Brent Simmons: You're listening to The Omni Show. Get to know the people and stories behind The Omni Group's award-winning productivity apps for Mac and iOS. Music.

[MUSIC PLAYS]

Brent: I'm your host, Brent Simmons. In the studio with me today is Ainsley Borque Olson, OmniPlan Prime Minister, Greg Titus, OmniPlan engineer and Ken Case, CEO of The Omni Group. And this is a special episode. We're talking about the story of OmniPlan. Say hello, Ainsley.

Ainsley Borque Olson: Hello Ainsley.

Brent: Say hello, Greg.

Greg Titus: Hello Greg.

Brent: Say Hello, Ken.

Ken Case: Hello Ken.

Brent: So general question, I think might go to Greg: How did OmniPlan start?

Greg: Back in the day when we had lots of spare time for spare time projects, I —

Brent: The early 2000s, roughly?

Greg: Early 2000s, roughly. I had been working on Graffle’s automatic layout and that got me interested in optimization problems of various kinds, and I started looking up papers and came upon project management resource leveling stuff that I thought was interesting. And so the original OmniPlan prototype was basically enough user interface for me to be able to mess around with writing the algorithms that I thought were interesting.

Brent: Now to be clear, this was you having fun.

Greg: Supposedly, yes.

Brent: I love your definition of fun.

Greg: Well, you know, yeah.

Brent: Yeah. “I'll read some academic papers on resource scheduling, and I can write an algorithm.”

Greg: I think that's fun!

Brent: So you wrote enough user interface scaffolding to be able to enter tasks and resources and things, right?

Greg: Right. Basically the bare minimum just for me to be able to mess with stuff for myself.

Brent: Did you give this a ... it must have had a name of some kind before it was OmniPlan.

Greg: I think it was actually always OmniPlan?

Brent: Really? You just said it's going to be...

Greg: Sort of the company culture at the time, everything was Omni-blank. We had Omni-everything. You look in our old SVN logs for various projects and stuff, and there's Omni... 120 different things.

Brent: That would be a fun bit a of data mining to do. I should write that blog post. All the apps that we didn't make, but could'a.

Greg: Yeah.

Brent: So about how long was this your kind of fun thing before people started to know about it?

Greg: I think that I probably worked on it for a couple of months and then it sat around for maybe six to nine months, maybe a year before anything came of it.

Brent: So what happened next? What took it from a thing that you had worked on for fun and kind of set aside, to it becoming an Omni app?

Greg: We had kind of a internal meeting on, “hey, this, we're doing products now, [becoming] a product company, what should our next product be?” and I pitched it. And I don't really know why, I don't remember what the situation was, that was the reason why we're doing that at the time, I don't know, maybe Ken does, but—

Brent: Was consulting winding down, Ken, or what was going on?

Ken: That was certainly a piece of things. We were trying to move out of the consulting business and just be able to focus on our own products, but we needed to have enough of a stable of products making money to be able to do that.

Brent: And there was OmniWeb and Outliner at the time already?

Ken: OmniWeb was our first paid product. We had OmniGraffle and OmniOutliner. Around the launch of Mac OS X and we also had OmniDiskSweeper, we were actually selling back then it wasn't just the free download.

Brent: OmniDazzle?

Ken: I don't think that was before OmniPlan. Maybe around the same time.

Greg: I think that was around the same time.

Ken: I think that was the same sort of pitch session. A short project doing long project. A supposedly short project. It was meant to be like a month or something.

Brent: Was there a meeting where everyone got together, an all hands thing, and people gave pitches, or ... ?

Greg: That's my vague memory of it. Yeah.

Brent: All what? 15 employees at the time.

Greg: Something like that.

Ken: We could certainly easily fit around a table, let's put it that way.

Brent: Oh, okay. Right. Yeah.

Ken: It was a big table, but...

Brent: That was several locations ago. University of Washington area?

Ken: Yeah. It was just north of U Village.

Brent: Yeah. Okay. Any idea of what else was pitched? I take it the company was into productivity so there could have been a spreadsheet app, for instance, or—

Ken: We weren't just into productivity at that point either. Like some of the consulting projects we were just coming off of were games. So that was some of the stuff we thought about as well.

Brent: So the identity of Omni as a productivity app company was still being molded at that time. Yeah.

Ken: It was not settled, yeah.

