6 ways organizations can support employee mental health and well-being
Keeping people engaged and satisfied with their work helps organizations build a sense of belonging and create a welcomi...
We have to face it. Cloud vendors give us a choice to be their guinea pigs with pre-release features, or be late adopters and use only what is ready to use and proven.
Before we begin, let’s shed some light on the phrase: “It’s in preview”, used so often in the context of cloud services.
Years ago I started my career with on-premises technology (there was no cloud back then), and the majority of my work was related to Active Directory.
Up to 95% of enterprises use it now, and if you ask me, it is one of the better pieces of technology created by Microsoft. Back then, it was the main point of the on-premises universe.
With every major change to Active Directory, we as consultants, and Microsoft as a vendor, had the same issue – adoption!
People were afraid of even touching some of the parts of Active Directory. The best way to scare the AD admins was to mention the “Schema Extension”.
For those who are not familiar with it, a schema is a definition of objects and attributes used in Active Directory. Once you commit a change to the schema, you can’t revert it.
It doesn’t mean you couldn’t do anything about it, but lack of information made people scared to death when a “Schema Extension” was planned.
To be completely honest, it allowed me to make a nice profit and work on many projects to execute schema extensions. Even if I tried to convince customers that their projects mostly were not necessary as there was nothing to be scared of.
Why did Schema Extensions pose a problem?
Every update to schema was created to enable some future updates and functions. Each new one built on top of the previous one.
Because people were scared to update their schema, they were falling further and further behind, adoption slowed down, the value of software updates and new versions was diminishing… the feedback loop was stuck in a vicious cycle.
Fun fact: because AD schema is closely coupled with Exchange server functions, it was also slowing down Exchange updates.
The Exchange team developed an LDAP driver, which faked schema extensions if they weren’t in place. If an attribute was not in the schema, Exchange saw it and reported it as “no value”. That was a long time ago but lessons were learned.
Think about it – what if there was a magic switch to “Disable all preview features” in your Azure or Office 365 subscriptions?
Most of the people responsible for it would set it to “ON” (problem disappears) and forget about it.
What would be affected? You guessed it – adoption and feedback loop.
That’s why (in my opinion, I’ve no insider knowledge about it) this functionality is not available, although I know such requests exist and are often raised by customers.
The cloud vendor gives you the opportunity to move faster and benefit from things that are still being developed, for the price of possible additional problems, services not being finalized, and not having a full SLA on them.
Among our clients, I see three major groups when it comes to approaching features made available in preview:
I think that the key to making this decision is knowing your enemy.
In short, they’re test services or functionalities which are still under development, and so:
These are the key elements you need to be aware of when deciding that you want to go with the flow – I mean, preview – in the cloud.
Sounds scary? Maybe – yet in more than 90% of projects we are doing in the Azure cloud, our clients use pre-release features in one way or another. I’m not saying it is without risks but there are some benefits.
Here is an example of a service developed on an extreme scale while using preview – I think that even the Product Team was a bit surprised that someone decided to use it so early.
When we got a request to solve for a pretty sophisticated scenario of knowledge mining, including things like searching through handwritten notes in multiple languages among various document formats, it sounded like “mission impossible”.
It was mission impossible until one of our consultants brought up this new, just released, preview feature in Cognitive Services.
Fast-forward one month, and the solution was born and deployed. It allows users to find the right information fast, using advanced analytical functions to perform intelligent searches on large quantities of business information. Here you can read a bit more about it:
You have to answer these questions to yourself and to your organization:
You need to find the answers first, taking into consideration legal requirements, business and technical risk, and your overall goal.
Pros and cons of pre-release features
Both approaches have their advantages and disadvantages.
Among the cons – you should definitely prepare for things not working as expected and a bit of re-work here and there. You also need to evaluate and mitigate the risks.
On the pros’ side – something less visible:
What is your take on it?
Are you all-in and using every feature or service you can get your hands on?
Are you lukewarm and trying some features in less risky projects that are not going to production anytime soon?
Or are you taking a safe bet and only using what is already fully supported and proven in the field?
Let me know. I’m curious to hear about your approach.
There’s a single question that is raised every time I speak with a client about using a preview feature. Maybe we will meet in the future, so let me post it here and answer it right away. Here goes:
Q: When will feature X come out of the preview and go into GA?
A: I don’t know!
Really. As consultants – we don’t know. As an MVP and Regional Director, I might have some indication of it, but I wouldn’t be able to share it with you under my NDA.
You can expect it to happen around some major events or the end of the fiscal period of a specific cloud vendor, although it is not certain. If it is really important for you, talk to your partner and cloud vendor. Maybe with an interesting use case, we can “work some magic” for you.
Read other similar articles