Hillel Cooperman and Tjeerd ("cheered") Hoek are two of the key figures in the Windows User Experience team at Microsoft, and they've worked on some the company's more advanced user interface projects over the past several years, including MSN "Mars," Internet Explorer/shell, Windows "Neptune," Windows XP, and now Longhorn. While my first (somewhat humorous) run-in with the Windows User Experience folks came during a Windows XP Beta 2 event in Seattle three years ago, the team has been working tireless toward Longhorn since the early days of Windows 95, when it moved Windows to the Explorer shell.
This interview is the first in a series highlighting the personalities behind the technology. It's often all too easy to depersonalize a company as large as Microsoft, but there are real people behind these products, and they care very deeply that the products they create are as aesthetically beautiful, functional, secure, and reliable as they can be.
Tjeerd: I started at Microsoft in 1994 as a product designer. My background is in industrial design and engineering. The key disciplines for product development that we have at Microsoft include product design, user research, program management, development, test, and UA (user assistance); at times we call it UE (user education) as well. UR is user research. Typically, the key team that is driving design of features includes one person from user research, one person from product design, one person from program management ... they try to get the plan together for what the product needs to do, and how it's going to do that. That design helps them develop a spec, which they give to the development teams.
I started with a product that was called At Work. We worked on printers and copiers and scanners ...
Paul: Right. I always thought [At Work] was a great idea.
Tjeerd: Yeah, we did too, but it never really panned out. A little while later, that was canned and part of the kernel actually became Windows CE over time. So I went to [the Office team] and started working on a little bit of Office 95, Office 97--it was originally going to be Office 96, but it slipped a year--and 2000, and I did early [work] on Office 10, which became Office XP later. And then I moved to Windows.
Paul: So you're not the guy responsible for all the crazy proprietary Office interface stuff, are you?
Tjeerd: Command bars?
Paul: Yeah. Every version of Office has to have some unique UI that doesn't exist anywhere else in any other Microsoft product.
Tjeerd: Right, well yeah, basically yes.
Paul: I think we need to talk. [Laughs]
Tjeerd: For example, I did File Open/File Save, and stuff like that. And I did command bars.
Paul: You're command bars?
Hillel: Go ahead, give it to him, Paul. Lord knows I have.
Tjeerd: Aww, it's not too bad. I wasn't responsible for writing the code ...
Paul: It's even worse if you're responsible for the vision of it ...
Tjeerd: So... [Turns to Hillel] He hasn't mentioned the Office Assistant yet, so that's good.
Hillel: That's not your fault.
Tjeerd: No, that wasn't my fault, that's true.
Paul: But arguably, you let it happen.
Tjeerd: I was there. I did vetoes of it. But I do still maintain that the Office Assistant isn't the terrible idea that people make it out to be. Actually, it's just the way [that technology] was used by many teams was worse than its actual implementation. It's actually just an animated puppet. And there's some truth to the idea of social UI. And some people really do enjoy it. Needless to say, they aren't the people who work in our industry.
Paul: Yeah it's also not the kind of people that spend a lot of time word processing.
Tjeerd: So I went to Windows to work on what we were calling at the time Neptune, which I know you're familiar with. Joe [Belfiore] was heading up that team for us, doing a lot of follow-up concept work to all the Neptune thinking that had already been going on. I joined the team at the time that we were trying to make Neptune real. And after that, there was kind of a reset, where we focused even more on shipping the NT code in [what became] Windows XP. That's actually the point where Hillel and I, and a bunch of other people from the team, when to David Cole. He had this ambition to build a client for the MSN service. And there was already some work going on at the time, to try and merge IE and instant messaging and media player...
Paul: Right. So this was the Mars project.