Greg: Yeah. We were doing a bunch of different things besides the consulting, or doing game ports from PC to Mac for a while there, and we had thought about a multidimensional spreadsheet as one of the other possibilities.

Brent: Does anyone know what that is?

Greg: I do! They're cool.

Brent: Back to your idea of fun again.

Greg: Yeah! Yeah. Oh yeah.

Ken: Yeah. Anyone who used Lotus Improv on a NeXT, I guess would know what it was.

Brent: Can it be explained understandably and briefly to our listeners?

Ken: So your traditional spreadsheet has your two dimensions, right? Your X dimension, your Y dimension, your rows, and your columns. And a multidimensional spreadsheet just adds additional dimensions to that. In Excel or something you would mostly do this now with extra sheets that you would use, of the spreadsheet, but in something like Quantrix, which was a derivative of Improv, that was produced by Lighthouse Design before they got acquired by Sun. Ancient history now.

Brent: That's why we're here.

Ken: In that sort of thing, you would... you could take your extra dimensions and you can put them along the horizontal axis or the vertical axis, and it would end up sort of grouping by that extra dimension. So you would see... I mean it's still obviously presented on a screen in two dimensions because that's what the screen is, but internally it was multiple dimensions, and you could easily have five dimensions going on, or more.

Greg: So like one of the, just as an example, a sort of a common thing to do would be have, you know, say products as one dimension, and then like months as another dimension, then years as another dimension. And so you have the expected arrangement of that. You could show like products as rows, and then months and years as columns, but then one of the great things about these multidimensional spreadsheets is you could manipulate the dimensions, so you like, you reverse the order of months and years, and you can see all the Januarys together, all the Februarys together. So you can see seasonal things as opposed to over time, just generally over time things. So the idea of it is to be able to rearrange the dimensions to see different things about your data.

Brent: Had you written a prototype of this already too?

Greg: A little bit.

Brent: Now, I was getting there.

Brent: But OmniPlan won. What was the status of OmniFocus at the time?

Ken: At the time OmniFocus hadn't even been thought of.

Brent: Okay. Did Kinkless GTD exist, or was that also still in the future?

Ken: People might have started using Kinkless by that time. I don't remember the exact timeline of when this meeting happened, vs.... But at the time I think we were thinking of, “well, how can we make OmniOutliner a better host for Kinkless?” Like make it faster, and so on. And when we got into those product meetings, we ended up deciding, “oh, let's just make that into its own app” rather than trying to make OmniOutliner be one app that does all of the integrations and things that OmniFocus now does.

Greg: As I recall, Tim [Wood] tried for a while to enhance the AppleScript support to make Kinklesss better on the existing Outliner and we were... we kept running into technical issues where it was just always going to be kind of slow to do it via scripts rather than natively.

Ken: Although he made it a lot faster than it had been.

Brent: At best though you've still got the Apple Events layer to deal with. Yeah. So OmniPlan got the nod maybe because it was most finished? Or it seemed less crazy than multi-dimensional spreadsheets, or...?

Ken: I don't remember the... I mean, as I guess I alluded to, multiple things got the nod, but this was the big project that really ended up, I guess, lasting. Let's put it that way.

Brent: Okay. Were side projects fairly common? I've heard OmniWeb described as Ken's side project in that sense.

Ken: It was a very short side project originally. Although it quickly distracted all of us. Well, we're coming up on the 25th anniversary of when that happened. Right? It's like in a month.

Brent: Yeah. Outliner was originally Luke Adamson and somebody?

Ken: Luke Adamson worked on it. It was originally some code that, I think Greg worked on some pieces. No?

Greg: No.

Ken: I worked on some pieces of that. It was originally a program that I started writing to help my wife keep track of our books.

Brent: That explains the multiple columns. Part of the original. Yeah.

Ken: And also to keep track of, then, our projects. So before we had OmniPlan, you know, it's a way to kind of do budgeting and so on. So that's part of the reason that OmniOutliner had math in columns. It's able to do sums.

Brent: I think it's the first Outliner I ever saw with columns, and I just thought it was really, really weird at first, but now I totally use that feature. It makes sense. So when OmniPlan came out, it was kind of the era of, where GTD was first getting a lot of buzz. Did people assume you had written a GTD app, and then...?

