skip to main content
Media Window
Author of the Resource Posted on by Phil Morgan in Product and company news

Highlights of Software2's User Day with Edinburgh Napier University

Another Software2 User Day, another success! December 3rd saw Edinburgh Napier University host Software2 and friends in their beautiful Rivers Suite for a day of discussion and collaboration when it comes to higher education software delivery.

We'd like to take this opportunity to thank Edinburgh Napier for their incredible hospitality as well as all of those in attendance, to whom our User Days owe their success! Videos of each of the presentations are available here; watch them below or use the panel to navigate.


What’s new at Software2?

 

Transcript

For those that don't know me. I'm one of the directors at Software2. I've been there about seven years. I was employee number eight. I'm actually based in Inverness. So this feels like my user day too. So I'm pretty excited about that. Out of all the people that we employ from Illinois to Austria, I'm our most northerly employee. So yeah, it's our 10th year anniversary. We've been going 10 years today. But I thought it was also interesting, it's taken us 10 years to hold a user day in Scotland. So when we finally done it. It's our first Scottish user day, we've got six universities at the moment in this part of the world. We might have seven if they're here today, so I won't embarrass them by saying who they are. But hopefully they come along and join us.

So just to go through some of the things that we've been up to this year. So one thing that we set out the start of the year, as most organizations do, we set our aims. And we really decided that we were going to refocus on higher education. So over the last two or three years there'd been interest from outside of higher education in our products and services. And we've kind of been intrigued and an interest in that. But we felt that as an organization, our values, our principles and our product really aligned with you guys. So we decided that we would refocus and everything that we designed do and develop and the way in which we operate and work is really about you guys. And about making sure that anything that we develop adds value to you and your student experience.

We've got some ambitious plans and some ambitious aims as an organization. So we've grown mainly in product development so we can enhance and improve the way in which we develop our products. To enable organizations to have more choice and more flexibility about how you work with us and Ryan will tell you more about that. I didn't do this deliberately, but I know after I put this together, the gray people are actually in finance. I don't know why they both have gray pictures, but I thought that was kind of curious after I put it together. So we've also added some people to help obviously as an organization that's dispersed over the world there's lots of different rules and regulations at different parts of the planet that we have to deal with.

I think excitingly we moved into our pub in the UK hot office. Well you recognize some of these people, but obviously Tony here playing darts. So that's a board meeting, a dartboard meeting. And you know that's given us a great place for us to call home. We were in a fairly bland ragis office before that. And now we feel that we've got something that represents our identity, which is kind of social and inclusive and fun. And we don't take ourselves too seriously. But we care passionately about what we do. So that's also given us the ability to grow. So we've got space in there for the future as well.

Just in terms of where we are. So this isn't an actual map. But obviously the Atlantic is a bit wider than that, but I've condensed it onto here. So you can start to understand this is our customer base across the Northern hemisphere. So the majority of our customers are represented here. We do have some customers outside of this zone. Especially exciting, we started with two universities in the Middle East this year, at Khalifa and Sharjah. So they're not on the map. But we're seeing the fastest growth I suppose that the most people coming to us from Europe and North America.

Just in terms of user days. This is our sixth user day this year. So we've done one in Toronto, we've done one in Germany, one in Spain, we did one in Middlesex. I know a lot of you folks have came to as well. So it's over 300 representatives from over 100 universities at our user days this year, which we're really proud of.

Okay. And then just to give you an idea of some of the things that we're doing with our customers. So really we see successes when we get to the point that an organization will work with us to help promote how they're doing things. How they've improved. It was wonderful earlier to hear that Claire was talking about her NSS results and that we may have been involved with it, the impact on that. Because really, our success is your success. At one show at the point that things are improving and people are happier with the service that you're offering them. We can feel proud of what we've achieved. If you want to go and check these out, these are on our websites.

So these are our webinars and videos that our customers have recorded about their projects and what they've done. Obviously this then starts to get picked up in the press. So university business and people like that start to want to talk to us about what we're doing. And share the knowledge that we have. Which is really the knowledge that you share with us and give to us. So it's a real group effort. And then there's plenty of ways to keep up with what we're doing and what our customers are doing.

Okay. So just a quick overview of what we've launched this year. So at the start of this year in spring we launched the packaging subscription. And the idea behind that was to try and enable people to be more efficient and not every organization packaging everything up from start every time. And really that was from feedback from these types of user days where we heard what you were saying, "Well actually why are we packaging the same app that the next university is packaging? Isn't there a way that you can help us out and deliver something that we can all use in a more cost effective and efficient way?"

At Middlesex, at the user day that we launched a version 2.7 which had a range of features. We also launched Analytics, which is a reporting engine that sits inside and monitors and looks at all the data and all the launches. And all the information that is captured inside apps anywhere. So that's really to enable you to get well, more return on investment from all the work that you're doing. And then today we're also going to share with you 2.8 and the features that we've been added to that. And we'll share some of the things that we've got planned for next year as we go through.

In terms of next year, we've got some really ambitious plans to grow. So just to give you an idea, over the last three years we added the same number of customers. So we have to support and develop for the same numbers customers as we worked within the first seven years. So we've kind of doubled the number of people we have to support and develop for. And we also need to find a better way to support and work with those people. Because well, they are all over the world. So it presents challenges to us and we've got some ambitious plans. There's some things that we're bringing into the product to enable you guys to have more choice about how the product works and how you stay up to date.

We want to maintain our satisfaction level, so we have a 98% net promoter score. Which is classed as world-class. But we want to maintain that. And the big challenge that we have and we're focusing on, is also ensuring that we don't lose too much of our identity. So as we grow we try and stay kind of software too and people represent our values and our philosophy and don't take themselves too seriously. We don't want to become another Microsoft or Citrix or a kind of faceless technology company.

So these superhero icons are where we've got open head counts. So people that we're looking to employ next year. We've also got some ambitious plans as I mentioned around the technology, around what our new delivery methods are coming out to the market. And making sure that we support and work with those about some of the integrations that we're going to tell you about today. And it's all higher education focus. So as I said earlier, everything that we're doing now, all the development is about getting more value, more features. More things that make hopefully your lives easier as a user of the product.

Okay. And just, before I finish, sorry, I'm going to embarrass you. But we love this video. And we think Michael, you produced so many cool videos about how you use apps anywhere. So we just thought we'd share this. If the sound works, I'll be amazed. But let's see. So this is Michael in his, is it a Land Rover? Looks like a Land Rover. He's in the middle of nowhere. And he's decided he wants to open a document or launch an app. So he's going to go into his car. And Apps Anywhere is here working from the dashboard of his a Land Rover. Is it connected to the internet? It must be yeah. So connected to the internet and launching apps from within is Land Rover. It's just proving that he is in the middle of nowhere. He's not cheating. But yeah, I suppose the challenge is if anyone can come up with a more innovative way of using Apps Anywhere, then we'd love to see it. We were amazed by this.

Okay, cool. It's on our website as well. And thanks Michael you're always kind of pushing the boundaries of what is sensible use at least. Okay, great. That's me. If anyone's got any questions or anything, wants to know anything about Software2, or any of our plans, then please feel free to talk to me.


Edinburgh Napier's software delivery estate

 

Transcript

Hello everyone. So, my name's David Pentland, and I am a client services team manager here at Napier University in the information services group. I just thought I'd give you a very brief summary as to how we've got from where we were to where we are, and then introduce you to Clem is going to tell us a bit more about what we've done with AppsAnywhere here.

So, when I started here many, many years ago, this was the sort of PC that we had in the university, no hard disk, no networking, and basically the only way to get your operating system and your applications on there was via floppy disk, which you put in, switch the machine on, and it loaded them. So, you'd have to do that every single time. It worked. There weren't very many of them around, and we had very few applications, to tell you the truth, and the applications were so small you could load them at that time reasonably quickly. But that was the way it was.

Then moving on into the 1990s, we got hard disks in the machines, still not networked, so we would still have to go around every machine, load the operating system on and the applications on. Though this time, three and a half inch disk were not so floppy anymore, but they're still there. But anything that needed to change, you'd have to go around every machine and redo it all over again. It worked, but upgrades and changes were not easy at all.

Then the system here, where the operating system and the applications could be loaded over the network, so we had networking now and we had hard disks. So, a server could be built up with a bunch of files on there, which was basically Windows 3.11. When you switch the machine on, they'd all get downloaded fresh every time, so any applications could be put down as well. Again, the applications weren't totally big, weren't very many of them, and the operating system was fairly small. So, that was quite a good way of actually making sure that all the machines were completely set up and were working every morning when they got switched on, and if there was any problems with the machine, you can simply switch it on again and you'd get a fresh copy of everything down.

Data in those days was held on things like these. I've got a zip drive disc if no one's ever seen one before, so quite an interesting historical item. I'll leave that around if you want to have a look at that. And software in those days came in boxes. Don't come in boxes anymore. This one says, "To save wastage, only one disk size enclosed." So, this one happens to have three and a half inch, which says you're having a [duel 00:02:21] box with five and a quarter in it as well. So, that was how we installed our applications.

That obviously wasn't workable when we got to bigger operating systems like Windows XP, so at that point, we created a reference image, we Sysprepped that. We used PXE boot disks, which were okay to get the image across using a product called Ghost onto the hard disk of the machines, and we'd do that for student labs once a year, usually in the summer. And applications could either now be built into the image, or we could actually serve them from a data store on a share on a server. So, a bit more flexible. We could update some applications on the server, but if we needed to do anything that was in the image, we'd have to basically rebuild the image and put the new applications in, so better but not ideal.

hen things are getting really complicated, so we managed to get rid of the PXE boot disks and do PXE boot over BIOS over the network. So, that was an improvement, much easier to manage, but we still had an image with, this time, Windows 7 in it. We're now using a Windows deployment services instead of Ghost, to move those images onto the machines. We've got a script that we ran after imaging called Wind [inaudible 00:03:35] which was basically a way of configuring a PC with anything needed in a specific area. So, we'd have to decide, "Oh, this application needs to be here or this application needs to be there." Wind [inaudible 00:03:45] would then install that locally after imaging. We've also introduced Novell ZENworks applications, which could deploy applications through either MSI or to simply link to them over the data share onto PCs. You could also do a bit of a overnight application deployment. And we also had a virtual desktop service by this time where we have a golden master, which has some applications in it, and we also ThinApp for some of those applications, which is a virtualization technology from VMware.