Tjeerd: Yes, exactly. So we did that, we worked on both the client and a bunch of the pieces of MSN that we started to pull in as we realized that we could actually get a holistic experience for users around those services. We needed to take more ownership of all these separate pieces of MSN. And then, after we did a couple of versions there, which of course shipped fairly quickly after each other, we were both called back to the Windows team, to merge back into the Shell team, and make it a lot broader. A whole bunch of people came back.
Paul: Right. Didn't a lot of IE guys come back at that time as well? Joe B...
Tjeerd: Yes. Joe Peterson was the head of that [new Shell team] and we basically created this Windows User Experience team under them. So I own the design team within that organization and I report to Hillel. We have about 34 product designers, which is a uniquely high number of people. This company typically considers 10 designers a strong, big design team...
Paul: You know, I never heard of you guys until the PDC. Where were you hiding?
Hillel: We try to keep the troll under the bridge, away from the public. Away from the press, especially. They're worried about what I'm going to say.
Paul: I see.
Tjeerd: So that's me.
Hillel: [to Tjeerd] This is your first job out of school, right? You came right to Microsoft.
Hillel: Now, I'm one of the rare people that actually worked in the real world before I got here. I've been a Microsoft employee now for six and a half years. I've been a Windows user for five and a half years.
Paul: Wait, wait...
Hillel: Ah. So I worked in a few small companies before I got to Microsoft working on Mac software. One company, we did database consulting and a bunch of small Mac utilities. Then I went to work for Aladdin Systems--StuffIt and all that--and then I got laid off there. And I actually said no the first time Microsoft offered me a job, and that was a big mistake.
Tjeerd: Well, it worked out...
Hillel: Yeah, it did, but trust me, it was a mistake. I've got other things to make up for it. But not things that were as good. I started at Microsoft as the program manager for Java for Mac IE.
Paul: [Laughs] Oh, man.
Hillel: That was my first job.
Paul: So your prospects were looking good!
Hillel: Yeah. Well, remember the days. Right? We had a browser, we were going to do all that stuff. And then within a couple of months, I took over as the lead program manager for Mac IE, period. So I was working on Mac Internet Explorer. I came at the tail end of when 4.0 shipped and really shipped 4.01, which was much better. Then a bunch of us quickly realized that the presence we had down in San Jose was nothing ... really, the action and excitement was up here. Really, it's different today. But back then, the excitement was up here. So we figured up here we'd have a much stronger presence. And so we all said, we're going to Redmond. And we're going to work on Windows. So we picked up and we left.
A bunch of folks went to work on different things, but I ended up working for Joe Belfiore on this pre-Neptune, researchy, next-generation Windows user interface project. But I was very na?ve about how things worked. So I didn't realize it was more of a research project than a shipping thing. And then we did Neptune, which was still a little more research than product. And then I convinced Tjeerd to come over, after begging him for more than a year. And then I was co-program manager for MSN Explorer for a couple of versions, so we shipped that, and that was very exciting.
And then I came to work on Longhorn. I'm what they call the product unit manager of the Shell Core team. Think of it, in a sense, as the classic shell, the absolute core user interface, both around the storage work that we showed [at the PDC], and the core user interface elements, like the task bar and the Start menu.
Tjeerd: But the cool thing, of course, is that the team, apart from owning a number of specific features, and having the devs directly build those, we are responsible for driving the more holistic user scenarios that exist in Windows for users. So we really try to step back and think more broadly about what people do. We go to other teams and try to talk to those guys to work out how we can make things more seamless for people.
Hillel: In addition to the Shell Core team, I'm also part of the MSX team, which is the product design/user research/user assistance folks.
Paul: What is that? Does that stand for Microsoft eXperience?
Hillel: Yeah, something like that. The interesting thing is that even though we're focused on Windows, because of the nature of the product as a platform operating system, we end up trying to think beyond Windows as a platform, both for the rest of the company and for the industry, which is kind of a scary responsibility, which we take on with humility but high aspirations. So that's kind of what I do.
And by the way, I'd like to point out that I saw comments that ... what was the comment on me? That I sounded like I was on acid or something crazy...