Ken: As we started getting close to being ready to announce OmniPlan, we started hinting at, “hey, we have a new app that's coming.” And one of the things we shared was the app icon, without providing any sort of name for what this thing was. And people immediately started thinking, “Oh you're doing a to-do app, a GTD app.” Well, no, that's not what it is.

Brent: At least you're not on record as saying “we will never make a GTD app. There's no market for that.”

Ken: I think by that time we were starting to think about what our next project might be, and I think we were starting to think about doing OmniFocus. So we ended up announcing OmniFocus earlier than we had originally planned, just to explain to people that if they were looking for OmniPlan to be that thing, that was not what OmniPlan was. That's not what we were building there.

Brent: So, how do you define OmniPlan? It is for... I mean, OmniFocus is for your individual work, and OmniPlan is for—

Greg: OmniPlan is for project management. I mean we do have users who just use it to plan their own things themselves, but I think that really the intention is that you have multiple resources, that there's multiple people involved, and multiple things involved in finishing your plan. The important thing about it is the relationships between the tasks that need to be done, and the people that are available to do them.

Brent: And when they're available. And things like that. Yeah. It finally ships 1.0 comes out in, let me look at the history page.

Ainsley: 2006. 2008?

Brent: 2006. You got it. Yeah. December 13, OmniPlan 1.0 release. So how did it do, off the bat?

Ken: It was a good launch, yeah. I was trying to remember, is OmniPlan 1.0 the app that we showed on stage at WWDC?

Greg: Yes. I'm not sure if it was 1.0, but it was it was the year that they were doing Objective-C garbage collection. And Ken went up and had a project plan that involved, “here's how you make an app,” and it had this big block for finding memory leaks and stuff like that. And, “but now with amazing new Objective-C garbage collection that Apple is introducing this year, you don't have that whole big chunk of code...”

Ken: We shrunk that section.

Greg: “And of course this copy of OmniPlan that we're running here on stage is built with Objective-C garbage collection. Yay!”

Brent: Yeah. Cool.

Greg: And I was sitting in the back row. I remember just covering my eyes, thinking “Don't crash, don't crash, don't crash.”

Brent: I was at that. I remember that one.

Ainsley: That's the first I've heard of it.

Brent: How was it to go up on stage during... It was during the keynote, as I recall, right?

Ken: It was during the State of the Union.

Greg: The second one, yeah.

Brent: So, still the big room and everyone's there.

Ken: Basically the same set of eyes. The press are gone, but all of the engineers and developers are there. Yeah. It was, I don't know, an audience of 5,000, 4,000, or something like that. I was a little nervous.

Brent: I would have been.

Greg: Better you than me.

Ken: A lot of practice, As Apple does for every—

Brent: So they helped you go through this, or...?

Ken: They ensure that everybody who's up there has done a bunch of practicing in front of ...

Greg: ‘Force’ is a better word than ‘ensure.’

Brent: Had it taken a lot of work to get OmniPlan actually garbage collected in time for this?

Ken: Yeah, there was a lot of work. And of course that version never shipped.

Ken: One of the hard things about garbage collection in the apps, as opposed to the automatic reference counting (ARC) that we do now — and that, in fact, we finally switched to with OmniPlan just recently — is that garbage collection was an all or nothing proposal, where if you wanted to have garbage collection in an app, everything the app touched and used had to be garbage collected. Whereas with something like automatic reference counting, you could just do a single class at a time, migrate it, and it could talk to the old retain-release world just fine.

Ken: So even the C code that is using Core Foundation and so on, had to be garbage collected, and it was not very natural to do that with some of the pieces.

Brent: Let's see, and OmniPlan 2 came out, the history page tells me, in 2011. So what changed? Ainsley you're not the PM yet when two came out, but you've studied the history.

Ainsley: I started working here shortly after we released 2. In 2 we introduced a collaboration via a publish and subscribe mechanism, as well as change tracking to go along with that, so that OmniPlan can now be a tool that multiple team members could use simultaneously on a project.

Ken: Is that when we added Microsoft Project import/export?

Ainsley: I wasn't sure on that, if that was initial in 2.0.

Greg: It was within 2.x, but I don't remember. I don't think it was right then 2.0 came out.

Ken: It might not have been for 2.0.

Ainsley: And that certainly is a big part of Plan to be able to open files from Microsoft Project users on a non-Windows machine. We have lots of users who find that very useful.

