Cloud-Based Services: Physician, Heal Thyself
Some of our readers may remember the phrase, “physician, heal thyself” from the New Testament of the Bible. According to Wikipedia, the basic intent of the phrase in the proverb (Luke 4:23) is, “Hey… before you can help us, you have to take care of your own problems!”
This saying came to mind when I was reading this AM that Google had experienced a system wide outage early on Monday morning. The problem had something to do with the user identity services, meaning that it affected any service that requires the user to sign on (Google Docs, Google Meet, Google Classroom, YouTube, etc.). Google had another outage not too long ago (September) that affected Google Docs and Meet; as a matter of fact, it’s been a rough year for the company – there were reported widespread outages in July and August as well.
The news media just loves this stuff, and so today’s headlines about the formal election of America’s next president, Brexit negotiations, Russian hackers, 300,000 dead from COVID in the U.S., and the start of vaccinations, had to make some space for a headline about Google. After all, Google and Facebook in particular have done (at least in America) what politicians haven’t been able to do, to unite Americans – unfortunately, in their disdain of the tech giants.
As bad luck would have it, Google’s troubles bear some similarity with problems right here at little ol’ RSR. Our website was re-hosted to a new service last week, and the immediate result was that no one could sign on (another identity services issue). Since none of us at the company have any development chops anymore, we were reduced to complaining, the equivalent of banging pans in the kitchen while the cook is trying to get the meal out. All of this reminded me of an old joke I used to tell on myself (a recovering CIO): “there is no person more technically incompetent or dangerous in the IT department than the CIO – he thinks he still is good at what got him the job in the first place!” Now that I think about it, my staff never thought it was very funny.
But underlying both Google’s very big story and RSR’s very little story is an underlying truth. That is, the “-abilities” still really matter! The top careabouts of any IT services provider, whether it’s an internal raised floor operation or a 100% cloud-based services one, must be reliability, recoverability, scalability, serviceability, secureability, manageability, adaptability (and probably a few more you can think of). If an IT services organization can’t deliver on the “-abilities”, then nothing else matters. No amount of state-of-the-artiness, cool functionality, or sexy interfaces matters one whit. If businesses have put their digital futures in the hands of an IT services organization, it must deliver five-9’s (99.999%) reliable service, or it doesn’t deserve the trust – or the job.
My partner Paula Rosenblum frequently makes the point that “the cloud” has been a vendor-led initiative. My own contention is the flip side of that same coin – that retailers that are in a hurry to get to new functionality would consider “the cloud” first, not because of the logic or beauty of it, but because they have no choice. One of the things that we’ve all commented on here at RSR is that COVID has only accelerated the move to the cloud, because functionality that retailers thought they were going to have time to develop is suddenly mission critical (enterprise-wide contactless selling and customer order management systems come to mind). The real driving energy for this comes directly from consumers – they live in “the cloud” 24X7 and expect businesses to be there too.
But it’s all for naught if the services provider isn’t reliable; then the client business isn’t better off, but worse! So, this is (or should be) the mantra throughout the technology world: “Physician, heal thyself!” The difference between now and Biblical times is this – we can’t wait around for a miracle. The hard work of making the technologies that the tech world has made our society totally dependent on more reliable, must continue.
One of the great thinkers of business computing in the 2nd half of the 20th Century, Gerald Weinberg, used to start off his lectures with a lesson for his audience (I paraphrase): “how many bits – off/on switches –are in IBM’s mainframe operating system? <answer: over 1 billion> How many have to be in the wrong position for the system to fail? <answer: one!> Therefore, computer science truly is more difficult than brain surgery, because the human brain will re-route signals when a problem exists in some of its cells!”
We like to pretend that this isn’t relevant anymore, but we’re fooling ourselves if we do. We used to hope the system came up in the morning. Now, we absolutely unequivocally need it to be there for us, 24X7.
There’s a long way to go before that’s the case.