So, basically, everything was getting really complicated because applications are now available from all these different areas, deployed in all these different ways, and unfortunately, it usually meant that nothing was quite where it should be for what the students and staff wanted. And it was really hard to manage. So, finally, about a year and a half, two years ago now, we decided that AppsAnywhere was a good replacement for all these different methods, and in combination with SCCM, and possibly Intune in the future, then we are pushing applications out through SCCM. We are still doing some imaging, but in the future we may just take the operating system that comes on the machine, make sure it's an SCCM and make sure the applications are then deployed in a scheduled task, and everything else is coming from AppsAnywhere.

And that's allowing us to serve all the applications to people where they need them on both physical machines, on both virtual desktops, and increasingly on mobile devices, which is a bit of a thrust for the university to try and make people more easily mobile with laptops and so forth, and working from home as well. As Claire said, we have a lot of overseas students, and AppsAnywhere is helping us deliver those applications to them as well. So, that's a whistle stop, too, of where we are, and I'll now let Clem tell you a bit more about how we've AppsAnywhere here. Thank you.

Hello everybody. I'm Clem Clark, lead technical developer in the client services team. I'm about to give you a quick idea of what we do here at client services team. Probably where most of you are really an end user computing area, dealing with all the desktops that are out there. Windows, Mac, all the MDM devices, tablets, and phones. We've also got a virtual desktop infrastructure, as Dave mentioned earlier, which is a little bit more of a challenge than the standard physical desktops that we have, and of course, we have to deliver the applications and licenses across all of these devices. We also have to deal with the printing and MFDs, as well as delivering everything from SCCM updates, Windows updates, group policy, basically just keeping all the physical devices and virtual devices up and running, which is a bit of a nightmare sometimes. I'm sure you will all agree. It's quite a lot of areas to get your hands stuck into here, and quite more often than not we also end up dealing with not just third line support, but second and first line support coming our way as well. So, trying to keep projects running is a bit tricky because we're always having to deal with support on the backend.

Thought I'd just take you through a bit more in depth from what Dave was saying. Maybe not go back to the 1980s to get to AppsAnywhere here. But quite recently we were using Novell ZENworks as our configuration manager. Novell ZENworks is a big configuration manager, a bit like SCCM, we were really using it more for applications and full application delivery. There was a couple of little tweaks that we do for modifying the registry and so on, but it was really our application delivery method completely. So, there was a portal of sorts that students and staff could use to be able to launch an application, and when you launch this application it would just, essentially, install an MSI or it could run an executable directly from the network. Obviously, installing an MSI on demand is not the best way for students to go about things when you're in a class. Some of these applications can be huge and take a long time to install.

So, unfortunately, we couldn't always use ZENworks to deploy that applications on demand. But it did have an agent on the device which runs the system user, so the students who obviously were not admins would not have any problems in actually installing any of these applications or did any registry tweaks they might have to do afterwards. It was very handy. If there was a problem, if an application was deployed, we could make a slight adjustment to the app in ZENworks, registry key, file delivery, permissions, anything like that, run a command line of any sort, and it would be delivered for the student.

The portal, as such, was a static window, which we made sure students couldn't close, but it also did publish applications in the Start menu as well. What we did have to do quite often, though, because we were delivering MSIs, most of the time through ZENworks, was not often builder on, although we have done it sometimes. We'd have to tweak the MSIs used to work correctly, so we used Adminstudio, which is the biggest, probably, product that you get for building MSIs. Most of the time we would just create an MST, which is a transform to see... If you wanted to install an application, you just needed to have a couple of registry keys or a property, and we use the MST to do that. And we use Adminstudio to build a MST as well. We had a license with ZENworks for that. It was a basic license. Sometimes we needed to do a bit more and we had to buy the bigger application, which was an additional cost on top of the ZENworks cost.

But as Dave mentioned, installing the MSIs is a huge task sometimes, especially if it's a big, big apps, so at Image Time, we did have this process of detecting what lab you're in, putting the big apps on afterwards, and making sure they were ready for the students as soon as they were there. One of the problems that we had is, with virtual desktop, obviously running shared CPUs memory on servers is a little bit more restrictive for the users, so we had to come up with some clever ways of getting applications working there. One of the security situations that we have here is that we destroy the desktops after you're logged off, so installing an MSI on demand not only takes a long time, but it was gone the next time you came to another desktop. So, you'd have to do it every time. This was not a good situation.

So, VMware, who we use for virtual desktop, had a product called ThinApp, which essentially, would capture your application, a little bit like you do with Clay Pigeon Studio. However, it was very, very clunky. It was basically a command prompt. It would say, "Launch your applications," launch it, and then you just, in the command prompt say, "I'm done." You couldn't really see what was happening. You couldn't really tweak the registry, couldn't really tweak the files properly. It was very difficult to use, and it wasn't very good, really. We could only get a few of the applications, a few of the easier applications, onto VMware. We couldn't put the big, big apps that everyone really wants to use.

And finally, it was licensed our VDI situation, but we couldn't use it everywhere. We could use it everywhere if everyone just paid for it, but it was very costly. So, we've come up to the situation that we're running MSIs for some people, we're running ThinApps for other people. Might having to install things at image time, and essentially, there's a bit of disparity between all the versions that are out there. Maybe the capture in ThinApp is not quite the same as the MSI that's been fully installed, so the students aren't getting the best experience overall.

So, I had this great idea to speak to Software2, get them in to see if they could unify all of this, basically. See if they could make it simpler to get the same application to everybody across the board, and they've done it. Simple as that. Instead of going to the AdminStudio to capture an MSI or to build a new MST to tweak the application, or instead of having to go to ThinApp to do it, we can just do it all in one place, keep it all uniform, make sure everybody has the same version across the board, and it has the same experience every time. We can deliver more on virtual desktops because of this. Essentially, the capturing process is much better. Much, much better in AppsAnywhere than it is in ThinApp, and we've got consistency. So, almost all our apps now can get to virtual desktop no problem at all.

What was really important for us is there is still a few apps that we want to run from a network share, small apps, want to run a few applications, and it has, obviously, it has that ability to do so. And bonus that we have is we can allow the students to access this from home, which we never had before, obviously. I'm sure management liked the fact that we have reduced some costs across the board as well. So, that is where we are. There's a few test ones in there because that's my own account there, but you've all seen the portal, of course, so I'm not going to dwell on it too much. But much, much nicer portal than we had than with ZENworks. And I suppose we could click on it. Will it work?... As live demos go, this is probably the easiest one to do, just thinking about it. So, this is obviously a Bring Your Own Device. I'm on the wifi here, so I'm not being passed through straight away.

And you can see the egg, as we like to call it, on the front screen there, which you passed this morning. It's very War of the Worlds looking device... So, hopefully it'll... Oh, yes, we're all good.

[inaudible 00:13:31].

So, as many of you will have, we can't have the full set of suites of applications available off campus or for Bring Your Own Device just due to some licensing restrictions. So, yep, another quick quick tour of how we got to AppsAnywhere, really. So, most people here are using AppsAnywhere. I think there's... I can see a couple of yellow ones who are just getting interested in it. I don't know how far you've gone with that yet, but most people will have experienced the portal. Most people have probably done a lot of Cloudified apps. I don't know if you've done any yourselves. So, I didn't want to dwell on much of that. I wanted to talk about things outside the portal and how we've managed to interact with Cloudpaging player rather than just talking about the portal and talking about Cloudpaging itself.

So, one of the things that Software2 provide is the PowerShell module. It's available in forums. Essentially, it lets you talk to the Cloudpaging player on the client. It's free to downloads, and here's a little snippet of it, if you're interested in a bit of code. You can see from this, it's just saying, "Stop Cloudpaging," and it's a simple function that will stop the Cloudpaging player. There's obviously plenty of functions in there, but we use this ourselves to try and give the user a better experience, particularly to add applications at login so they're available for the user. So, the example I've got here is Endnote, I'm sure everyone knows Endnote. It's got a Cite While You Write plugin for Word. So, if the user wants to use that plugin in Word, if it's not been loaded previously, they would have to go to the portal, load Endnote, which will no doubt bring up the entire Endnote application. They don't want that. They then have to go to Word and then run the Cite While You Write plugin to do their work.

So, we were thinking, should be put in the image? Should we make it always available? So, I came up with the option of using the Cloudpaging player to do the work for me, so when the user logs in, it puts it Endnote into the player and it's all ready. It doesn't launch it. It's all ready to go, and so the Cite While You Write plugin is ready to go. We could do other things as well, so we can detect what's in the player already, what's running, what's not running, and stop and start services as necessary on the machine. And we can also remove applications from the player, as well. To launch the application, all we need to know is the S2URI, that's the link that you click when you click Launch in the portal.