Brent: Collaboration implies some kind of shared server or something. Did it work over WebDAV or...?

Ainsley: Yeah, so this is before I would say a lot of people were using a service like Dropbox. We didn't have iCloud yet to collaborate on work files, so you have a system that syncs to either a WebDAV server, our own Omni Sync Server, or when we shipped it, it could sync with MobileMe, which is long since discontinued by Apple, and the precursor to iCloud.

Brent: Post-cursor to iDisk.

Ainsley: iDisk was part of MobileMe.

Brent: It was part of MobileMe. Oh, okay.

Ainsley: I think .Mac was the precursor.

Brent: Oh, .Mac. I don't know why this has had so many different names.

Ken: And there’s iTools.

Brent: iTools, yeah. And iTools was also the name of the thing from Tenon, which did Unix-y Mac things.

Greg: We had a different model too from things like Dropbox and iCloud drive now, where you're sort of essentially sharing the same file, but it's intended for one person to share the same file, but just between your own devices. And so there's kind of less worry about, “hey, what if you change the same file and two different incompatible ways on two different devices?” because presumably you're one person and you're not simultaneously using both hands on two different devices. Whereas OmniPlan explicitly wants to deal with the situation where you have multiple real people who are making changes, that potentially conflict with each other, to the same project. And that's why change tracking was such an important part of the sync strategy.

Ken: It's a lot more like Google Sheets or something, which didn't exist back then either.

Brent: Were the apps still run by the lead engineer at that point? Or were we using product managers?

Ainsley: I checked with the former project manager over lunch. Aaron Kwong took over as PM in summer 2007. So, shortly after OmniPlan 1 shipped, and then I took over from him three years ago. Shortly after we shipped out a OmniPlan 3.

Brent: Before that—

Greg: Before that it was whatever my whims were, is what we did.

Brent: What features do we need? Whatever sounds fun to Greg.

Ken: And I was sort of acting as PM for all of the products before we split up that into some real PMs.

Brent: And the company was small enough, and everything. That made sense at the time. Did any of you ever go out and actually meet with project managers and get feedback, and do that kind of thing? Or help them get set up?

Ken: Between 1.0 and 2.0, yes. Not really for 1.0, that I recall, unless we started hearing from some project managers.

Greg: Once 1.0 was out the door and was kind of getting some uptake, we had a couple, some customers that we would go and we gave a couple of trainings to and you know...

Brent: Including some fairly large customers.

Greg: Some fairly large customers, some fairly large groups at a time, you know, in front of a classroom type setting. And I'm teaching people how to use the app, and taking questions and stuff. And so you get some good feedback that way about which parts are hard to understand.

Ken: When we shipped 1.0 we actually had hired somebody to help us work on it, who had previously shipped to his own project management software. And so he had a list of customers that were interested in him building a new version of MPX, I think was what it was called. And so, you know, we offered all of those folks an upgrade.

Brent: Oh, nice. Yeah.

Greg: And so we had all of their requests for however long it was that that had been out.

Ken: So it wasn't in a complete vacuum.

Greg: Right.

Brent: Right, yeah.

Ainsley: Now we have 12 years of user feedback in our bug database. And then I've taken a project management course, so I've interacted with other potential PMs as they learn to project manage — and this is a course at UW — but how the UW teaches the skill set that you would use OmniPlan for, which is really interesting.

Ken: And I grew up in a family where my dad was a project scheduler for Boeing, so he would be bringing home these big Gantt charts that were just rolls of paper and then work on them at home in the evenings. So, long time exposure to some of that world.

Brent: OmniPlan on paper though.

Ken: Like for a 747 or whatever.

Brent: Yeah, wow. Then the next thing was probably the iPad app, as part of the iPad or Bust initiative. And were you working here, or right after that?

Ainsley: I started a month after I think we shipped OmniPlan 1 for iPad.

Brent: And so was that a huge job, making the iPad app?

Greg: It was a fairly big job, obviously, but I think for OmniPlan, compared to like Outliner, and compared to Graffle, that we were doing at the same time, I think that OmniPlan was actually easier. We were very fortunately kind of written the Gantt, and a lot of the display stuff for OmniPlan, in Core Animation. So lower level display, API, then AppKit from macOS. And so we always, for all of our apps, we share the model code that actually deals with the innards, but generally have to rewrite most of the user interaction display code between macOS and iOS, and OmniPlan, fortunately, a lot of the actual display stuff could also be shared between both platforms, and the only thing that you had to rewrite was the interactivity.

Brent: Ah, it's kind of lucky then.

Greg: Yeah.

Brent: Core Animation to the rescue. In our pre interview you mentioned your most favorite thing in all this was the animating labels as you scroll, is what my notes say. So, what's that?

Greg: Yes. So as you have the Gantt chart on the screen, it keeps the labels visible as as you scroll horizontally. The labels move to stay displayed, and then they stick to the edges of the bars. Then as you keep scrolling, they go over the bar when that's available. And then on the other side... I really like it because you expect Gantt charts to be this very stuffy thing, and I just liked to play with scrolling them around. So it's a little bit of fun that you don't really expect in that sort of area.

Brent: So we shipped OmniPlan 3. When was it Ainsley? When did OmniPlan 3 come out?

Ainsley: 2015.

Brent: What happened in 3? What was new in 3?

Ainsley: So, in OmniPlan 3 we introduced a feature split for the first time between a Pro and a Standard set of features, which at that point we'd already done in Graffle, and this allowed us to offer a slightly lower entry price point to customers who didn't necessarily need the collaboration features. So OmniPlan Standard has the full scheduling functionality that Greg was describing, but it doesn't include the collaboration. So if you are, say, a one person team or you only have a few resources, it's a good fit for you. But if you need to export to Microsoft Project, or collaborate with multiple team members, we were able to offer a Pro version, which... this was the, I think maybe the first app that we had a second version of to sell on the App Store, if I remember my release timeline right. So that allowed us to offer an upgrade path for customers who had purchased OmniPlan 2 through the Mac App Store. Maybe Focus had done it first, I'm not sure.

Ken: I think Graffle had maybe done it first, but this was pretty soon after.

Ainsley: Oh Graffle probably had, between 6 and 7, yeah.

Brent: Ken, was that part of the reason for doing Standard and Pro, just to be able to do some kind of upgrade?

Ken: At the time, we hadn't landed on our current solution of just offering the download in the App Store for free in the first place, which lets us kind of do whatever we want with our upgrade pricing. But that was what we had landed on then to let people have some sort of upgrade pricing, was you can use the in-app purchase, and it's required if you want Pro, to buy Standard with the download, and then buy the in app purchase as Pro. But if you owned a previous version of the app, if you owned OmniPlan 2 already in the App Store, and you download OmniPlan 3, then you would get a free upgrade to pro. And so we can give you basically half off.

Ainsley: OmniPlan 3 also introduced network view, which I think was one of our more commonly requested features before then, as well as a multi-project dashboard.

Brent: So what's the network view?

Ainsley: The network view ... Greg can probably describe it better than me.

Brent: Is this something that sounded like fun to do and you did it?

Greg: Oh, of course. Well the network view is, actually that's one of the things that a lot of customers ask for. It's also known as a PERT charts, incorrectly known as PERT charts. It's a different method of laying out the tasks so that instead of emphasizing the scheduled order and dates, it emphasizes what depends on what else. And so that was, yeah, one of the big missing pieces in OmniPlan 1 and 2 was that sort of display.

Ken: If you go back and look at the early project management software on the Mac, like MacProject, or even earlier than that, LisaProject, that what you see on the boxes, or their typical screenshots of those project plans was actually this network view.

Brent: Ainsley, was my favorite feature included, The Monte Carlo simulation?

Ainsley: Monte Carlo simulation was introduced in OmniPlan 3.

Brent: Yay. I just had to mention it because, listeners, there's a video, a marketing video, that goes along with that, that will be in the show notes, and you should see video cause it's awesome.

Greg: It's the best explanation of what Monte Carlo methods are, ever.

Brent: It's quite special. Starring Aaron Kwong, the previous prime minister.

Ainsley: Aaron Kwong's now our primary tester on OmniPlan. So he hasn't strayed far.

Brent: He still knows the app quite well. Yeah.

Brent: So that gets us through 3, and most recently it seems like it's been, cleanups, consolidation around the sharing and syncing kind of stuff. Right?

