June 7, 2020

Digital Experience Monitoring (Lifeguard IT Recast)

Digital Experience Monitoring (Lifeguard IT Recast)

*** Recast from the Lifeguard IT Podcast ***
Digital experience monitoring solutions empower IT with data and analytics to understand the intersection between end users and technology. With this knowledge, IT can act to better align technology to user ne


Related Links:Security Now PodcastAuthor Ryk BrownRyan's Goodread ProfileRyan's LinkedIn ProfileHeather's Linkedin ProfileLinda's LinkedIn ProfileIT Lifeguard Podcast FeedThe first episode recorded on the Future of WorkFollow us on Twitter: @thedwwpodcastEmail us: podcast@digitalworkspace.worksSubscribe to the podcast: click here

Transcript

Ryan Purvis  0:00  
Hello, and welcome to the digital workspace works podcast. I'm Ryan Purvis, your host supported by producer Heather Bicknell. In this series, you'll hear stories and opinions from experts in the field story from the frontlines, the problems they face, how they solve them. The years they're focused on from technology, people and processes to the approaches they took, they'll help you to get to the scripts for the digital Express inner workings.

Heather Bicknell  0:29  
Hey, there, it's Heather. I'm just jumping in here to let you know that the episode you're about to listen to is actually a rebroadcast of one that Ryan and I recorded in early 2019. For my other podcast lifeguard it, we thought it'd be a great fit for our listeners here as well. So without further ado, enjoy the episode.

Hello, everyone. I'm Heather Bicknell, Product Marketing Manager at Lakeside.

Linda Tsao  1:05  
And I'm Linda Tsao. I'm an engineer at Lakeside.

Hey, guys. 

Heather Bicknell  1:09  
And together we are the hosts of the lifeguard it podcast. Thanks for joining us today. We have a very special episode for you this week. If you're a longtime listener, you might know that we alternate our format between sort of high level discussions on topics and in depth interviews. So it's always a treat when we get to interview someone in the field. For this episode, we had the opportunity to speak with Ryan Purvis, who's the CIO of HiLo Maritime Risk Management. Ryan has a background as an IT leader at both enterprises and startups, which we'll hear more about in the interview. Specifically, we wanted to know what challenges are top of mind for CIOs in 2019. And also any advice Ryan had for tackling large projects?

Linda Tsao  1:57  
Yeah, Ryan, so it was really fun to interview who it was super interesting ideas.

Heather Bicknell  2:02  
Yeah. And he's also a longtime super user of our products SysTrack. So in this interview, he also talked about how he's used to track his previous roles, how he's using it now, and what plans he has for it in the future. Some of the topics that we cover are gamification, we talk about automation and artificial intelligence. And we also go into how to better understand users needs and experiences. So if any of those topics are interesting to you just keep listening. I'll also link any additional notes as well as how you can get in touch with Ryan in the show notes. So without further ado, here's the interview.

All right. Well, we're here with Ryan Purvis uses CIO of HiLo Maritime Risk Management. Thanks for joining us on the podcast Ryan. 

Ryan Purvis  3:01  
Thanks for having me again. 

Heather Bicknell  3:04  
Ryan's been on the podcast with us before actually about a little over a year ago. So this is good timing is great having you back. So just a little warm up question. I know when we had the past interview just on our past conversations, it's been really apparent that you like to read a lot. And I'm just wondering, how do you find the time you know, in your busy CIO schedule? And also has there been anything particularly good that you've read recently?

Ryan Purvis  3:33  
Yeah, it's one of the things I always make time for. I've always always been a big reader. And I made a conscious effort or decision maybe two years ago to cut back on Netflix and the like, to rather read. Sorry.

Heather Bicknell  3:52  
I said that'll get you there.

