Identity and Access Management

I’ll admit this article made me shed a tear for Novell eDirectory. Still better in various ways than what Active Directory is even now...

Back in the day, we were a Novell shop and had a new payroll application come in that was promised to work with eDirectory. Come the week before implementation, the implementation team got very clear that Active Directory was required (“oh, not sure why the sales team said that...eDirectory support is a ways off”). That was a long week/weekend.

Our first AD domain was called “MAD” because that’s a solid acronym for Microsoft Active Directory, doncha know? ;)

I think one of the things you can look for, which was pointed out in this article that Forrest referenced (https://www.forbes.com/site... ) is the presence of companies dedicated to migration. In the case of the Oracle article, the author points out the existence of entire companies dedicated to helping customers transition from Oracle to AWS, Azure, etc.. The lack of companies finding demand for that suggests that companies do not legitimately need that service at the moment.

One might say that this is only the current state of the cloud wars if you will, but if all of the talk about how terrible vendor lock-in is hasn't led to anyone having a real job, how long will we have to wait for it to become an industry unto itself? Given the fundamental flexibility of serverless architectures, and especially if you use the microservices design pattern, this shouldn't be something you're unable to rearchitect without rewriting everything.

Just my 2 cents.

So you're saying that the cloud alarmists are right, but that it doesn't matter. I think for such alarmists, it's supposed to be self-evident that proprietary lock-in is bad. The story of FOSS is deeply ingrained in our trade, and seeing the "end" of the story being that FOSS gets repackaged and made proprietary is a cruel twist of fate.

I'd say it's concerning for one firm to control all computing because of the general principle that power corrupts, and absolute power corrupts absolutely. Neither is this an experiment where we can see how things go and reroll the die if it doesn't shake out the way we want.

It's entirely possible that it's in the interests of each individual tech company to embrace something which is globally suboptimal, if not catastrophic. That's something like the Tragedy of the Commons in economics 101.

But maybe computing technology is cheap enough that if aws ever really turned against the world, it wouldn't take people too long to come up with alternatives.