Is Google’s new ‘Dynamic Data Masking’ solution an innovation or is it a sign that the once cool, agile start-up has become an uncool, corporate monolith?
By Jonathan Farina, Chief Technology Officer, WCKD RZR
How do you know when a “cool”, innovative company has become a large, slow, uncool corporate? When they start pulling their engineering punches or start pushing one of their products for the wrong reasons.
Rather than ask “Is this going to solve a real problem for my customers?” they add features for the sake of making a bit more money. They can see a real-world problem but won’t try to solve it because it’s too hard or too risky or worse.
Yes Google, I’m looking at you.
What am alluding to? Let me start at the beginning, are you a Software Engineer/Inventor or just someone who likes to think ‘outside the box’?
As a software engineer at heart, I often have an idea, and normally I go through what I like to call the “great idea” cycle. It goes a bit like this:
- Have a good idea!
- This is a great idea
- If that’s the answer, then everyone else would be doing it.
- *Quick web search*
- Damn they are doing it
- Go to 1
The great idea cycle has happened many times in my career, as I am sure it has to you.
Next, the “If only I had” variation?
- Have a good idea!
- This is a great idea!
- If that’s the answer, then everyone else would be doing it.
- *Quick web search*
- No one else is doing it, I should do something about this…
- Do something else for a while
- Damn, someone else just launched my great idea!
- Feel sad
- Go to 1
There have been a few genuine times in my life when I certainly went through the latter cycle… I had a great idea for a website where friends keep in touch…kind of like a circle of friends/family…and then Facebook came along.
Or the time I decided what the world needed was a way for people to “digitally” read books and magazines, just before eReaders were a thing, then there was the time I decided a “digital” bank would be amazing, you know, disintermediate the traditional banking model…hello Monzo, Revolut etc.
You get the idea; I am sure I’m not the only one, but I will say I try to learn from my mistakes.
Now, an interesting thing happened a few weeks ago… whilst surfing LinkedIn, I came across a post from Google. They’ve just launched “Dynamic Data Masking” in BigQuery, oh wow! I thought.
This is it, Google have gone and done it, finally beat me to “my great data masking idea”… But wait… I re-read the post, read the documentation behind it and then realised….no, sorry, you’ve simply got it wrong. Right idea but you’ve restricted it to BigQuery only. Seriously. Let me say it again…BigQuery only! As my kids would say, #epicfail.
I’m not trying to be a “Cloudist” about this, it not just Google, Amazon, Microsoft, and the others don’t even have this sort of technology yet. I should know. In my former life I desperately searched for such a thing.
The truth is, if you understand the real world or the businesses that operate in it, the issue is obvious. Imagine you’re a company who has just picked Azure or Amazon Web Services…or even if you’ve not started their cloud journey, perhaps they have on-prem data stores like Oracle, PostgreSQL, Mongo etc…., that is not really a very big leap of faith is it?
No company only has a single data store, especially not in a “cloud” technology like BigQuery, literally no one. So, this great idea solely focused on a single Google product is singularly short sighted and frankly…makes no sense.
This is a real problem; unfortunately, I’ve experienced these issues first hand across all the organisations I’ve worked at. So, what can you do? You have two choices, build it, or buy it.
So, here’s “my great data masking idea”:
Think of it in three easy steps, Find, Govern and Access. Ensure you have technology that can store the structure of your data wherever it lives (on-prem or in the cloud) and make sure you can search that technology easily. That’s Find.
Now ensure you have some technology that provides a single policy governance engine across all your data, remember the key is to write the policy once and apply it many times. That’s Govern.
And finally, invest in technology that provides a single logical access point to all your data. The key is that it must ensure it across all your data and data sources. Once you have all three pieces, you have a Data Enablement system that will solve your business problems consistently, securely and if built or integrated correctly, in a manageable way.
See! Now you understand why Google have failed so spectacularly. So perhaps I really can make something of this grand idea I have. It’s the kind of solution which a new, cool, innovative company might dream up.