Ainsley: Yeah. We've been looking at where customers run into trouble using OmniPlan, or what they'd like to see improved. So last year we did a big overhaul of the publish and subscribe collaboration mechanism we described earlier. So it still has the same underlying functionality, but we made the interface for setting it up a whole lot more discoverable and easy to use. Because we found customers were looking to do what OmniPlan could already do, but were struggling to set it up in a way that worked for them.

Ainsley: We're also, in OmniPlan 3.12, which we're currently working on, and is currently available in public test. It may have released by the time this podcast comes out.

Brent: Yeah. This is probably March 20th. As people are hearing this, I think.

Ainsley: Yeah, we'll see! We're making some updates to the network view. When we released it in 3, the most common thing we heard was that folks wanted a way to see long task titles in the network nodes instead of truncated short titles. So we're introducing that in 3.12.

Brent: Can you talk any about the future of OmniPlan without, I suppose, giving out all of our fine tuned details.

Ainsley: Well, Greg is currently working on automation on OmniPlan.

Brent: Okay, so that JavaScript...

Greg: So we'll be adding that JavaScript support for automating, which we've already shipped for OmniGraffle.

Ainsley: And then we don't have any sort of timeline yet, but we would like to improve integration with our other apps like OmniFocus.

Brent: Sensible.

Ken: As we add sharing of tasks in OmniFocus between different OmniFocus databases, we want you to also be able to share those tasks with OmniPlan, or more specifically, sort of the other way around. Right? Like you're, somebody on your team has set up a shared project plan and your piece of it you would like to see in OmniFocus.

Brent: Oh, that makes sense. So pushing from OmniPlan down to—

Ken: But then as you check things off, your status updates can flow back up to the OmniPlan project. So they automatically get the updates.

Ainsley: Folks have been telling us they'd like to see this for a very long time.

Brent: Oh I bet, yeah.

Greg: And it was on the road map earlier, and then Apple shipped the iPad, and we did the whole iPad or Bust deal and that put us off of doing that...

Ken: We had to set it aside for a while.

Brent: So is the code actually mostly done in one or other of the apps? Or is this a case of where you, nah you really need to get started?

Ken: An iteration of the code was done.

Greg: It was kind of working for a little while and then things have moved on, and I don't think any of the almost-working bits are left, and would need to be redone. But it's certainly, we've proved that it was technically feasible.

Brent: Yeah. Wow. That's cool. The [2019] roadmap talked about focusing on polish to a certain extent. Bugs and performance, and that kind of stuff. Is OmniPlan taking part of that or?

Ken: Oh, absolutely. Yeah.

Ainsley: Yeah, for sure. I mean, so unlike a lot of our other apps, OmniPlan has a really specific customer base. We primarily have project managers. Like you need to do project management in OmniPlan. Whereas say, OmniGraffle, you've got all kinds of different use cases for the app. So looking at how we can make project managers’ lives easier by improving features that we have.

Brent: Cool. So for people interested in helping Beta test, where should they go?

Ainsley: Unlike our other apps, as well, we have a pretty small pool of folks who are willing to run test builds of OmniPlan. As soon as we start testing something in OmniFocus, we get tons and tons of emails. We started our OmniPlan 3.12 public test recently, and I've only heard from a couple of people. So, if OmniPlan customers are interested in running tests builds, those are available from a link on our site: omnigroup.com/test. Links to all of our actively running public tests, and it's a great place to check if you're interested in testing what's new and hot in OmniPlan.

Ken: And if you're interested in this integration between OmniPlan and OmniFocus, please send email to omniplan@omnigroup.com, and just let us know, and that way we can add you to a list and tell you specifically when those builds are ready for testing.

Brent: Cool. I'll make sure all this stuff's in the show notes, so people can click on the right things. All right, well I think that pretty much covers the story. Thanks, Ainsley. How can people find you on the web?

Ainsley: You can find me on Twitter and Instagram. At... I can't even say my user name, @Ainslaw.

Brent: Well thanks. Greg, how can people find you on the web?

Greg: I am @gregtitus on Twitter.

Brent: All right, thanks Ken. How can people find you on the web?

Ken: I am @kcase on Twitter.

Ainsley: I'd also like to thank our intrepid producer, Mark Boszko. Say hello, Mark.

Mark Boszko: Hello Mark.

Brent: And especially, I want to thank you for listening. Thank you.

[MUSIC PLAYS]