Tjeerd: Who wrote that?
Hillel: It got posted somewhere on some Web site and there was a discussion board and they were like, "Who's talking? That guy sounds crazy." And by the way, Paul, I blame you, to be clear. There's something about the way you shot the video, and acquired the audio, that really transformed me into sounding weird...
Paul: Yeah, that sounded nothing like you. [Laughs]
Hillel: I'm standing by that claim.
Paul: Well, I apologize for that, I guess. But ...
Hillel: Just kidding. I have to take my lumps.
Paul: Well, you can tell you're not from Microsoft [originally], right? Because most people from Microsoft are ... are ... what's that word?
Hillel: You know...
Paul: No, I mean, in other words, you get the typical [Microsoft] executive up there ... and it's not very exciting.
Hillel: Oh, come on. Watching a Group Vice President code? That was cool.
Hillel: I thought it was exciting. And what about Ballmer? There's no one more exciting than Ballmer.
Paul: Yeah, so anyway.
Paul: Let's just say there have been moments ... in my many experiences sitting in the audience at a Microsoft event, where they will cart out some person who is not from the [computer] industry. They're just interesting because ... all of a sudden you realize that there's been a standard here in the computer industry, especially with Microsoft, where these people are a certain way, and they are what they are. What was interesting about your talk [at the PDC keynote] and then also your session, was that you don't have the stock Microsoft PowerPoint template going, and ... it's interesting. You really stood out.
Hillel: Cool. Well, I appreciate that.
Paul: It bodes well for Longhorn. You don't have the "three bullet point" thing going that we always see.
Hillel: It's a funny thing. It's very easy to look at a company--and I'm not saying you're doing this, but I did do this--and see some of the very obvious spots where we could be less boring, less formulaic, or whatever those things are. There's something underneath all that, which has resulted in a lot of the success the company's had, and I can't take any credit in that. I got here and realized there were a lot of things that I personally cared about, and I was certainly not alone or being original; I wasn't a genius for thinking of things I wanted Microsoft to do. I found lots of people who got here out of school, who cared about the same things. But I eventually realized that you didn't need to replace anything that Microsoft was good at to do them. It was an "embrace and extend" strategy. Things sometimes feel a little conservative.
Paul: You don't really notice it until someone else shows up on stage ...
Hillel: So there's contrast, yeah.
Paul: And then you really see it.
Hillel: Oh, I noticed it.
Tjeerd: That's another thing we were trying to do at the PDC. The User Experience team had been appearing there [at past PDCs] and had often had messages that were just, "don't do this. Don't do that." They were negative, prohibitive kind of messages. And we were trying to turn it around, and do something positive and, very deliberately, do a bit of whipping our own backs, recognizing that we really haven't lived up to the things we've asked of the ISVs.
Paul: It's a little malicious though, because you showed off all this great stuff that you weren't actually handing out at the PDC.
Hillel: Yeah. But if we hadn't showed it, you would have said we were mean for not showing our plans, right?
Paul: True. You're the bad guys either way.
Hillel: And we know that. It's OK, we accept it.
Paul: So the User Experience team started up in the XP time frame, right?
Hillel: There's always been a user experience team for Windows. It was just called the Shell team. Back before XP, I suggested--I worked for Joe Belfiore at the time--that we rename it the Windows User Experience team. But it's been the same team. The transition that happened, the growth that happened in terms of the charter and the actual size of the team, happened after XP, for Longhorn. But the attitude--and you can see it even in the XP UI--is an attitude of, hey, it's time to really start evolving forward in terms of the UI. That definitely started with XP. So we renamed the team, and we finally saw Microsoft investing in aesthetics. You know? It's counter-intuitive, compared to what Microsoft's done in the past. And that started, not with Longhorn, but with XP.
