the relevant data sets,’ according to MongoDB’s CTO.
Decades into what should have been PostgreSQL’s dotage, developers keep reimagining what it can be.
It’s a mistake to believe that running open source in the cloud will avoid vendor lock-in. But open source offers freedom and independence.
Cloud computing was like rocket fuel for software developers. Scientists and engineers are the next in line.