Commentary: Companies have been selling miracle cures for technology lock-in for years, but customers have higher priorities.
When are we going to learn to stop trying to sell ideologies? Customers should want open, unified access to their financial data stored across different banking providers, as promised by the U.K.'s Open Banking initiative. But, apparently, they don't, with adoption of open banking still anemic after three years. "Customers are not influenced to switch banks because they can take their data with them," Starling Bank CEO Anne Boden said. "Customers switch banks because they want a better service."
At larger scale, and for similar reasons, enterprises should eschew the lock-in that comes from buying into a particular cloud vendor's services, especially serverless functions that tightly tie a customer to the cloud provider's infrastructure. And they should never choose proprietary software over open source. But enterprises keep buying into cloud, with adoption of serverless options accelerating, and they keep using proprietary software.
This isn't because people, or enterprises, are ill informed about their options. Not usually. No, it means they have different priorities. And the top priorities seem to be speed, flexibility and productivity.
Business outcomes, not religious dogmas
Take the big contract U.K. spy agencies recently inked with AWS. As reported by the Financial Times, "The contract is likely to ignite concerns over sovereignty given that a vast amount of the UK's most secret data will be hosted by a single US tech company." Actually, it's more likely to spark lawsuits from rival cloud providers that want the business (because that's what the big clouds do with their lawyers—see here and here, for example).
No one seems to be concerned with the reason for the cloud spend, however. Ciaran Martin, former head of the UK's National Cyber Security Centre, said the deal would enable the spy agencies "to get information from huge amounts of data in minutes, rather than in weeks and months." It's not about national sovereignty; it's about getting stuff done.
SEE: AWS Lambda, a serverless computing framework: A cheat sheet (free PDF) (TechRepublic)
In a similar manner, an executive from a large insurance company, speaking on the importance of serverless to Simon Wardley of DXC Technology, stressed that the decision isn't about technology, per se–it's about focusing on business outcomes:
During the time it took for the vendors to come back with quotes to the RFP, we had the system built in serverless, in production and reducing cost per transaction from $20 to eventually 8 cents. There's a cognitive burden. You can have a serverless team focused and taking on reducing costs of transactions, speed of claims and the tasks at hand, or you can have a team focused on the intricacies of container distribution. We're an insurance company. We care about customer outcomes, not infrastructure clusters.
Years ago then CoreOS CEO Alex Polvi called AWS Lambda, the company's flagship serverless offering, "one of the worst forms of proprietary lock-in we've ever seen in the history of humanity." He was perhaps right in terms of licensing and hardware (because serverless locks you to a data center, not just a piece of hardware, and you never actually get to access that hardware), but wrong in terms of the importance of that fact. A company could avoid that lock-in by ... writing all the underlying software (event model, deployment model, etc.) But that's a lot of work, with a relatively puny payoff. Hurray! You avoided lock-in! And for your efforts you bought yourself irrelevance as more nimble competitors build on the differentiated services that cloud/SaaS providers offer. Small wonder, then, that a Datadog analysis showed that companies are invoking Lambda serverless functions 3.5 times more often in 2021 than two years ago.
All of which is a long way of saying that if your vendor's primary selling point is an ideology, rather than a quick path to positive business outcomes, take your business elsewhere.
Disclosure: I work for MongoDB, but the views expressed herein are mine.
- Cloud lock-in: Why the microservices cure might be worse than the problem (TechRepublic)
- IBM study: Single-vendor approach to cloud is dead (TechRepublic)
- Multicloud: A cheat sheet (TechRepublic)
- 5 programming languages cloud engineers should learn (free PDF) (TechRepublic)
- Power checklist: Local email server-to-cloud migration (TechRepublic Premium)
- Cloud computing: More must-read coverage (TechRepublic on Flipboard)