Christian Buckley interviews Naomi Moneypenny

Christian Buckley MVP, Microsoft Regional Director, CollabTalk LLC interviews Naomi Moneypenny

Video Transcript

Christian: Alright here we are again we’re here at the European SharePoint Office365 in Azure conference in Prague this is ESPC19. I’m here with Naomi Moneypenny hello.

Naomi: Well hello Christian how are you?

Christian: It’s great to see you again.

Naomi: It’s wonderful to be here thank you so much.

Christian: It’s only like I just saw you in Orlando before that I saw you in Redmond at the airlift I see during the airlift and inspire in the summer as well so seeing all the major conferences so it’s like we’re hanging out all the time.

Naomi: I know it’s good isn’t it I’m excited to be here at ESPC this week.

Christian: Well it’s great well so I know you were talking in the keynote this morning about your sharing about project cortex I’m really intrigued by kind of the background to that we’ll get get into what that is but maybe you talked about the problem spacey we’re trying to solve for before because your background maybe you could tell the world for those that don’t know you some of your background and what led you down that path and when you kind of talked about the background for the project cortex.

Naomi: Sure I’ll be happy to. Yes so for the folks I haven’t met before my name is Naomi Moneypenny I have the privilege of leading the product development team for project cortex and I worked for many years in the machine learning area so before I came to Microsoft I was working in expertise understanding and identification from collaboration networks and so I worked a lot in machine learning for about 15 years or so before I came here and so it’s interesting to think about that with sort of the background with project cortex and how we came to some of the areas that we were doing now there’s some fundamental challenges that haven’t changed in many many decades inside of organizations and we think it’s really important to finally think about how we can address some of those core problems. So you’ll recognize a lot of these right these are things that we’ve seen a lot so for example does this connected silos of information it’s really hard to be able to find the expertise that you need inside of an organization and then of course you have this whole motivation around how do you do knowledge sharing inside of a company.

Christian: That’s right there are there organizations with who are better at it than others because they’re just very social in nature and that’s actually what’s a we kind of got to know each other around this social collaboration space and giving your background in search and knowledge management but because social was leveraged as early on in this kind of you know history right of a way to add context to a lot of those knowledge assets that are being created because in that if we’re talking about it sharing it within that social activity that’s a big part of kind of where cortex is now solving kind of another problem.

Naomi: Yeah that’s right I think it’s about connecting a few things right you think about layers if you will within the Microsoft graph right think about there’s a layer around the documents and the storage essentially kind of the nodes if you will in some of that network and then we’ve layered in a layer of people it’s an important piece right it’s like the internet connecting people together thinking about that layer of connectivity and on top of that you have a bunch of signals that are really coming from all of the activity between those people and the content and. The conversations the comments the tasks the calendars all of those signals. And so you can think about that as the Microsoft graph increases what we’re doing with project cortex is essentially helping you to mine dynamic entities if you will from there and that’s really helped to create topics maybe it’s a project maybe it’s a capability a skill a campaign an initiative anything it’s happening inside your organization but things that are very organizationally specific and important to your company and so we’re mining through all of that wonderful content in the Microsoft graph and creating this whole new rich set of connections and so we really think about it as a way to connect knowledge together the people and organizational information and then to be able to use all of that collaboration data to really make it keep it up to up to date.

Christian: What I think it was important about it because obviously in the knowledge management space in building out portals for many years decades you know we have gone in and manually gathered data in people profiles and other information and built out pages if you think of like a Wikipedia type knowledge base of information that’s out there and then tried to surface that out to users yeah but we’re talking about now is for the system to automatically go and build some of these components for that discovery process again to be in this intelligent system that is able to say okay Naomi and Christian you’re you’re doing a lot of interaction talking about this topic I see that it’s now a formal project it’ll then go and generate then the pages with that content, add us to it and then we may even be unaware that there might be conversations happening halfway around the world and another project team that is relevant to this and it will surface that content if it thinks it’s relevant to the conversation.