And this is probably the most exciting slide of the day, but this is an actual link from the portal. So, that that URI will launch Cloudpaging app, but maybe we don't' actually want to load the app. So, like with Endnote, we don't actually want to bring it up every time a user logs in, we just want to make sure it's loaded in the player so all the resources are there. To do that, unless I want to create a duplicate, which I don't want to do, I have to offer multiple delivery methods, which is on the left hand side there, for Endnote, and then create a new Cloudpaging delivery method which is set to not auto run. And when I do that, and it got to the portal, I can see multiple launch methods right there. And actually, if I hover over the bottom one, you can see in the link bar at the bottom, you can see that URI that I just posted earlier. And that's, essentially, what's happening when you're clicking that, it's just launching that URI.

Once we've got that, we can get rid of the multiple delivery methods, because we don't want to show the student that on the portal, but we must keep the delivery method there because the URI will still be active. So, here's a little example, PowerShell script that I have. Basically, a Start app is a function I've written, and I'm passing a couple of things to it there, the name of the application, as it appears in the Cloudpaging player and, of course the URI. And the commands that I use in my function, the first one simply detects if the app name, being Endnote X9, in this case, is found in the player, and I can use a simple PowerShell start process to run the URI, which will load it up. And that's part of my function there.

Let's have a little example, shall we. See if it works. So, this is the virtual desktop service. Hopefully... All right. We are actually, pretty much, Windows 10 across the board. I've just picked this. This is just one of my test pools. I picked Windows 7 because I've got quite a lot of stuff set up in it, just to make it easier for the presentation, but of course this works with Windows 10 not problem at all. So, this is how I was connected to the Windows desktop service. I haven't done anything though, and the app is launching right there. And unfortunately, I can't see the Cloudpaging player. I'll have to pop it here. Try that again. It's not loaded down there, but it should be. And you can see Endnote X9 is there and ready to go.

So, the next thing we've also done is use the network cache functionality, so rather than caching your applications directly onto the PC itself, we use the cache directly on the network. Increases performance slightly on large apps because it doesn't have the cache on demand. Although, I have put in there, we haven't measured the speed increase, so we take Software2's word for that, that it is working. But for us, it's very important because of the VDI situations. We want to increase speed as much as we can. VDI, when you get a new desktop, and you don't have something previously cached. Also, it obviously reduces our IOPS, which in VDI is extremely important. However, it does require a little bit of extra disk space, and you should have a a good fast connection.

So, we're not using this for home use. We're using this on VDI alone. There's a few things we need to get it done, which is the GUID of the app, we need to have the STC file, and we need to have a network point to store it on. So, basically, we need to get the GUID. Now, there's probably a better way of doing this, but the way that we would do it was we would either find it while we were doing the Cloudpaging capture in the studio, or even when we start the application itself, we can see it there. So, that's a standard Cloudified app just running directly in the player, and we can have a look in the log file and get the GUID. So, once we know that GUID, we need to get the file.

So, here's our STP, our Cloudpage application. If you unzip your STP, this is what's inside. If none of you ever done it before, I'm sure a lot of you have, maybe some of the newer guys, you will need to know some of this information for your exam if you do it. But the interesting ones there, the CAEs, probably, you can actually see the batch files for activation and virtualization. You can go in and have a look and can actually see what they're doing. You've got the icon file, you've got the token file, which is your license, as such, to run in the Cloudpaging player, and the STC, which is the big bulk of the application. So, we copy that STC file to our network share, we rename it with the GUID underscore file, and we'll leave it there. It's now available to use as a network cache.

So, we just need three registry keys on the client to actually make use of it. We enable the cache, we say we'll use it as a T-drive, is where we want to use it from, and the location of where it is on the network. When you're using it, you do see this, unfortunately, in the Explorer, so we use an extra little register key just to hide that drive to make sure it's not visible to anybody. If you do see that in the Explorer, you can actually navigate down and see the the cache files. So, to test it, quite easy, clear the cache on the clients, we're at 0% used, launch the application, and come back here and it should remain at 0%. And we can confirm this, as well, by looking at log file, multi cache added, and you can see a couple of lines above that it says, "App multi-source T-drive. And we can also see in the registry that it has added it on.

So, right, I'll quickly go through Jamf Pro. If you're not familiar with Jamf Pro, is a MDM solution for Macs and iPads. We use it for Macs specifically. And apart from just change a lot of settings on there, we reuse it to publish applications to Macs. Unfortunately, users have different experience with Macs than they have from PC. They need to get their apps from a different location. There's another portal, as such, for Macs using Jamf Pro. And they may actually still need to use the AppsAnywhere, as well, to get some of the published links that we have there.

So, AppsAnywhere can integrate it, and we can publish everything that we have in Jamf onto AppsAnywhere. So, this is what it looks like in Jamf Pro. This is our admin portal. You see we've published some applications here. We allow them in the self-service portal. In AppsAnywhere, we can link our Jamf Pro server environment here using a connector, and it's simply a case of pointing to the server and adding a service user that has access. So, basically two boxes to get it all linked up. Then we can add a new delivery method. We can just select Jamf Pro policy, and it's the usual Cloudpaging page here. But instead of having your Cloudified apps in the list, it sees automatically your Jamf applications in there because we've already created the connector. And there we go. So, what we had in Jamf we can see directly in AppsAnywhere.

And on the Mac side, as simple. We've got the published app, the second one in there is Adobe Acrobat. That's the published app for Mac. If we hover over that, you can see it says, "Launch as normal," but it says, "Via Jamf policy." And you get the standard splash screen. And if we're not convinced about what it's doing, we can have a look in the log file and see that it's actually just calling the Jamf binary. But if we have a look at the Jamf logs, we can see that it's just been invoked as normal, so it's quite a nice little connection.

So, there was a couple of apps that have lots of entry points, lots of links to start different parts of the application, and we didn't like to clutter up the portal at that point with lots and lots of icons with the same name, more or less. So, I developed a little application launcher, which essentially, just sits in the same directory as the shortcuts that you have, reads the shortcuts, and displays them nicely, and you can launch directly. And finally, we'll want to do some parallels soon so we can get all our PC apps onto Mac. At the moment, we're still restricted with Mac apps on Mac, but if we can get all of our estate onto Mac, or whatever device anyone wants to choose, tablet, then we really will be truly AppsAnywhere, and that's where we're hoping to be very, very soon. And I think that is me.


AppsAnywhere & the University of Dundee

 

Transcript

Yeah. What I'd like to do is tell you about why we made the decision to move to AppsAnywhere. How we actually have implemented it. I'd like to spend a wee bit of time on the challenges that we've had with it as well and finish with what our next steps are. These are the obligatory slides about Dundee and what we are, what we look like. We're a small university by most university standards, but we're very compact. We sit in the center of Dundee so it's got a lot of good things going for us. We're not nearly as spread out as some of the others. You can kind of walk across the campus quite easily. And there's just some numbers to give you an idea of what we are like. The important one I think that stands out is the research income one.

We are a very intensive university. This is reflected in the income that we get, which means that we've got a lot of challenges as well and until we manage two distinct sets of users, you'll all be the same learning and teaching and with research. But research is absolutely massive, So we do have to spend an awful lot more time in the area.

So what were the requirements? Windows 10 was our biggest driver. We were moving to Windows 10 so we were going to take that as an opportunity to fix an awful lot of things that were broken before. We had a real issue with Windows 7 and Microsoft products. They were running extremely slowly in a lot of circumstances. When we went to Microsoft, Microsoft said, "You're using such a wide range of stuff, it's not ours that's broken. It's somebody else's piece of software that's broken so we can't really help you."

So we thought like, "We need to simplify things for Windows 10." We decided then we would do this initially on the student desktop because you've got that natural break in the summer for when you can introduce it and also because our student desktop's much more tightly controlled than the staff desktop was.

The other issue that we heard was around hard drives. And I've had a couple of, you mentioned this already, that the hard drive, the image gets bloated, it gets bigger and bigger, there's more stuff going on and you really just can't fit everything on. And we were in a position where we were having to say, "Oh, if it's going to go here, we'll have to take this bit of an application off to put something else on because nothing fitted." And we were bringing out new machines. We were going for these HP all-in-ones that were 256 SSD drives, 16 Giga RAM, i5 Processors.

So we decided then we need to address this issue about bloatware and how we're going to fit everything on for the students. The rebuild time as you've had earlier as well was taking ages because we had to do all these different things and we decided what we really want is just to clean the image and then deploy the applications to the image. So we were looking to do that as well. So you can see it's all starting to come in. Yeah, AppsAnywhere was ticking the boxes.

Finally, our director had kind of made a promise that we were going to introduce, bring your own device and we were all just thinking, "Well, how's that going to work then?" How are going to manage that?" So we needed something that would kind of go down that path and AppsAnywhere was the natural choice for us to move to that.

Once you decide that you're going to do this, you decided you're going to simplify things, it's going to have every single part of IT and all the different areas that I've listed there are the different sections of IT that come together for this. The end user services team and the service desk, they are part of the same group, they own AppsAnywhere. That's the home of AppsAnywhere. So they do most of the packaging. They do most of the administration. Transition are the technical teams. So they're there really to say, "You have an idea, we're going to make that happen, and then end user services will run it." So there had to be a coming together of all these areas with a common goal, a common set of objectives that was simplify everything, make it quick, and allow the applications to load quickly.

So when we started the Windows 10 project, we were very specific about the users we wanted doing that. We didn't really want friends and family, we wanted everyone that had complained to everyone that had mourned because they were the ones that would give us much better feedback than IT going "Oh yeah, it broke but i fixed it." That was no use to us. We needed that user experience. So we gathered these people together, identified the apps that they had, and we made it real use of Snow.

I don't know how many of you use Snow, but we installed the Snow Client on every single device. So Snow feeds back, tells us everything that we've got. It told us we've got 8,000 to 9,000 applications across our estate. We're never going to manage all of those. We're never going to be able to control all of those. So we decided then that as we moved forward with this, if we identified an application that was used by more than 10 users, we would include it in this. If it was used by less than 10 users, we wouldn't, and we would use another system affect or Defendpoint to do installs with that.