Ryan Purvis  3:56  
Yeah, so um, and actually at the same time, found out, I listen to a lot of podcasts as well. And I guess the time that I have issues is on the commute or or at night. And one of the recommendations was for an author or a series written by an author, Ryk Brown, or Rick Brown, I'm not sure I pronounce it. Which was it a science fiction futuristic series. And the podcast, I think was SecurityNow. And and the guys who sort of regulation said, you know, he picked up one book that he couldn't put it down. And I obviously took it and started reading the same series. And I think I finished the entire series in about a week. And it was it was quite bad in the sense that I would not sleep because I'd rather just do one more chapter one more chapter, one more chapter. And that sort of kicked started me back into my old reading habits and my mom would say I of all her kids, I'm only one that really reads a lot. And I read anything, provided it's in book format. I struggle with newspapers and struggle with magazines, but if it's his knowledge that I want to get, it'll come out of a book. And I find it's a nice way to to get a familiarity with a subject before going to talk to someone about it, because the next part obviously, is to talk to someone. Yeah, so last year, I set a goal of 52 books, which went past a little bit. This year, I've set myself a target of 60 books, which is basically a book in about a week. I suppose. The other thing to mention though, is I read more than one book at a time. So often, if you do see my I think they post on Twitter through Goodreads. They seem to close off a lot at the same time as because I probably finished three or four books that I've been reading slowly in parallel. And I finished them all the same time as well.

Heather Bicknell  5:43  
Yeah, I keep falling short of my Goodreads goals. So some great tips. Yeah, I think definitely cutting back on the Netflix and just prioritizing it more is a good thing. Yeah. So getting into some of our interview questions. So you know, we've talked a little bit before this episode. So last time we talked to you you're, you know, working for a large bank. You kind of have a background there. But you're, you've transitioned to being the CIO of a startup now, I'm just wondering kind of what does that change been like for you? How is it then to go from, you know, managing thousands of users to being at a start up?

Ryan Purvis  6:30  
Yeah, so actually, my career started in the small business. Business sort of 10-15 years ago, was a thirty man organization. And when I moved to the UK, that was really my first taste of corporate life to two banks, one real estate company, and another startup slash cloud business in between. And when this opportunity came along, it was it was going back to things that we're more comfortable and no, no, that was uncomfortable in the bank at all, but when you're in a larger organization, there tends to be a lot of people that are, are doing pieces of a role or a process. Whereas in a startup, you tend to be a multi hatted, multi skilled, because you have to, there's no one else to do the work. And that really appealed when the opportunity came along, to get my hands dirty, but also have a more strategic view on things. Yeah, so that's that's the the driver was to be most it was to try and find the balance between the two. And to be fair, you know, my last organization that was, was going in a good direction. So it was quite a tough decision to make to move, but it's all about timing.

Linda Tsao  7:39  
have you noticed any technology changes in your new role that maybe you didn't experience in your past role?

Ryan Purvis  7:46  
Well, it's funny because, you know, HiLo is a risk business. And it's specifically focused on on making the the shipping industry safer through risk management, so taking information in processing and putting it out through the risk model. So we're tending towards the banking mentality anyway. Because because banks tend to mitigate risk as a primary objective to service the customers as well. I suppose the nicest thing is being able to put in technology quickly. Without the bureaucracy of a big corporate. I think that's just, you know, the start up nature but we're attending to all my experience from working in the in the banking sphere has has come up quite a lot because we, you know, we do an ISO 27,000 certification at the moment, which has meant going through audits and making sure you have the policies and procedures, the controls in place, which are all, you know, day to day, activities in a bank anyway. So in that sense, the only real change is being the flexibility to put in a product because either you know, it's all it meets the need and you want to do it quickly. Because this is a lot to do in the startup world.

Heather Bicknell  8:59  
Do you have anything projects on the horizon for 2019. I know you mentioned getting that certification.

