VMGuru

Everyone is talking about it: Chatbots are going to be the next bing thing. If you don’t believe me, check out the Forbes article on why chatbots will be your new best friend. A quote from that article:

Chatbots came on the scene in 2011 as business intelligence, artificial intelligence and messaging platforms combined into new forms of responsive technology. New ways were needed to support companies interacting with buyers and provide customer support that aligned and could evolve with changing communication habits.

The way we communicate with technology changes. Where we first interacted with a computer by punch card on a very low level, like programming, the communication evolves into a more natural form of communication with a human-like interaction. Our technology more and more ‘understands’ us and knows what we want, sometimes even before we know it ourselves.

Chatbots and natural language processing are a sign of that change in communication.

So, what’s a chatbot then?

A chatbot is a messaging application, also known as a conversational interface. It’s sole purpose is to simplify complex predefined task(s) based on natural language. A couple of ‘famous stars’ in the conversational interfaces are of course Star Wars C-3PO, Amazons Alexa, Apple’s Siri and Microsoft’s Cortana.

But if you think chatbots are just the latest hype, you’re wrong. In 1964 ELIZA was born in the MIT AI Lab. ELIZA could be seen as the first chatbot. ELIZA is able to have a conversation with a user, or least the appearance of a conversation, the same like the chatbots of today.

Fast-forward to today. There are different types of chatbots: chatbots that schedule your appointments, chatbots that help you select the right Valentines present and, like OpVizors OpBot systems that provide you with information from different knowledge systems.

To learn about chatbots and how to create them I started on researching natural language processing and chatbot frameworks. To put that knowledge to the test I created Jeeves, the VMware information chatbot.

Jeeves, Your own VMGuru chatbot in your chats

Jeeves is a chatbot that you can add to your Slack team. It will provide you with all kinds of information contained in the sizing guides from VMware. For now his knowledge is limited to (most) information on the maximums from vSphere 6.5 and (most) information on maximums of NSX 6.3 (Thanks @smitmartijn).

When you add Jeeves to your own Slack team you get your own VMGuru, with the knowledge already installed. When we add knowledge to the database, it will directly be available for you in your own team without installing an update.

So, what can you ask Jeeves?

You can ask Jeeves about all kind of metrics for VMware, like how many VMs you can place on a host, or how many IDE controllers you can add to a VM.

You can talk to Jeeves in English if you like, but if you just want the short version, you can: VMs per host

As I said, Jeeves will, for now, only answer questions on the sizing of vSphere 6.5 and NSX, but will in the future answer questions on various VMware related subjects. Because of that unknown queries will be stored for further analysis.

Cool project, I want to help

Putting all that knowlege into the databanks of Jeeves is a lot of work. If you want to help, or have information that could be nicely integrated into Jeeves just reply below or send a tweet/meesage to me. Together we can create the best VMware information butler there is.

Where can I find Jeeves?

At the moment Jeeves is not yet available in the App Directory at Slack, but hopefully will be in the near future.

If you want to add Jeeves to your Slack team, just click the button below.

What’s next

For Jeeves the next step is to ‘know’ more about various kinds of VMware products so it can help you in correctly sizing and designing your infrastructure. Now it is only available via Slack, but as Jeeves learns more and more it will be available on more platforms. Connections to Facebook Messages and Lync are already in the works.

For what the conversational interfaces itself I think that the sky is the limit. If you can imagine it, you can build it. It looks bad for people in support organizations, because if you can capture it in a workflow or ‘call script’, you can automate it with a chatbot. Only for some hard questions you still need humans to do the answering/troubleshooting, at least for now. There are already conversational interfaces that do the most of the talking, and only for the hard parts it gets help from a human companion.

I think that Jarvis isn’t that far fetched as we might think.