The Last point I've made, there's the annual software review. I don't know how the rest of you work, but the way we used to work was every year in the summer we done an annual software review. We wrote to everybody in every discipline in every school and said, "What software do you need for next year?" And then we would chase them up because there were all on holiday or we would be hunting them down to see, "Do you want the same as you've done last year." And It was absolutely awful. A.

And there was a special team had to be put together to do all this things. So we decided that what we'll do is we'll stop this and we'll follow the Microsoft model that they would introduce me to Windows 10 where we would do small increments. There would be no big bang and each year you would get exactly the same as you had last year unless you told us that you wanted something different. That was the approach that we decided to take with it.

And it wasn't a great because you just don't get results. People just don't get in touch with you. And I'm sure you've got loads of people that at the start of the term, they'll say, "Where's my software?" And we'll say, "You never told us you needed It." And he'll explain, "Yeah, but I need it." So we held the team back to help us with that. But we also thought, "What else can we do to help support this?" And this is where we come on to how we've put everything together.

Microsoft Teams is really the core of how we communicate within the actual IT team. Microsoft Teams is what we used to collaborate with the users when we were first rolling out Windows 10. And with that team, we joined that to the new packaging service that Software2 were offering. And that really helped us. Because that meant when we got these emergency ones that had to be installed from next Thursday, we could start looking at, "Okay, can we put a resource on that and the work that they were doing, can we actually offload that to software2 and get them to do it." Also, we'd bought a bundle of credits to get this done and we discovered when we bought this bundle of credits that gave us access to lots and lots of other software that we didn't have. And we thought, "Well if we're getting it for free, we'll just put it on the desktops." That was a bad decision because once we put it on the desktops, we fund we had with nobody to test it because no one had asked for it.

So packagers are tasked with making sure things are UATed so they will then say, "I'm not doing this because..." Or actually it was in my name. So they were coming to me and saying, "Will you UAT this?" "I don't know what that thing is. No, I can't do that." So we're all back a wee bit from that.

But as I said, we'd set this threshold for students. We set it at five users, so anything over five users that students used, we packaged up, we put it through here. We've got a kind of choice of an order that we put things through. AppsAnywhere is our first choice. SCCM is our second choice, Defendpoint is our third choice and our final choice is manual. There's no getting away from it. We have to do manual installs every now and again. Something comes up. We want too find a way that we're doing that we can avoid that.

So once we've done that, we start putting stuff through AppsAnywhere. Our resources though by now are really stretched because they're having to put up new machines. It's the same team that are actually building all of this. We've got stuff coming in from AppsAnywhere starts helping with that demand.

But at the same time the university decided they would have a new network installed as well. So there was a new network came in, there was new firewalls came in. The new firewalls really messed with our license servers as [inaudible 00:08:34] will tell you. We had a lot of issues with that because applications just stopped working. They couldn't reach the license service. So we've had a few challenges that IT created themselves. But I think that's a common thing as well.

The main thing though that came out of it was that we discovered that we had to start pinning responsibilities on different people or different areas, different roles. So then we started saying, "If you ask for a piece of software, you are now that software sponsor, you've got to give us the license information, you've got to tell us who will UAT it and any new versions that will be up to you when that version gets installed." So once your name's in that sponsor box, you can't get away from it unless you pass it to someone else. Without that, we wouldn't package an application. So if someone comes to us and says, "Will you make this available for my students?" We'll say, "You're the sponsor." "No, I'm not." "Then we're not doing that. We need to know who's going to do that piece."

And that's kind of fed through. Now we've got Senate learning and teaching. We went to them, they endorsed this, so now we're able to say, "It's not us, learning and teaching are saying that you have to do this sort of thing." That's been really helpful. This year one of the biggest helps that we had to forfeit was for the Adobe apps. If you're familiar with Adobe, you'll know they changed the licensing model and Adobe's been an absolute... It's been a difficult one this year and also we didn't find out how they were going to run the new licensing until around August of September.

This really put us in a hole because that's a busy time for everybody that's involved in this. So what we've done is we went to AppsAnywhere and AppsAnywhere were able to provide all the applications, all the Adobe applications. So we've got them in UAT just now with the new device license model. All the UAT that we're doing with them is running quite well. Adobe's a strange one. Where UAT actually rather than it just be a single sponsor, there's lots of different areas in our university use it, so we've got lots of people that want to know that it's working properly for them.

So that's taken a little bit longer. I don't know if some of you were affected by the Adobe license issue that happened just in November there. Anyone who had licensed it 2017 the license has run out. So from today or yesterday there's been a flood of calls coming in. People saying, "I have not got this." So what we're hoping to do is make AppsAnywhere available to them. They [inaudible 00:10:51] have the new versions and give them some limited help with that.

An important part of this was that we wanted this to be consumer grade, so we didn't want having to give anybody lots and lots of help with how they're going to run these things. This is the only instruction we give our students for it. We said this is how you use it and they were able to use it, it worked. Now, when terms started, we thought, "We've done well. We've done our UATs. We've given this to students, they know how to get it. They fired up and some applications run like an absolute dog and we were flooded with calls from people saying, "This is taken 15 minutes to load. I've got a class to run and what's going on here?"

Then we started looking at that and thinking, okay, did we go in the classes first and fire everything up and get it running and that. So we've had to really look at this quite closely and adjust the way that these applications run, make use of the technologies that we've got. We've also decided with some of them that AppsAnywhere probably isn't the best way to deliver some of the applications. We'll still present them through the AppsAnywhere portal, but we'll take a different route with them. We've got engineering applications that are over a hundred gig and then have serious plugins that really have to sort of meld with the system and there was no way we could get this to run properly.

So I've actually got someone working on that right now as a different way of doing this because that has given us quite a bit of grief. One thing that we found after we went through it, was I was able to get this and like numbers. So we were able to see that the students were actually using the thing, and I thought this was particularly interesting. The heaviest usage we had was on Valentine's day, 6,680. I'm not sure what that says about our students, but that was the day we were most on it.

Because we've had it running for the wee while now, we're able to make comparisons between previous years, previous days even to see what the combinations look like. We can drill down into when applications were used to see if the classes that were expecting things to happen can happen. We found that four applications that we've been running for over a year have never ever been used. So why were they requested?

So we'll be able to feed that back though. We've got 50 applications that in total have run for less than five minutes. So we could look at them and we can feed that back and say, "Are you really sure this is the correct choice of applications that you have?" We've got 453 applications that are coming through AppsAnywhere, so it's quite a hefty number. We do though, encourage everyone to use the search box. We still go along to see people and they're scrolling down and down and down. [Inaudible 00:13:38] hundreds here, just hit that search box.

So move on now to our next steps. As I said, we want to fine tune the delivery methods. We're really looking at how we're going to use SSCM more. We restructured SCCM and we rolled out Windows 10. We've now found that we didn't do that the way that suits the delivery of applications. So we're no going through a second phase of rejigging SCCM. So we've be able to deliver more applications that way.

We are relying more on Software2's, packaging service to give us the demand elasticity that we need to be able to cope with things. So we're now able to do things a bit quicker because we can offset it and we can concentrate on... Which we're splitting packaging out and to ones that we can hand off that we know that that can be dealt with and the ones that really need that local knowledge and experience and we want to retain that experience. So we make these decisions.

Actually it's the application packages. When they get the package, they make the decision about the delivery method, how easy, how difficult it would be. Then we look at the workload and then we decided we're doing it internally or we're going to hand it off. We've got a nonstandard requests, just now, for piloting AppsAnywhere in parrallel. We think this is going to be the next stage. What I was really hoping that we'd be able to see what we're going to do with it, but the person that's dealing with this is involved in industrial action, just now, so I don't have any answer.

I think they're back next week, so maybe we'll find out then. I want to make better use to analytics. So I'm really pleased that we're going to see a bit more of it. And finally I shared an office with the director and every now and again, he says, "When are you going to switch on? Bring your own device? So I have to keep saying, "It's not just a simple thing of switching on, and then there's lots of other things that we have to take in account." So we'll have to keep referring back to that though we're not cast to be the target that we're aiming for. And that's me. Okay.

Thank you.


Software2 Packaging Service

 

Transcript

As most of you know, at the beginning of this year we introduced our new look packaging service and we had kind of two clear goals. First of all, we wanted to reduce the repetition across all of the different universities and colleges all packaging the same applications. The other thing we wanted to do was to package applications more efficiently and then we could pass on the savings essentially, to you.

So, the first thing that we did is we introduced our core apps. We actually, I think, introduced the world's first self-service packaging app store. This was the first time I think that anyone's done this kind of thing. So, we had 30 free and open source applications ready to download and that we're keeping up-to-date and we just call this our core apps.

And then we introduced a menu of over 60 applications. Essentially, we created an on-demand service so that you could log into the self-service portal and you can request the applications that you need. You responded. So, over 50 universities have signed up to use this service already which we think is fantastic. So, as Matt said, Dundee were one of the biggest users of this service this year, but we just really, really pleased that so many of you responded to this. We created a banded kind of price system. So bands A, B and C, A being the simplest apps, B a kind of mid-range and then C the more complex. But yeah, over 50 customers, 50 universities using this now, which is really cool.

So, what does that mean? Well, as you know, it gets pretty busy in the summer. This was Matt Rogers at the start of the summer this year taking on all of your packaging service requests. Not actually Matt Rogers although it does look a little bit like him so put that disclaimer in. And then of course it got busier because when you all get into July and August then it gets really busy and everyone's trying to get their applications done for the start time and this was what we looked like in the office at that point.

So, we kind of understand the situation. It's what Matt talked about. You have these periods of the year where it just gets really busy and you've got to get everything ready and we want to be there to support you with that. I like Matt's phrase, "Demand elasticity." We're there to help you. You don't have to put everything through the packaging service, but if you need us, we're here and we can help and we can do some of that work for you. I'm pleased to say that our team delivered every single one of those packaging requests in time for the start of term. Everything that was asked for was delivered on time. Okay. So, pretty happy with that. And it was busy, but we did them all so it was good.

So, what's new for 2020? Because so many of you are using this service now we can go further. This is what we're going to do. We're going to add another 20 core apps. We're going to take it from 30 to 50 that are included. Hopefully, some popular titles in there that people are using and will benefit from. Obviously the one that we've been asked for the most PacMan. So, we're going to take it from 30 core apps to 50 core apps, hopefully give you some more value. You're going to have a higher hit rate. It's more likely that you're going to be using some of those 50 applications.

Some of these things they're pretty simple applications, but it just doesn't make sense for everybody to package the same apps. We can do this for you so we want to make that easier. We do give you all the parts of the application, the recipe so if you want to take something that we've done and then modify it and change it, add your own plug-ins, whatever you want to do, you can do that. But hopefully we've saved you that little bit of time.

And the other thing that we're doing is we're working really hard behind the scenes. Umais, who's here today he's been working really hard to try to automate the packaging of some of these apps. And what that means is that we can update them on a more frequent basis. So, really we want to be doing these at least every three months so you've always got an up-to-date version that you can just draw down. And it's really because of that that we can do 50 core apps now plus 10 of your own free open source apps that you choose. And we're doing it at the same price as last year so no price change.

So then the bands. We're also going to add more value to the kind of packaging menu as well. So, we're still sticking with the three bands, A, B and C. There actually were only two apps in Band C this past year so it's not like everything is the most expensive. There's just two in there. But as we're getting better at this, we're getting more efficient. We're taking more requests from you. We're seeing more of the same applications being requested. That means that we can do things quicker in the Software2 office and so we're going to reduce the price of some of the apps. We're going to move Visual Studio, Nvivo and Arduino down to Band A which is the cheapest band. Okay? So, if we didn't help you with those this year, then hopefully we can help you with them next year.

We're going to add another nine apps to the menu as well so we're going to give you a kind of fixed price for another nine apps. Our goal is to try and build up this menu to add more value so a lot of these apps that you're all using, we can package them with Software2. We can do it faster and we can try and pass on these kinds of cost savings, these efficiency savings to all of you. So, that's the whole goal of what we're trying to do with packaging services.

Mike talked a bit about Adobe CC, so we offer this as a kind of bundle. It costs 3000 credits and we include 18 apps so I think that's pretty much everything in the suite is in there and we update those twice a year so that you've always got a latest version of those. It's a kind of significant saving over what it would cost to do these individually.

This year we've been doing the Named User Licensing because of the way that that works. Early in the year, we recognized that we could do that more efficiently and the users essentially just log in with their Adobe ID so we did that first. We have now cracked the Shared Device Licensing so for next year we will do Shared Device Licensing at the same cost. You can either have Named User or you can have Shared Device, whichever one you've gone for. We'll do the whole suite for 3000 credits essentially. If you want them both, if you want the whole set of apps with both Named User and Shared Device, we'll do buy one get one half price. We'll do for 4500 credits we'll do you both versions.

We're going to add a new bundle for 2020. See if you can guess what this one is. It's AutoDesk. There's a lot of apps in the AutoDesk suite. We've picked what we think are the most popular 12 apps that people are using, the ones that we were getting asked for over the past year, and we're going to do the same thing. We'll do the whole lot for you for 3000 credits and we'll keep them up-to- date. So again, rather than go through all that kind of time of packaging AutoDesk, we can do that for you. Again, we give you all the parts of the apps and the recipes so if you want to take what we've done and add some different plug-ins or to make changes, you can do that as well. But we just want to try and save you the time on these kinds of apps that really everybody is using.

Okay, so this is something that you want to do for the coming year. Essentially you choose a subscription. We've got the core subscription now includes 50 core packages and 10 of your own requests. Or, you can go for the plus subscription which is everything from core, but you get 6000 credits included. And depending on what you choose, you're going to get up to about 30 apps from the menu. Or, you might decide to swap those credits and take Adobe or AutoDesk, however you want to do it. And if you want to buy more credits, obviously you can buy some more credits, but that's essentially how it works. So yeah, you can just add, if you want to use the credits for a bundle, you can just add a bundle. Essentially now if you take a plus subscription and you choose wisely from the menu, you can actually get over 100 apps packaged and ready to deploy from the plus subscription. Okay?

So, this is what we're trying to do. We want to try and make it easy. We want to kind of pass on the time savings that we're making with this. That's essentially what the packaging service is all about. I think this kind of seems like an odd thing for me to say, but we didn't create the packaging service to make money. This is not a kind of part of the business that we see as something where we want to make a load of profit. We see this as support in this community. Everybody's packaging the same applications and we want to kind of make this easier.

So, even if you can only take the core service for the coming year, I would love as many of you as possible to do that, then we can continue to add value to what we're doing with this. I think we will continue to add more apps and we will continue to try and pass on the efficiency savings as we go along.

If you do want to package your own ups, obviously we have the Software 2 library of recipes and we keep adding to those all the time and we're feeding in all of the recipes that we're creating from the packaging service get shared onto the forum as well. So, there's around 450 recipes on there now. I think that has probably doubled just in the last 18 months. I seem to remember standing here probably before Umais joined us, we've maybe had just over 100 recipes on there. We're up to 450 now and we've tried to keep those up-to-date. If you want to do our kind of training and become a certified packaging associate, we can do that too.

We've done a lot of classes this year. I think we've done at Essex University and Lancaster to date in UK and Dundee, I think, this year. And then I think six or seven of our customers in Germany, I think this year have done training courses as well. We now have about 200 certified packaging associates, which I think is really cool. There wasn't a certification for packaging in the past, we've created one. There's 200 professionals doing packaging, which is great. If you've not done that yet and you want to do it, generally you need probably 6-12 months experience using Cloudpaging Studio. We go into a lot more of the advanced features. We look at all the latest tips and tricks and techniques that our team are working on.

So, it does get quite busy this. We get booked up quite quickly because typically everybody wants to do this in the kind of holiday periods when the students are away and it's a bit quieter. So, if you'd like to do this in the new year, it's probably a good idea to book it early if you want to try and get into January or even at the kind of Easter time. If you want to do training then, let us know.

And then next year we're going to go a step further. We're going to create a Certified Packaging Expert course and this will be a real kind of virtualization deep dive. I like to say we're going all the way down the rabbit hole. This is just kind of simple packaging, but looking at some of the really complicated applications [inaudible 00:10:19] got services, drivers, Java. The ones that Umais loves to see in the office. All of Umais' favorite apps will be covered in the CPE costs. And also looking at kind of in-depth troubleshooting about how if you do have issues, how to deal, how to solve those issues. We're hoping to have that available early next year as well. We'll post more information on the forum in due course. All right. Brilliant. Thank you very much.


AppsAnywhere Analytics live demo

 

Transcript

So, hello, everyone. My name's Keegan. As Phil said, I'm the Account Manager for the U.K. over at Software2. What I'm going to do today actually, as Phil said, I'm going to give you guys a little run through of AppsAnywhere Analytics. I'm going to show you some of the dashboards that we've created and actually go into the explorer function as well and do a little live demo. Fingers crossed everything goes to plan!

So just to give you a bit of background actually, we released Apps Analytics a few months ago. For us, it really feels like the missing puzzle piece that is AppsAnywhere. That's because over the last, say, three to four years, we've integrated loads of different delivery methods to ensure that you guys are carrying on delivering applications to your users in the most effective way.

But with the old reporting module, we're only really able to report on Cloudpaging data. So there's lots of data that had the potential to be captured that we wasn't reporting on. So now with Analytics, we're reporting on every single delivery method that you guys have set up in AppsAnywhere. What that means is, not only will you have a lot more data to work with, but that means we can start using that data to tailor an IT experience for your students, start improving that student experience as well. We think along the way we can save you a little bit of money in doing that as well.

As I said, there's a few dashboards already created here that are ready to go straight out of the box of Analytics. We've created these dashboards based on data we think that you guys will want to report on.

So this is just a landing dashboard. So this is the first dashboard that you guys will see when you log on. So straight away we can see some stats at the top here, so the number of active users, and how many times those users are logging in the past 30 days. So some really useful information straight out of the bat. We can see that our users are logging on to AppsAnywhere multiple times. More interestingly still, we can see that they're logging on on multiple devices as well.

What we've got here, if we go down a little further, is our top 10 resources for the month. So these are our top 10 applications. Over here we've got a really interesting pie chart. The reason why this is interesting is we're looking at active devices by operating systems. So we're actually breaking down who's logging into AppsAnywhere per operating system. So this allows us to start asking the question, are we actually delivering applications the way that students are trying to access them?

So we can see here that a large proportion of our users are trying to log in with a Mac device. With this information, we can see that we should probably be utilizing maybe Parallels RAS or [JAM 00:02:50], so we can cater for these users. So this goes back to my point and saying that we can use some information to start tailoring our IT services to make sure that we're delivering applications the way our students are trying to access them, basically.

If we scroll down a little further, here we've just got a heat map based on where our users are launching AppsAnywhere around the world. So this is particularly relevant in today's climate where universities and institutions have campuses all over the place. So we can just see here where students are launching applications. So that's basically the first dashboard I'm just going to show you.

The next one I'm going to show you is just the monthly management summary. So for this one here we're looking at our most popular resources for the month of November. So we can see again we've got the number of active users, so we've just seen that, but this time we've got how many times those users are actually launching applications.