Ryan Purvis  9:06  
Yeah, so it was when I joined HiLo, the biggest problem they had was getting data from the customers in a, in a regular uniform way. So the last two, three months have been about building an automation pipeline to bring the data in. And because a lot of the data we get is human entered, there's a need to interpret that information. So we'll be going down the route of an NLP or natural language processing. And we need to scale we've we've got a rather large footprint of clientele to to onboard data from and quite a lot of volume as well. So that'll be the main driver for us is to is to automate using NLP and machine learning. So I AI sort of technologies that you need nowadays it needs to be in the basics, and then we'll be expanding globally so there's a need to look off distributed users and distributed customers shippings actually quite different to banking. In essence, it's very global, you know your clients all over, because we're, this is more client focused. So when I say bank is different, it's different for me in the sense that we were dealing with the clients all over the world. So you have to be conscious of what the demands are from a legislative point of view or even from a cultural point of view, and how we can building trust with them to get the data that we need in order to process for the risk model.

Heather Bicknell  10:32  
I'm interested in hearing about that. So the outcomes of that project because of course we've done some of our own work with NLP and are starting to dip our toes into AI with the new stuff coming out of SysTrack. You know, speaking of our products, SysTrack I know you've been using it for many years, but um, I've never heard the full story. So how did you how did you start using SysTrack and kind of when  did you learn about product?

Ryan Purvis  11:01  
Yeah, so I mean, he strikes us as the UK. We, when we made the decision to move here, and one of my old clients who was with the bank that I ended up working with, he had the opportunity to come on board as a product manager looking after SysTrack the sort of characters was a self healing desktop project that they wanted to pick up and run and subtract was the key component to that. So when I when we came over and when I had to make the decision between corporate life or staying with the, in the sort of solution vendor, system integrator space, it sounded like an interesting opportunity. So, I came on board, SysTrack, it already been deployed to a small number of users. And one of the main targets was to push it out to to the rest of the bank. This is all the investment bank side not not the retail. And the reason why that distinction is that obviously this There's different factors you think about when you're dealing with retail branches, they typically have low bandwidth lines and that sort of thing. Not that that was a reason not to deploy it was also they had a different product out there Aternity. And there was a need to do some, some test building before switching one after the other. You know, we went from sort of 80,000 desktops turned 250,000. In that year, there was quite a lot that I needed to learn around Citrix architecture and how to build up the environment. But the the interesting stuff actually was around the self healing, and very much with 8.4 is now we were trying to do that with the product then. So we're sort of talking seven years, seven years ago, six years ago. And it was a lot of reverse engineering in senses and obviously with with leg size involvement, but connecting to the local database, doing some work there taking the resolve rules, trying to do automatic actions with that, and my sort of experience as a developer helps just writing the code and, and doing some of the checks. And we worked on how do we give the end user a perception or, or at least, if it wasn't a perception of they had a good desktop, we would know what was wrong with it. And this was also tied in with a VDI deployment going on at the time. And non persistent VDI, which was moving away from persistent VDI, and moving users off of physical desktops. So that's how it all started. It was it was a nice blend of using all my experience skills and learning a whole lot of new stuff that I had no idea about.

Heather Bicknell  13:38  
That's very cool. Do you remember what any of those automated actions that you that you were programming in?

Ryan Purvis  13:46  
So we had a few pockets? What what actually ended up happening is I almost ran a program where we tied together multiple solutions. So over time in any organization, you'll have teams that will build their own solutions to to meet problems. So a large part of what we were doing was to pull the data together from these different solutions. So we had some operational guys that looked after the, the non persistent environment have built their own set of tools, which used to strike data. So we would do tying they would, they were doing automations on the desktop to fix a profile or, and there'll be the user profile or the add the profile, there'd be fixes as you look for in the mounting of applications. So you'd be looking in the registry to fix a key or change a key, they will automatically reboot cycles. So forcing a user to, not just forcing them without communication, but they get reminders to say that they're going to reboot the machine in two days time, one day time. And then today, reminders around saving documents, giving them a title so that automatic saving would work for solutions. So part of this was one getting the users to understand what they had to do. The other part was to give the operational soldiers sort of level three, and even engineering resources away to deliver solutions to the desktop so that on a trigger or an alarm, you able to do something to fix the desktop. So something like a process taking too much CPU, or too much memory, the tooling would receive an alarm from SysTrack who is doing the monitoring that will trigger a pop up to say, you know, Dear Mr. Customer, your your outlook is using too much CPU. Can we close it? Or are you do you want us to wait, and then if they keep waiting, there'll be a retry cycle if that CPU is still being held by that process. And part of this was was a lot of this was educational, the end users too, because of you moving them away from their own isolated desktop, you know, a laptop or a physical machine to virtual desktop, which is obviously a shared infrastructure, and where those sort of runaway processes could consume all the resources on a rack. Other cleanup things that we So with Internet Explorer, various versions of that, not all of them close their sessions correctly. Either to do to buggin into Explorer or bed, Cody aware applications to have to clean those things up. And it was about finding those things. And that's why it's this check was quite useful because you could look at what was going on to the user session to the high amount of memory we use what was causing that and then put in place solutions to to fix them.

Heather Bicknell  16:28  
I love that you're kind of ahead of the curve there because of course, we're you know, we're dedicated now to to building out more of those automated self healing actions and the product. So it's exciting to see but I love that you were doing it. Before before we started making it even easier for people. Do you remember any of the outcomes or like, kind of what benefits you started to see once you had implemented that?

Ryan Purvis  16:55  
Yeah, I mean, the main driver for these things was ticket reduction. A ticket would have evaluated dollar value every time one was created. So if you could avoid a ticket, and I think the average is one ticket was a goal was one ticket per user per month. And any way you could reduce tickets to meet that, and I think the L three got it to that level or just under a new one. Bear in mind, some of these tickets are going to be noisy tickets to the sense that they are. If they haven't received a response to the first ticket, they're creating a second ticket or a third ticket. Sometimes it's I need help with the business application, which has got nothing to do with the technology platform. But then you do have the tickets like I can't login, my password needs to be reset, those sorts of things. And that's why automation really came in quite quite well and quite strong. I think password resets is so one of the biggest top talkers when it comes to tickets, but even then, you're sort of next ones or faulting applications or, or problems with add ins and that sort of thing. In that way, there was some good results, the reboot of a machine that after after a week of running would definitely give a performance improvement. So we did see a down, downsize and that I can't give you exact numbers because I can't remember the stats. I mean, in some sense, our automations created tickets to because users do like the pop up messages or didn't like the reboot cycles. But that still sees education. And that's one of the things we did in the last bank really well was was communicating with a user showing them what we could see. So showing them something like SysTrack resolve, showing them the black box data recorder, showing them the memory usage, you know, being much more scientific about what spec machine do they actually need, as opposed to the opinion of well, I need a big machine because I do big things. The other data have really helped to to crystallize them. So I think there was a lot of a lot more transformative of benefits than actual just ticket reduction in the end.

Heather Bicknell  18:59  
Yeah, I think what you pointed out there, but that communicating with the users around changes is so important. So they're not like this now where like what's happening on my endpoint?

Linda Tsao  19:11  
Yeah, so you're talking a little bit about or you mentioned a little bit about your big projects. So like desktop transformation projects, or migration projects. So I guess I'm curious, what are some of the challenges that you've experienced from them? And do you have any like tips for anyone that might be attempting these projects themselves?

Ryan Purvis  19:33  
Yeah, I think one of the biggest wins for us was knowing our customer. And I'm not talking so much about software besides the face to face, knowing them. But understanding what the profile actually was today, what kind applications they used, what machines they had under the under their name, how you know, instances where users have multiple laptops or multiple desktops, what kind of files this stored where your network folders, shared mailboxes, and so on. So knowing knowing that aspect of any co user so almost moving towards personas, that was one of the first key things to to get to grips first. And then looking at those applications specifically around what would be good for the platform's. specifically looking at licensing as well, have you got the right license for that user? So an example would be something like Adobe Reader can be used for most use cases, but there's very, very rare that you actually need Adobe professional. And obviously, if there's a need for professional making sure that the use case matches the license cost. The same with something like office 365, what sort of Cal's should you be looking for an E3 versus an E5? And really just being being clear on when something goes onto the new platform? You know, are you Is it the same software just been repackaged? Or is it a new version that as potentially a new interface or new performance load. So one of the things that we did quite well, in both organizations where we did this was having an application testing function on the build. As a build goes through this process of being put together in any of the business core business applications are put on top of it. There was profiling done of how did that impact the performance of the of the build. You're trying to keep that footprint as low as possible, with tuning down the sort of noisy things like antivirus, for example, from making too much or taking too much resource when they were were a lot of that with the anti malware, anti spyware, software that needed to be tuned as well. And some of the newer technologies like

VMware,

or they call them application, application stacks. There's a lot of exclusions you need to put in place so that you know just knowing what those Things look like. And testing that a lot before it goes out to customers is quite good to do. The other thing, obviously, is to have waves, you know, wave zero wave one, wave two, where customers will get the new build different different times. So sort of having a wave zero for the it folk and then wave one for the friends and family wave two for the next small batch people. So if you just need to roll back, you've got an opportunity. I think we did really well which, which was to compare old builds versus new builds and physical versus virtual. So if we had a user that was on virtual, we'd know what the physical footprint was like. You know, obviously, one of the first sort of challenges you'll get is it on my desktop was better than this. And in most cases, you're you're looking at the data. It wasn't you could see clear things were the VDI infrastructure was performing better. But because they didn't have the laptop or the desktop underneath the desk, they felt that they were There was somewhat robbed, in a way and yourself is that there's the converse, there will always be there will be some good, good, worse experience. I mean, something like Visual Studio on a VDI, it's quite a big application, you really need to fine tune it to run on the VDI. So you can see, you know, the converse sometimes. But that's the point is you're looking at this from a scientific point of view as opposed to opinion. So that's and that's the, you know, the one of the keys as well is to keep it unopinionated as possible and as factual as possible.

Linda Tsao  23:33  
So I guess what did you do with users that had all the resources but were migrated over to VDI and therefore they were kind of like you said uncomfortable with the whole situation.

Ryan Purvis  23:45  
It depends on the situation. Often we would show them the data we had. We did three stages of an analysis we we looked at the sort of three common four common resource pool is IOPS for your desk, CPU, memory and then applications that were in use at any given time. Some of this was it was educational user. So teaching or showing them and things like closing too many windows that open. Not having too many applications open at any given time. There was some gamification that we did with with a portal. So almost scoring them on good corporate, good IT corporate practice. So if you, if we took, we took the subject health score, and we merged it with the almost the recommendation health score. So if we gave you 10 recommendations, and you've been your middle 10, you got 100%. And that was average that was the Health. And that gave what the intention was to give the end user some some skin in the game, to to be part of because it's sometimes it's been part of their part of the infrastructure now that the service and the infrastructure if everyone behaves all the infrastructure performs on their own but barely we have the converse. And then went off pretty well. In some cases, some cases people didn't like being scored. In some cases, there was competition around the school, which was the right thing. But at some point the you could only get that score so good that I think we only really touched on the on the edge of what we could have done was gamification. We could have looked more at leaderboards and actually across sort of non non technology related items such as service requests or ticket creation, you know, losing points or creating unnecessary tickets, for example. Yeah, it the gamification I think, will will be more prevalent over time. I think it's if he's proven itself to work, and you'll see it I think we'll see more and more in business applications.

Heather Bicknell  25:40  
I think that's good. I think it can be, you know, it makes it makes it so you have some awareness, I think of what's going on with your, with your computer without being you know, too technical or overbearing in some way I think. And then it allows sort of that we talked about the concept of level zero, which you where, you know, users are using self service and taking the resources that it gives them to remediate things themselves, they don't actually even enter into that helped us cycle, right. But it is really important to, you know, find clever ways like gamification to make that actually possible and at least fun for some people. One of the reasons we wanted to do this episode of us that you've recently deployed to SysTrack, check out your new company. So I'm just wondering kind of what is that deployment process been like for you?