Naomi: That’s right so it’s really about getting over those two big fundamental problems one of the reasons I think knowledge sharing efforts of the past have not been able to scale there’s two things that it relied on one is maybe you thought about like things like expert systems in the past so there are very rules base they were very dominated by a hierarchy or they had to be very formulaic in the way that they were answering they couldn’t adapt they couldn’t iterate. The other big issue is sort of knowledge sharing relies on wonderful people like us who want to share our knowledge but do it on top of our day jobs right and that’s a really hard thing to do so it’s a hard thing to ask people to scale that up

Christian: These people should go in there and apply metadata.

Naomi: It’s really and so we didn’t think about that so when we’re looking at this issue is sort of for project cortex we’re like fundamentally how can we change this how can we change the game of how you do this and so AI has to do that heavy lifting right it has to go through all of the organization’s content because you don’t know where to start as a person how would you ever start right. The the problem seems so hard they would never get started with it and so we need AI to go through to identify all that content in your organization figure out what projects are being a surfaced from there and then you of course as a person can go through and curate that and help to make it better. But we really want AI to be able to do that heavy lifting and to keep things up to date because heroic efforts of the past they last for one year or something like that and then they go stale as well and then it feels even worse when you try to do it again and so we need AI to basically help that keep that semblance of connecting and discovering new content and keeping it up to date and surfacing it up to you as a person to say is this the right data did we get it right are we doing are we doing something valuable here and so there’s a kind of that whole feedback evolution that goes on between AI discovering the new content surfacing up this information to keep a project up to date and then the subject matter experts you can update.

Christian: The curation aspect of it. This is the hardest part about all this is that I’m excited about hearing about this project yeah it’s a very limited access to the early you know the program it’ll be expanded out uh you know in in the we talked about timeframe for those things because I know all those things kind of fluctuate but it’s so we talk about those things it’s again one of those examples I’d like I’m hearing the right things I want to get my hands on it and play with it I mean I have a lot of individually questions about the curation process because a you and I have had these conversations in the past about artificial intelligence and how much trust to hand over to AI to go and do and and how much of that heavy lifting you know can be done i I think that the curation piece is bigger.

Naomi: There’s a lot of things that we’ve done there so there’s a few things we do there’s kind of the cold-start problem if you will so how do I make sure that I’m scoping it to information that I want my not doing that’s a good start so we want to make sure we are able to do that we also need to compliment customers who have invested in knowledge management in the past so and that’s also a big sort of structure and the culture and the behaviours that they’re getting there as well and so it’s both the technology problem and a human problem and we want to make sure we bring the best of those worlds together. So you can do things like scoping the mining for example of like do don’ts just want certain sites within your organization to remind and that helps you to gain more trust and confidence in the system so you can actually look at this and surface it up maybe to a knowledge manager first and make sure that they’re verified mr. projects feel good maybe it’s something you want to think about is authoritative inside of your organization and mark them as such and sort of have a curation policy around those and so we think there’s a lot of governance there’s a lot of best practices that can be applied in there these same things you might do for hub sites right now in other areas we’re using that kind of mechanism. So we think about that sort of mining piece. There’s also a whole area around curation and disambiguation I’m merging in topics of all those issues –

Christian: Oh like a scenario for example of like an acquisition I think it says suddenly and there’s a this noon you know data silo some knowledge base or something that suddenly is added in well you might have dozens of topics that are out there hundreds of topics created but suddenly now you have this new data this information sources all of these other resources and for the AI to be able to go and apply it like pull consume that apply that and add those assets in yeah at all those different locations is something that’s very exciting but you know from AI so I started my career as a business analyst in a tech writer and the process of constantly having to follow and be aware of who’s done what so I know what to go and change the documentation and where it needs to change. I mean you’ll be able to you know from a change management standpoint in organizations something changes you got to understand you have insight into what’s impacted across all of those different projects and the people yeah.

Naomi: So there’s it’s interesting to think about because there’s that kind of notification and I’m following and I want to keep up to date with the topics of things that are interesting to me but we also want to think about what are your contributions to the knowledge Network. How does your contribution actually weigh and how did you help with other folks as well and the things that you’re sharing how valuable is that inside of the organization? Right because you get a great sense of achievement from that right and we get.

Christian: It’s not just how many blog posts you’ve done how many articles you’ve written how many documents you’ve authored but the ones that are actually being consumed you might have somebody who who authors or co-authors you know a tenth of the articles of another person and yet that be seen as the subject matter expert or that content so you can start that attribution which is incredibly hard in systems and marketing systems right I’ll be able to see that in the value of content that is incredibly powerful.