Once again, we can see our top 10 most popular resources but this time we drill it down a little bit further and we're actually asking the question, how are those most popular resources being accessed? So we can see here whether those resources are being accessed on student-owned devices or whether those devices are actually organization-owned.

If we scroll down a bit further here, this time we're looking at the number of launches and active users, but this time we're looking at it over the period of November. So we should be able to see some peak times in launches and some dips in launches as well. So we should be able to identify when the weekends are or when the best time would be to schedule any sort of maintenance, any sort of upgrades, or any sort of downtime required. So we can actually start using real data, real accurate data, to schedule any sort of maintenance that we may be planning as well.

So just the final dashboard I'm going to show you. So this time we're looking at, for the month of November again, same stats we've just seen, so number of active users, number of launches, and number of devices, but this time we're actually drilling down and we're going to ask what we mean by a device. So we can see here we've got the device OS breakdown again. So this time we can see whether those devices are user or organization-owned or whether it's on or off the domain.

So we're starting to build a real, clear picture of our whole computing environment. So we can actually start asking some really important questions of how students are accessing AppsAnywhere, on what device they're accessing AppsAnywhere from, where they're accessing AppsAnywhere from, and who owns that the device they're accessing it on as well. All really important questions.

We can save these dashboards. So all the dashboards you can save as a PDF or a CSV file. We can also schedule them to be sent over to license managers or IT directors, whatever that may be. We've got a delivery schedule here. I wouldn't personally be sending this to your IT director every day at 6:00 A.M. but we can set that up at specific months, or whether that be at the start of term or the end of term, whatever that may be, we can set that up and we don't have to worry about it again.

So that's essentially some of the dashboards that we've created, but I know everyone in this room will want to get behind the hood and see how we actually manipulated some data and looked at the data and how we use that data to actually create some of these dashboards and that's where the explorer function comes in.

So these are all different categories of data that we can use and look at. The first one we're going to look at is daily metered concurrency summaries. As catchy as that name may or may not be, this, essentially, is looking at concurrency per application. So this is probably our most popular data point we've been asked to report on, plain and simply because it's the most obvious way we've identified we can try and save some money, in terms of licensing.

So if I do this correctly, which is a big if, but if I do this correctly, we should be able to see all our applications and the peak concurrency for each of those applications. So we can see which applications are over-licensed, or which applications may be under-licensed as well.

On the left hand side, these are all different categories that we can... all different dimensions and measures that we can pull data from. Just run that as peak concurrency. If I run that query, we've just got a list here of our applications. So this is just a demo site so there's only two applications in it. Your environment will look a lot more interesting than this. As you can see here, we've got a peak concurrency for MATLAB and [NAT JIS 00:07:44], and then we can add these as visualizations fairly simply. Then we can also save this. So we can save this as a look which means that just I will see it, or we can save it as a dashboard so the whole team can see it as well. So that's just the first example.

The next example's a little bit more complex. I mean, can't get much simpler, but the next example's a lot more complex, and this time we're going to look at launches. So in looking at launches, what we're going to do is, we're going to go beyond licensing, we're going to go beyond usage, and we're going to actually delve into your physical environment. So we're going to actually look at how your physical machines, your lab computers are being used. So we should be able to see peak concurrency for your physical machines, so we can see which machines aren't being used to their full potential.

Just go into device info and device name. Launch information. Number of distinct users. So this time, on this measure... I mean, on this dimension, I'm going to add a filter, and I'm going to add it as contain. So this is under the assumption that your physical machine starts with the prefix of LAB. When I run that, we get a list of all my physical machines and how many times those physical machines have been used. So this is giving me the information to go out there and investigate why some machines may not be used. They might have slow performance, they might not be positioned in a great place, but we're giving you the data there to go out and do those investigations.

What we can even do as well, we can go a step further and filter per lab. So if we look at, for example, just LAB A, we should be able to look at a specific lab. So there we are. Right, okay. So if you want to click off any dimensions and filters, you essentially just click it off on the left-hand side. It's as simple as that.

For this example we're going to go past physical machines and we're going to look at delivery methods. We're going to look at each delivery method for all of our applications, and we're going to actually look at how those delivery methods are being used or not used, whatever the case may be.

So [inaudible 00:10:19] basic info again and we'll just go to [inaudible 00:10:23] name. This time we're going to go to delivery method descriptor and launch information. Number of distinct users. Run that. What we've got here, we've got a list of all our applications, and we've got every delivery method for that application and the usage of that delivery method as well. So it's some really useful information again.

Once again we can filter on this. So we can look at specific applications if we like. So if I just want to look at MATLAB, and I run that, that gives us the information just for MATLAB as well. So really useful information that we can use in our everyday work.

As you can see, everybody, there's so many different measures and dimensions that we can report on. I've barely scratched the surface of it. I'm sure that everybody in this room will find some really awesome use cases for this. We'd love it, once you guys get your hands on it, if you'd share any use cases you guys have found for it in particular because we'd love to be able to create a dashboard and share it with the wider community, everybody can benefit.

Thank you very much, everyone.


AppsAnywhere's year in review and visions for 2020

 

Transcript

It's always a pleasure to be here on days like this and to see you all again. As you may have noticed, and as Marcus mentioned, there's been a lot of changes at Software2 this year, one of which is I had the great privilege of being made CTO. I don't know whether you may or may not know, but I've been here with Software2 since day one. This is my 10th anniversary as well. I was employee number three officially. Up until now, I've been very much on the cold face of Software2, managing the development team day to day. So, I see my transition as a really incredible opportunity to help drive our product forward even faster and really focus on the changes that bring the most value to all of you guys in the room today.

So, this is going to be a little bit different to my usual development update. I want to talk more about my priorities as CTO first. And then, I'll get into the nitty gritty after that. Okay. So, our number one priority, as Marcus mentioned, was investment in development. So, one of the big challenges when developing a world leading product like AppsAnywhere with such a small team, I will never downplay the achievements that we've managed over the last couple of years. But there's definitely so much more we could have done with more resources available. So, deeper integrations. All the integrations that we've done, we'd love to go more into those. Lots of new opportunities that we could take advantage of. And generally just adding more polish to everything we do. And also, as a team, working at max capacity all the time and still struggling to meet the deadlines that we set for ourselves has always been a struggle. So, yeah. This is a really good, exciting opportunity to for us.

Again, as Marcus mentioned, this was an immediate start. So, we've doubled the size of the development team within two months this year. So, it's been quite an intense last half of the year. But that growth always continues into 2020 as well. So, we've got another dev team member due to come onboard after the current onboarding process. And then, we've got some summer internships on offer for next year as well. So, if you know anyone that's interested, let us know.

To start us off, I'd like to introduce Lewis. He's joined us as our new user interface lead developer. Lewis has actually worked with Software2 on a contractual basis since our earliest projects ten years ago. He was solely responsible for developing the user interface for AppsAnywhere and Software2 Hub as well, for any of those that remember that one. He is a great friend of mine. We actually went to university together. And he's incredibly good at what he does, as you'll see later in this presentation.

We've also got Raihaan and Adam that have joined us. These are two incredibly bright and enthusiastic young graduates. They're working under very close supervision of Emily on the backend development team. They're currently upskilling their knowledge of our technologies, processes, and the product itself. And they'll start contributing early next year to the product. They'll be starting with some automated test packs to make sure all our releases are doing exactly what they're supposed to do. And then, obviously, they're going to make a huge difference in the speed that we can get out new features in 2020.

So, my second priority for the latter half of this year was resolving defects. So, over the last 12 months, we've built up a backlog of unresolved defects. Our focus has been very much on really big new features and the new analytics project, which has taken up a massive amount of our time. And those are obviously great for you guys, but I don't think that's what you see as the most important thing. Most likely the most important thing for you guys is actually the things that are causing you problems every day, the things that mean you take three steps instead of one, or potentially things that are stopping you using parts of our product for what are often really, really minor reasons. So, defects are often the easiest thing to fix. And that's reflected in the content of our next release, which I'll talk about in a minute. And we've got a strong focus on resolving these little issues that keep getting overlooked. And we've also got an ongoing commitment now to maintain and keeping on top of those as we go forward.

And then, the final priority, as Marcus discussed, was setting a clear direction for the product. Up until now, we've been very much focused on the next feature, but we've not necessarily been looking on where we want AppsAnywhere to be in like three to five years, what we want the product to do. So, we've still developed great features, but we haven't been looking at the big goals and the big projects. And it's hard to get the revolutionary changes out when you're just focus in on a month to month. So, we sat down as a management team and we agreed some really high level goals.

So, the first one, making the product easier to deploy. So, a lot of you guys in the room will be past your initial deployment, but you still need to keep things up to date. You still need to have updates every couple of months. And we need to make this as easy for you guys as possible. And again, I'll talk about that in a little bit. This includes making the product easier to support. So, putting in more login, more testing and debugging tools. SAML SSO is always a great example of something that a lot of our customers struggle to get the configuration right. And the product could be a lot more intuitive in helping you debug problems with things like that. And the product plays a big part in this.

So, the second priority is providing more value to you as a customer from AppsAnywhere. So again, this is partially about getting you up and running as quickly as possible. But, even if you've already rolled out, there's a lot of features that we're talking about in this area that I think will really benefit you. So, ideas such as important resources from all your connected environments, like Cloudpaging, VMware, Parallels, detecting locally installed applications and just import in them into the product for you so you don't have to go in and set all those up. Things like integrating with packaging services, so getting set up with an initial subset of apps is just a case of connecting to the services that you've already bought. And then, just bringing the value straight into AppsAnywhere from the start. Basically, anything that makes your lives easier is also a good bonus for us.

And then, finally, again as we've mentioned, maintaining our customer focus. So, we remain committed to being focused on higher education. You guys are our champions. You are the ones that are using the products and AppsAnywhere every single day. We want to ensure that we continue to provide everything you need to succeed in that sense. So, we're committed to helping you stay on top of the latest challenges, the latest processes, following the latest technology trends. And we want to provide you guys with the best experience of our products and services.

Okay. So, now the interesting stuff. So, AppsAnywhere 2.8. So, as promised for the last couple of months, our focus has been on defects. So, we've investigated, replicated, and resolved over half of all the defects that have been raised against AppsAnywhere since the beginning of time. What was that, 2016 now? If you want the Apple version, because our marketing team put these slides together, that's the most defects resolved in a single release of AppsAnywhere ever.

Okay. So, the sort of things that includes, further accessibility improvements, some major performance improvements that we've come to with the help of the guys at Durham. So, those guys have what, five, 600 apps in AppsAnywhere. Of course, that works fine in Chrome and Firefox, but they were noticing some performance issues in IE and Edge. So, they were really helpful in helping us get those resolved. And those are all in 2.8 now. Support for more than two automatic dependencies at a time. A lot of updates around the mass action session management. Support for domain aliases in SAML, which was causing problems for a lot of our customers. So, if your SAML response is returning a domain that's different to the one that's configured in [inaudible 00:08:12], you can now set that up and it will behave properly.

Some improvements around session expiring in Cloudpaging, a lot of people getting two and five minute warnings from Cloudpaging when they shouldn't have been. And the unknown user timeout issue, which I know has affected quite a few of you. So, yeah. I think it's 35 defects that we've resolved. And that's all the blocker and critical level defects. And there's only five high priority defects left in the system now. So, it's a really, really cool release that I'm glad to be able to get out. And I hope you can all get your hands on it.

What a Christmas present that is. And so, we wanted to make sure that it was available for the Christmas upgrade window. So, it's just in the final processes of internal testing now. And we'll be announcing that probably in the next week, I would imagine. So, yeah. Like I said, we want you all on this release. We want you all up to date with the latest version to make sure you're getting the best experience.

If you haven't yet upgraded to 2.7, I'll just quickly recap. A big part of 2.7 was the analytics feature. So, we're tracking all the users, the devices, all the sessions, and all the launches that happen in AppsAnywhere, storing that information in an analytics database, and providing it to you in a way that you can browse the information as much as you want, build your own reports, share those with management, and just see that data in any format you like. And we provide a bunch of built in dashboards and reports that we think are useful for you as well to get you off the ground. And we now have some training to help you get up to speed with analytics and get the most out of it. So, if you're not in 2.7, you're missing out on that.

We also implemented an app lists feature. So, users of the system can share an app with their friends or colleagues. But more importantly, academics or course administrators can go in and create a list of apps that are relevant to the course that they're teaching and they can share that list with the students. You're just using a simple link. So, that's a really cool feature. And I'll talk about a little bit more how we're expanding that in a minute as well.

And, of course, the latest OS support. It's the most trivial part of any release, but it is a massive thing for you guys. If you're still before 2.7, then all the latest releases for MAC OS, iOS you're not going to be supporting those. So, yeah. The sooner you can upgrade the better.

So, given those high level goals that I mentioned earlier, I'm going to talk a little bit about how that translates into what is next. So, the first thing is our improved upgrade process. This includes installs as well, but obviously the main part of this for you is the upgrades. So, upgrades currently take around four hours per server for us. It's completely out of your hands. You completely rely on us to complete those upgrades. We want to hand over some control for that process. So, it doesn't matter what we develop. Obviously, if you guys are able to upgrade, whether that's through process or it's too technically difficult, then you're not getting the benefits of what we're putting out there.

So, this is a little snapshot of our current status of all the customers. So, we've got about a third there on 2.7, but the vast majority of our customers are on 2.6 or older. If you're in the 2.4 and older category on that chart, then you're currently missing out on 38 new features in the product, 51 defect resolutions, and analytics, and possibly more if you're on 2.3 still. So, yeah. We really want to get you up to date with the latest versions of our products.

So, in order to improve that process, we need virtual appliances. So, I mentioned in June that we started with analytics on a virtual appliance. I mentioned that this will be coming up to AppsAnywhere. And a couple of you guys asked me about it last night as well. So, I'm happy to announce that it now exists in virtual appliance form. We are testing it internally at the minute. And the main benefits of virtual appliance, it all revolves around the fact that we can automate the build of a server. Okay?

So, that means that everything is guaranteed to be installed and configured, not just correctly, but the same for everybody that deploys the product. So, that means it takes much less time to get it running on the site. It's just a case of putting it in your environment and doing the last bits of setup. And most importantly for us is we can test those builds of the appliance internally. And what we're running all our tests against is what's running in your production environment. So, it's much easier to guarantee a working product on all the customer sites. And there's a much reduced chance of environment specific issues going on when we've got AppsAnywhere running in production.

It's easier to upgrade as an appliance because you can just swap it out for the latest version. Or, if you don't want to do that, it's an appliance so we can script most of the upgrade steps to make that process a lot simpler. The performance of it is... I don't want to say amazing because I haven't got any official figures for you yet. We're still working on it, but it's pretty impressive. The first time we've got this running on appliance, it was noticeable how responsive AppsAnywhere it was. And that's mainly due to the fact that it's now going to be running on Linux. So, Linux is much better at handling the massive number of files that are involved in AppsAnywhere. There's like 150,000 files in the AppsAnywhere project. And you can't even zip for that up on Windows. It just crashes the whole machine. So, Windows trying to serve all those files is just a hit to performance.

AppsAnywhere has actually been running on Linux in our development environment since it was first created. All our development is done on Linux. So, we're even happier with how the product works on Linux than we are on Windows. And, like I say, it just works. It's more stable. It's more reliable.

It also means a move away from Xen Server. So, Xen Server we initially started using to help us with the deployment. So, it helps us install all the components as a single installer and it helps us with deploying AppsAnywhere to those servers. In a virtual appliance world, those issues are sort of moot. We don't get the same benefits. Xen Server is also the number one reason for upgrades failing amongst our customers at the minute. It has a habit of entirely trashing servers just as and when it sees fit. Sometimes you can try and run a Xen Server upgrade and that machine is just irreparable. You have to revert to the snapshot and start again. So, we're removing this from the component stack, in order to get those performance benefits.

Xen Server also really restricts us in terms of upgrading components for security reasons. If there's a new version of Apache or open SSL that you have to have to pass a security report, we can't provide you those updates until Xen come out with a new release. So, again, removing that middleman in between us and the components will allow us to be a lot more responsive, from a security point of view as well.

So, yep. Release, we're saying early 2020. It will definitely be available for the summer upgrade window. We plan to be selecting a couple of customers for the Easter upgrade window to take this as an early access and make sure it's exactly what it needs to be. Obviously, this is an entirely new way of doing this in production. So, we'll be making sure that this is production ready before we're putting a firm release date on when it's available. But, like I say, it already exists. It's already running. So, January, February next year, this is going to be production ready.

Okay. I've already hinted at customer managed upgrades, giving you the control to run upgrades yourself. So, once we can easily deploy an appliance into your environment in a managed way, in a scripted way, we can start handing over the control of that upgrade process to you guys. So, we understand the current barriers to upgrade AppsAnywhere as a critical service. Downtime is incredibly hard to justify. There's huge risks of anything going wrong. And those risks are hard to mitigate, hard to accept. And, like I said, they have a high impact on your users. And, as I mentioned before, at the moment you don't have a lot of insight into that process. You're completely relying on us to do that, although we obviously do it perfect most of the time.

But, yeah. We want to give you control of that process. Take back control, that's the mantra at the minute. So, we want you to be able to upgrade when it suits you. We want you to be able to upgrade more often, as I've said. We want you to have more confidence in the whole process. And we want to come up with a process that means minimal downtime for your users. So, this will be our next focus area, once this appliance is ready.

We do have a couple of options. There's a split in the road ahead of us and we're very much interested to hear your feedback on these options over the rest of the day. So, there's two ways to upgrade an appliance, as I've hinted at. You can import a whole new set of appliances and then just swap them over. Or you can upgrade all the services, we can build in some sort of scripted ware that you can upgrade the service in place. And there are pros and cons to each.

So, if you want to swap the new appliances, number one on the left there, this is the safest method because you create a new environment, you can verify that environment, and your old environment still exists. So, the downtime is minimal. It's just the amount of time it takes you to run the Upgrade Wizard, you switch to the new environment, and, if everything's happy, you can delete those old servers and use the new environment. If you're not happy, then you just switch back. And it's a quick process.

The cons to that are you need to import new servers into your environment. There may be red tape involved in putting new servers in. It is extra resources you have to account for. Even if it's just for a limited time during the two or three weeks you managing the upgrade process, you probably still have to go ask someone and beg for some extra memory from the hypervisor. And, overall, it's a longer process.

If you go with option two on the right, then we don't need extra servers. So, it's the simplest process. You just take your system down, you log on to each system, you run the upgrade scripts, and everything updates, and then you bring your system back up. The problem with that is it is more downtime. So, obviously the system is down for the amount of time it takes you to upgrade all the servers and verify everything is working. There's more room for error, like I said, because you can't verify that your new system is working before you switch. And it's relying on you guys to take service snapshots and have those available if anything goes wrong so you can move it back to the environment yourself and go back. So, like I said, those are our two options. I'm not stating my preference. We're just interested to hear your feedback on how you think you would see that working in your environment.