Ryan Purvis  26:36  
Yeah, it's a lot of fun. We're trying to use Intune to deploy it. So that's that's had some interesting challenges, primarily because I don't think Intune works as well as it should, or at least from it should be a lot easier than it has been for us to deploy that. The main reason for deployments this track was around the ISO 27,000 process and what I expect to have from an order point of view and information security requirements. But it's in place now, where we're using it as part of diagnostics. So because I have people that travel quite a lot with laptops, if they have problems, I can at least have a look at what's going on. I make sure obviously, from a compliance point of view that all our patches deploy latest versions, that sort of thing. And before we lock down the desktop completely, knowing which applications users are using, so they can be packaged to be deployed so you can tune because that actually works really well. We're doing it is working, getting applications down. Even for us more user base, it's taken a load of me to make sure it's happening. And even with super the team we're using, it gives them some comfort that they know that they've got the right version of Chrome with the right extensions to use. Most of the most of us are safe seafarers with with very with a life technical background to me and technology background. And they're just more comfortable knowing that they've got the right stuff available to them.

Heather Bicknell  28:06  
I guess what are some of the so we've talked about, you know, how you've been able to kind of meet those, you know, start thinking about compliance and kind of audit everyone's desktops and also play off at your server location with SysTrack. What are some other ways that you're hoping to make use of this track in the future?

Ryan Purvis  28:29  
Well, as I say, from an audit point of view, it's around proving that we are monitoring the desktops we know what's happening on them. Anything it's installed we know about, from from as we grow and expand, there'll be a need to monitor desktops to support the users on them. I also would like to at some point, see them on ships, see SysTrack on ships, where we can also get some some insight to what's going on there because again, we're dealing with incidents that are happening on a vessel And as much as we have the human input, we're looking for the telemetry as well, to merge that together. So mechanisms that collect that data for us, would be helpful. So so that would be the ideal dream, at least is to get a more complete coverage

Heather Bicknell  29:19  
of the image of that use case of

on the vessel.

Let us know when you get there. It's very

Linda Tsao  29:26  
company name to Lakeside.

Yeah.

Heather Bicknell  29:33  
So our customers use digital experience monitoring tools for lots of different kinds of projects. Right? So we've talked about, you know, how you can plan a migration, how you can augment what the helpdesk can see, use it for root cause analysis, those kind of things. But these use cases can be so broad. There's a bunch of different ways that customers tend to quantify the impact of a digital experience monitoring solution. So I'm just wondering what is successful Em, it's kind of like to you How have you noticed benefits.

Ryan Purvis  30:06  
So I think this level of transparency that's important. Knowing how, when you when you look at the service are getting hard to how it fits together. And maybe this is the technology view. But I'd like to understand an end to end view of what I'm working with. So when something's not working, I'd like to know why it's not working and where it's not working primarily so that I can potentially get around it to get something out of it, get something working if it's broken. An example would be if I can log into my desktop because there's no internet in the room that I'm with because I use my laptop remotely. Sometimes. If the machine is down or the Wi Fi is down, I like to know what's broken. So for example, I would look at when I look at resolve, resolve comm connect to the desktop and over desktops open, but there's something else that's been causing the problem. So that's where I look at it. I think the other thing is that when, when you're working on on a device, you want to know that it says it's best to optimum state for you to be working on it. And if not, it's nice to know that at least the monitoring or the resolution is aware of what what the potential problem is and how to fix it. And I think that's something that I'm looking forward to seeing more and more with with SysTrack, in the sense of self healing or self repairing. And even with some of the stuff we did at in the banking, at least, where we can notify proactively, that there's a problem before it becomes a problem that someone logs a ticket for. So if the we had an issue that went down recently for office 365 I think was out in the States and affected a bit of Europe. If you were notified that your exchange was going to be a dowel, your emails gonna be down, so most of you would know the exchange. As he said, you wouldn't waste it. Trying to get it to work, which means you can focus on something else in that time. So that when it comes to experience monitoring, it's it's so much sometimes it's just about knowing that something actually is broken and someone actually is going to repair it, as opposed to us struggling with the problem, trying to figure out what's wrong.

Heather Bicknell  32:18  
Yeah, I definitely think the ability to kind of surface those really business critical insights more and more is going to be very interesting and very helpful for large organizations especially.

Ryan Purvis  32:32  
And I think there's even a level of just knowing today your experience is worse than yesterday, because of something we know it's because we had the windows 10. release a couple of months ago, I think we actually made it worse for users. There's some strange bugs that came in. If you know that that's happened and you know, the next one comes out in and there's a level of of differential Comparison to so today's version is better because we've removed 2% of the codebase. Because it was bad code. I don't know, whatever whatever the metric is, that's also helpful to know that things are improving or not improving why they're not improving.

Heather Bicknell  33:16  
Yeah, definitely. I remember. Yeah, some of those people's cameras stopped working. And that was interesting about a second happen off of new windows builds,

Ryan Purvis  33:31  
Sorry, one of the things we don't know was was to push out a monthly reports of load times for core applications, how long does it take for Excel to load take for words load, which is really about efficiency, or effectiveness of the platform. And it is ways to improve that. Again, a lot of these are, you know, affected by by a third party add ins. Because if you had nothing and if you use the sort of cool windows build with the core Office applications you wouldn't expect to have much delay because it should be it's Microsoft working with Microsoft, but the minute yiu put in other variables macros and and third party components. That's when you start seeing delays. And again, someone to be aware of that to be pointing out and monitoring it, you at least notice that there's a bit of visibility, a little bit of focus.

Heather Bicknell  34:24  
Yeah. What are some of those KPIs that you look for when thinking about improving digital experience?

Ryan Purvis  34:31  
So application load times is obviously one boosting scooter law one less that's one to look at as well. Average app average uptime. Windows devices tend to work better after a regular reboot cycle, ages of the build. So you typically find that a Windows device and this is anecdotal To be honest, not really scientific, nice to rebuild once here if you keep it on a desktop, whereas obviously VDI you rebuilding monthly, so it's less of a problem. And the other things are more around how the user uses the desktop. So they leave applications open and running in the background forever. And that could be a day, three days a week. And depending on reboot cycle, I have seen instances where we've had applications running for a year Just holding on to resources, and also not restarting the device. So you know, updates that need to to be run or installed. And that rely on that, that restart. don't fit don't finish the install. So that's that's another thing to sort of be aware of. Age of the updates, so things like antivirus updates and the actual version of the operating system are important to look so that doesn't matter whether it's Mac or Windows, making sure we're always running the latest. So if any of those things are not happening, to be signs of bubbling problems to come Then, of course, talking to the user. So I think the surveying tool, which is why I think it's an 8.3 feature, was something I was quite excited to push out. Even just talking to the user on the phone, sharing the screen, seeing what they do seeing what we see in the data, to correlate with what they experiencing. So for example, we might see the health score at 85%. And that's obviously the SysTrack health score I'm talking about, but does the user feel like that's a real, a real number to them? They feel like they're going to an 85 to 90% experience. And if they don't, you want to try and find out why they don't think that. And there might be a good reason for it could be that they, you know, have a certain way of working that the platform doesn't meet

Heather Bicknell  36:42  
you guys, we could probably just transition to sort of the predictions question unless there is anything else you kind of had lingering on your mind.