Tjeerd: Well, it can go to a new level. I think when we're pushing products, it's often hard to get the right dev done, make the right investments, and we knew for XP that we had to go to a new level. And now with Longhorn, we will go to another whole new level. And that's where the Avalon investments really come into play.
Paul: Did you look at XP, to some extent, as just testing the waters? To see how people would respond to a richer UI, or was it just the first step towards Longhorn?
Tjeerd: It did what they could do, I think, at the time, with the technologies ...
Greg Sullivan: Yeah, I think it was the first example where this more holistic approach came into play. It wasn't "holistically holistic," but it had a couple of key scenarios where we did a better job on the end-to-end. Photo acquisition is an example. It's not just incrementally better. We said, let's take a couple of scenarios, and improve the end-to-end experience, and recognize that this is the model we are using to approach the industry. It's not vertically oriented. We don't make the hardware. We don't make the devices. But if we are not the place where these things come together in a more end-to-end way, then it will never happen.
Paul: So clearly, in Longhorn, there is going to be way more of that [end-to-end, holistic solutions], it won't just be a few key scenarios?
Hillel: I don't want to over-promise, but we have big dreams. We're really trying to have a level of attention to detail that surpasses anything out there. And we think we're resourced appropriately for it, and it's really just a matter of time. We're not going to get to everything, but we're going to do a bunch of things we can be really proud of, and that customers can respond it. We're working really hard on it. The thing that's heavy lifting for us is that we don't just have these user experience goals in a vacuum. If our goals around Longhorn were just about improving the user experience, we'd be done a lot sooner. There's an ecosystem to involve first, and to aspire with different innovations, inventions, and platforms. But I'm glad [Longhorn] isn't just [a new user experience].
But it makes our job harder, right? The more powerful you make something, the more difficult the user experience is going to be. That's just a fact. You want to make something simple? Make it do less. Well, our customers have made it clear that they want our software to do more. How to balance those things has been a challenge for us. We don't want to ignore our heritage, our strengths, our quest for the ideal--on paper, anyway--user experience. Because we think that new user experience is going to accomplish a lot for users.
Paul: Sure. So what changes? One of the examples you talked about was the different ways that you can handle things like CD burning. In other words, for a power user you might have this right-click method, but for other people, it steps you through a wizard. Where do you draw the line for this stuff? Is everything in the UI going to be this two-tiered structure for doing things?
Hillel: There are really two things to compare and contrast here. One angle is, OK, what are the general tactics that you have in your toolbox that you're going to use? And you brought up a few of them: Right-click, wizard, click and just do it, et cetera, and so there are a bunch of those. We look at that as, we have a canvas to paint on, and we have a bunch of paints to use, and those are the different paints.
I'm mixing metaphors all over the place. [Laughter]
There's a separate issue, which is, how do we focus our efforts there? And it really leads back to what Greg said. Well, we pick the scenarios we care about. For example, managing tons of information. This is one we're going to nail. Make sure we bring to bear all the resources we have, all the tools in the toolbox, and really embody the value system that we articulated at the PDC in that scenario. There will be a couple of scenarios where ... yeah, OK, we didn't get there this time. But we want to get there at some point.
The scenario focus, I think, is the most interesting, because it lets us really measure things. We said this at the PDC. Microsoft is very good, in my opinion, at giving you the pantry. Here's one thousand ingredients. You can do anything. And it's very powerful. And we've been fortunate enough to be successful, and I think that's part of the reason for it: We give customers choice, opportunity and power. The flipside is, we have not always been good at saying, hey, if you need to do X, here are the pieces, pre-connected for you. But we did a much better job of that in XP, in my opinion. So I think XP certainly shows the start of that. Before XP, we were not very good at that. And I think Longhorn will take that to the next level.
So we measure things not by, hey, does it follow the guidelines for right-clicking, or a wizard, or whatever, but more on, let's look at the scenario, solve it end-to-end, and let's make sure that the customer can find it, use it, and get their job done. Finally, they should be psyched they used it.