So, I do have a couple of questions. I don't want to get too interactive here, but if we could do a couple of quick show of hands. Please, if you could raise your hand if you would need to ask someone else to take a snapshot of your servers. Okay. So, that was about 20%, maybe. Okay. Raise your hands if you need someone else to take a snapshot of the database. A few more. I'm surprised at that. Okay, cool. Raise your hands if you need to involve someone else to remove servers from a load balancer or add servers to a load balancer. Yep. Cool. This is going to go the same way for all of them, isn't it? Yeah. Okay. Creating new servers, even temporarily. Yep. That's most of you. So, who prefers the idea of creating new servers and swapping them out? 50%. Who prefers the idea of not creating new servers and just upgrading the ones you've got?

[inaudible 00:20:11].

Yep. 50/50. Perfect. That's that decision made. We'll keep working on that over time. Okay. So, that's that track of development. And we're also working on a great project at the minute with the University of Bergen. We're working on expanding the app list feature that I mentioned earlier. So, we're implementing that into the Canvas environment. So, University of Bergen and I'm sure a lot of our customer sites, the first point of contact for a student will be a virtual learning environment or learning management system. So, we want the users to be able to go in and get all their resources for their course and be able to see the list of applications that relate to that course as well.

Yep. Great project with the Bergen guys. That's underway now. So, that'll be available early next year. And that involves us becoming an LTI provider as well. So, at the moment this project is Canvas. But, in theory, we can expand that to any learning environment that supports the LTI as well in the future. So, that's a really exciting project.

And I said I would very briefly mentioned Windows Virtual Desktop as well. So, we're keeping an eye on the Windows Virtual Desktop project. We've done some internal proof of concepts software too. And we're talking to a couple of our customers at how they might be implementing this. We've got a proof of concept scoped out for how we would integrate with Windows Virtual Desktop. And I look forward to hopefully being able to show you that at the next user day. But, if you have strategic projects with Windows Virtual Desktop, we'd be very interested to hear how you plan to use that and how you see it working with you with the delivery technologies as well.

So, Marcus promised that I would talk about the future of AppsAnywhere as well. I'm going to add a disclaimer to the front of it. There's a lot of concept and visionary work in the following slides. All the design shown may not reflect the actual implementation of the product. But these are just some cool ideas that we've had.

So, first and foremost for me for the future, is remaining committed to keeping up to date with your latest tech trends, remaining committed to keeping up to date with the challenges in what you're working with, as I suggested with the Windows Virtual Desktop project. But committed to our focus of any app, any device, anywhere, anytime. Okay. So, for AppsAnywhere to be successful, it needs to be everywhere the student is, as we've talked about with the Canvas project. And also, in order to do that, we need a more flexible user interface. So, more integration means more delivery methods, more of options for the user, more confusing information that we have to throw at them. And, as always, we want to make that as simple as possible. So, the challenge is presenting them that information in a simple way. And we believe there's a lot we can do in that area to make that even simpler for the user.

So, this is a little bit extreme. I understand that. It's a concept, bear with me. But the easiest way to stop a user getting overwhelmed with too much information is to just take it all away from them. Okay? So, if you think about it, in 95% of the cases, all the user wants to do is find the app they want, click launch, and then know when it's ready for them to use it. So, as we've got a screenshot here, we've just simplified the AppsAnywhere interface down into a big search box. You've got your favorites, recents, and you can tab things underneath. But chances are you just want to search for the application. And then, as soon as it's found it, you just hit enter or click launch and then the application is there ready to go.

So, everything these days is driven by search. Even when I go to my phone, I don't know where any of my apps are on my phone. I try and organize them, but they just get lost. You just swipe down, you search for the app you want, and you launch it. And everybody is used to seeing this page every time they open a browser. You just open a browser, it says, okay, here's some popular stuff, but you probably just want to search and go to the first thing that you need. If they do need more info, like the 5% of cases, then obviously you can just click the information next to it.

And we started thinking that, in the current user interface, if you've got 500 apps in your list, people are probably just searching anywhere instead of scrolling through the list. And, if users aren't viewing that whole list, then we're also using a lot of server resources and a lot of browser rendering time, as the Durham guys found, rendering a list that people aren't even looking at. So, why not just focus completely on a basic search? And, like I said, those 95% of the cases, the uses just got the simplest experience that they can.

If you click launch on words, we can also provide more feedback within the browser. I'd like to see us using the AppsAnywhere client to keep the user interface up to date with the launching status, as you've got at the bottom here. And then, once the apps are ready to go, you can just see the application was launched and got a notification style approach to keeping the user up to date with what's going on.

As I already mentioned, if you want more info, it's available. But again, if we're talking about now the 5% of cases where people want more information, I would imagine that in 95% of those scenarios, the only information they actually want is, why is it unavailable. So, we could show that in the results there. We would just add a Y button, and that could be as simple as a little dialog box that says you need to be in an app. And the user has got all the information they need. They turn it off, they go to an app, or they find another app and they launch it.

So, I said that was the extreme scenario. If that's a bit too much for you, we can still smarten up the original interface. We can get rid of that full list of apps that people aren't browsing. We can provide more useful categories on the left hand side. We can highlight search more. And we can still do all the things around feedback in the current style of the user interface. So, again, interested in your ideas of just simply down the interface and making it simpler user experience. So, that's any app, any device, anywhere, anytime. As I said, that will always remain our foundation.

The next point for the future that I want to make is focusing on a great experience for everyone. So, we've always been about the student experience. It's core to everything we do. We always focus on getting a student the ability to launch an app in as few clicks as possible. But I don't think this has always manifested itself in the administrative experience, the experience that you guys get in the back end. There's still parts of the admin experience that require you to copy IDs in and things like that, stuff we could make a lot simpler for the administrator as well.

So, we could start by just polishing things up, simplifying the navigation, making data easy to read, like just adding the icon in next to the app name here makes it easier to spot the app that you're looking for and find it in the list. And we've got these quick action buttons. I'm basically just reeling off all the ideas that Lewis came up with, so I don't want to take his credit too much. But you can see that it's a much nicer experience for the administrator. It makes things a lot easier to find and a lot easier to navigate.

This transcribes quite well to the delivery method management as well. So, where these changes, these subtle changes make the biggest difference, I think, is places where there's a lot of information to present. So, this is where we need to think about design to make the data more easy and quick to find and simpler to manage. We need this interface to be intuitive. So, you can see here, once you get into editing delivery methods, you've got multiple levels deep into information. Just presenting it in a way that allows you to quickly find the setting that you need is tricky. So, we've broken down the navigation here into very distinct different layers and it makes it easier to find what you're looking for. And again, Lewis has put in a nice sort of use of icons so you can see exactly what you're looking for, which are the Cloudpaging delivery methods, which are the downloads, for an example.

And finally, we want the admin UI to be API driven. In order to rebuild the UI in such a way, we want to have it all driven by APIs in the backend. And then, that gives you the ability to configure this and do it in any way you want and build that into your existing systems. It's something we get asked for more than anything, I think, from the backend UI point of view. Okay. So, that's the second goal, a great experience for everyone, not just the end user.

I've talked about making the product easier to deploy, but I also want to see easy to expand as well. So, AppsAnywhere currently connects to eight different external services. There's a potential for two more with Windows Virtual Desktop and packaging service. You can run your servers on premise or in the Cloud. And most of our customers have multiple instances of AppsAnywhere across production, staging, and test. So, that's a heck of a lot of environments to manage and a lot of connectors just to manage connected to AppsAnywhere.

I imagine a world where we take as much of that as possible off your hands. So, we start by linking to your hosting environments at the top. We help you monitor and manage the apps anywhere I'm environments. There's a little hint there to the customer managed upgrade process. So, here you've got your production environment. Here you've got your staging servers. One of them is ready to go. Two of them we're still in the process of setting up. And, once those are all ready, then that swap with production button could be enabled. And that could just take you through the process of swapping the environments around.

We can show you the status of all your connectors and allow you to easily add new ones. And, if we were to connect to all your hosting environments, then there's potential we could automatically create a new environment for you, such as the test environment, which we don't currently have in this example. If you clicked add there, there's no reason why we couldn't spin new up a server in as you, deploy the appliance for you, and set it up as the test instance for that AppsAnywhere environment.

So, as another example, let's say you don't have analytics yet. Maybe, rather than deploying it, configuring it, and connecting it, it would be great if we could do all that for you, like I said. So, you just click add and it says, where do you want that? Do you want that in your on premise environment or do you want it in the Cloud? Same for Windows Virtual Desktop. You could just click add sign in to connect into your environment. And there's no reason why we couldn't handle the rest of the setup for you. So, I see you've already configured Windows Virtual Desktop. Would you like to use that environment or would you like us to configure up all your host pools and deploy your base image and just get you up and running from the start and let you handle it from there?

So, they're easy deploy, easy to expand. We want to make using AppsAnywhere and getting the benefit from it as simple as possible, not just for people coming on board with AppsAnywhere, but for our existing customers as well. Like I said, there's so much you can connect AppsAnywhere to. There's so much value there. We want to make it as easy as possible for you to see that in the product.

And finally, just a quick note on that proven return on investment, once we've got all that information connected, once it's all going through AppsAnywhere, we've obviously got the AppsAnywhere analytics product. That was only just released this year. We're only just starting with analytics. So, I'd like to see all delivery methods metered in analytics. There's no reason why we can't show you the usage across all your delivery methods, improve the return on investment on those delivery methods, and then maybe one day even tie that into licensing costs and show you that usage against the costs that you're paying for those applications.

Okay. So, let's quickly wrap up. Very exciting times for Software2. As always, I'm really excited with the direction that we're going in for the next year, the vision that we've set out, and what we can do with this product, and the value that we can bring to our customers. I'm looking forward to giving you that value more often or more frequently with our new processes. And, as always, taking any feedback from you guys to help us drive this process. As I've said, the next year's direction is very much dependent on the feedback we get from you and how you're going to use these features. So, please, as always, get involved. Come and talk to us on the customer forums. Drop me an email. Thank you very much, guys. Cheers.