Ryan Purvis  36:52  
Yeah, so I'm actually thinking. I mean, this is probably the most interesting question in some senses because I think As much as things move really quickly, I think they're quite slow as well, I think this year, and then the next two, three years, we really around AI becoming more and more prevalent, or the barrier to entry will be a lot, a lot less. And I think that's going to appear more in the codeless. way. And if you look at how Microsoft has released Power Apps, with dynamics, and dynamics AI, as an example, I think you'll see the ability for for anyone who's got a bit of time, and is willing to learn a little bit that they can actually deploy something. Some applications they're trying to use, or they need with a back end AI. And to be a simple AI will be something simple, like recommendation engine or q&a tool or whatever it is. I think you'll see a lot more of those arriving, which is tough for business sometimes because users can do stuff in the consumer space they currently do in the business space because of all the lines downs and, and restrictions protect the business. The other thing that I think would and I've just experienced as myself buying an iPad Pro, is that the ability to work or for sort of an intermediate device, which is what I consider this to be, and then have a hosted desktop, we actually do feel more heavy work. So have that hosted in AWS or have it hosted in Azure will give you much more flexibility to work anytime, anywhere. And then carry obviously a list. My, my laptop was pretty nice as it was, but, but it was a big thing to carry around. But an easier commute or a more flexible way of working, which I think is quite exciting. And then I think the other thing which we're seeing, some of the work I'm doing with some other companies is around boutique subscriptions, and having those become more granular so that you can build your own subscription and you probably find Maybe subscription aggregation did a lap saw happening? Very much like Netflix is as done with with content. And obviously now making their own content. I think someone's going to figure out how to do subscriptions that way. Which should be interesting.

Heather Bicknell  39:13  
 Definitely. 

Ryan Purvis  39:14  
Oh, there's one other. So one last thing was was gaming as a service. So we've talked about gamification. And I'm interested to see how, with the likes of Microsoft gearing up in AWS, I think it's just announced as well, to provide better mechanisms, stream games, again, or any device anytime. But I think they'll open the door for better for change in education and retraining and changing in teaching. Where I think a vocation obviously, on its own, already has a huge footprint. But the way we do education is going to shift more and more using the sort of tech better of flavor.

Unknown Speaker  39:53  
It's funny because you said gamification, I was thinking, Oh, like SysTrack, I can have its own like leaderboard. Turning SysTrack into a game I wasn't thinking about

Linda Tsao  40:02  
the actual, like video games aspects? Well,

Ryan Purvis  40:10  
that's what I did. That's that's exactly I thought you could I don't know. I mean, I don't think you need to write a 4D game. But I think you would be able to have you know, your your certification, be a first person experience where you go in, in this in this realm handling a technical problem with with the leaderboard and who does the end and you could set some interesting puzzles.

Heather Bicknell  40:40  
Yeah, I know there was kind of an article we covered a

while ago kind of related to this idea of bringing more tech into schools where I think it's a Google technology where you could basically do like a remote field trip and explore like, you know, the pyramids of Egypt from, you know, your eyeglass or whatever you have in the classrooms. I think that definitely makes sense. I haven't heard people haven't heard a lot of stuff about that, but I can definitely see that happening. Yeah. So I guess was a if people want to get in touch with you, Ryan, is there any good way to reach you?

Ryan Purvis  41:16  
Yeah, I'm on the usual sort of social media channels. So Ryan Purvis on Twitter. We're at Ryan Purvis. If I'm on Twitter, LinkedIn, find me there as well. And I think I'll just give you I'll give you the link so you can put them in the notes. And that's probably the best place to get hold of me.

Heather Bicknell  41:34  
All right. Well, it's been great catching up with you, Ryan. Thanks for coming on again.

Ryan Purvis  41:38  
Thank you very much opportunity over boys. Enjoy. Well. Thank you for listening to today's episode. Hey, the big news app producer, editor. Thank you, Heather. for your hard work on this episode. Please subscribe to the series and rate us on iTunes or the Google Play Store. Follow us on Twitter at the DW w podcast the show And transcripts will be available on the website www digital workspace works. Please also visit our website www dot digital workspace works and subscribe to a newsletter. And lastly, if you found this episode useful, please share with your friends or colleagues

Transcribed by https://otter.ai