Naomi: And then having that sort of message for if you think about it so if we look at sort of the next iterations of the work that we’re doing in project cortex is really about understanding like how is that work being used so think about it like you have an authoritative definition of a project right you might have a slide that you use a lot and you want to say if this is the slide that we want to use where we talk about this area well I want to be able to surface that every time we create a PowerPoint presentation about these things we want to be able to suggest that we want to be able to take a topic page for example and be able to create PowerPoint slides from it. Right because why shouldn’t I be able to write the same ways we have a QuickStart a function right now for PowerPoint where it minds through Wikipedia and creates a PowerPoint deck for you. There’s no reason that we couldn’t do the same thing from a topic page right so as soon as you’ve got that information it’s being curated in one place we want to make sure we’re empowering the people who are the subject matter experts to be able to reuse that content in lots of different ways inside of your organization. The goal here is that learning to learn faster skill. Right organizational agility how do I build that into it and so making sure I understand this information how can I repack it and reuse it and go faster inside of my company instead of all this kind of communication issues and there’s a redundancy that goes on right now.

Christian: Right well that was you know I think most people that are newer to this space will probably became familiar with some of those concepts in the early with the first delve release for example. So we’re where the problem with search was that you know search works if one if you have the right permissions to get to the content the first place to that you have used the right keywords to find that content three whether those people have you know the owners of the content have marked it up properly there’s all of these kind of barriers to finding making that search experience work and so by automating a lot of that capability it answers that kind of that basic need of of search of the system understand what I’m trying to do Who I am the context in my conversation and surface the things that I’m asking for and not asking for because I don’t know to ask for the right thing. Exactly so putting that intelligence behind that searchability findability discussion.

Naomi: Now I think it’s a great point I think I love it best with Simon Denton one of our customers from Mott Mc Donald he has a great quote around this because he said you know search is only 50% of the answer and he’s totally right right because it’s like if I have to search in some ways I failed right because I can’t find the information I need it didn’t come to me automatically and it should right I should be able to find that information and I should have it suggested to me proactively as well and so understanding topics for projects having that discovery ability right there the ability to take expertise and mash that together with the content is a really important breakthrough.

Christian: We need to get rid of the word search as the term for that because it needs to be more you know what I’m talking about that good joke of like it’s not about searching it’s about finding I know so we needed a term that’s like find based. I mean what else can we expect someone from the program kind of whether the official notes on you know people want to find out more obviously we there’s the the aka project cortex people get more information about that what’s the kind of the timeframe for expanding that the preview?

Naomi: So we had a preview that was been running obviously before we announced the project and so it’s been a great way to sort of really test every step of the wave of what we’ve done with project cortex with some of the largest companies in the world and to be able to bring them on that journey and so when we announced a month ago or so at Microsoft Ignite it was a great way to say okay you know we’re gonna have this capability are you really interested in working with us hard to make sure that we get that preview information and we got to work with an expanded base of customers. So you can apply it’s a aka.ms/projectcortex and so you can apply to form online there there’s also a bunch of demos and project videos and all kinds of stuff and the blog post and all the information is there so it’s been a really great resource just to keep up to date with what we’re doing. This is a sort of next wave of the private preview is really happening towards basically at the beginning of January so that’d be it and then we were hoping to general availability by the first half of next year so it’s very soon now so even though you know you may not get your hands on the bits immediately but it’s coming very soon just within a few months.

Christian: Well obviously with that diet timeframe you have a lot of free time on your hands she’s gonna go lounge around for the rest really appreciate your time and joining explaining more about project cortex and especially if your organization has done any of the because this is a premium service that we’re talking about so we’re not gonna find it in like the e1 or even possibly the e3 I know we’re not gonna get in the licensing discussion where it is but.

Naomi: I just build the product right.

Christian: But this is but if your organization has done you know more of the custom work historically around knowledge management and and search and are interested in finding out more about it you definitely gonna need to go take a look and pay attention to what’s happening with project cortex. Naomi thanks a lot for your time.

Naomi: Sure, thanks so much Christian has been great to see you and great to be at ESPC.

Christian: Right thanks a lot.

Share this on...

Rate this Post:

Share: