Season 2, Episode 4: Sarah Barrett
Sarah Barrett is the senior information architect for Microsoft Docs, where she works with content creators, engineers, and executives to bring order to one of the largest sources of technical information anywhere. Her practice, developed on projects for companies like Adobe, the University of Washington, and Expedia, melds the lessons of old-school information science with enterprise-level data analysis and just enough research. Sarah holds a master’s from the University of Washington in Library and Information Science.
Dan Klyn: Good morning everybody. Welcome to another TUG Sunday Service talking about information architecture. And today, I'm so happy to get to talk with Sarah Barrett, who is the Senior Information Architect for Microsoft Docs, which, if you've not seen it, you can go to docs.microsoft.com. It describes itself as the home for Microsoft documentation for end users, developers, and IT professionals that encompasses almost everybody that I know.
Thank you for being here with me today, Sarah, and I'd love to hear about what, what even is information architecture at Microsoft? As soon as I saw your name, and your attribution as that out in the world ( I think it was when you gave a conference presentation and I just happened to see that), I'm like, "Information Architect at Microsoft. I didn't know they even had those." So, I'd love to just know about what IA is at Microsoft and, and, and what you do as an information architect there.
Sarah Barrett: Sure. yeah, as far as I know, they didn't have information architects before they hired me. I was there on a really brief contract and then, there was a full time job posting that was for an information architect I couldn't believe. And it was actually a great job posting, which never happens in IA. [crosstalk 00:12:24] I felt really lucky.
I think Microsoft had previously had a very robust practice and then in the kind of early 2000s, it was de-prioritized and everybody kind of either had to become something else or left in a really tragic way. And, in, for whatever reason, the, the desire was there about when I started. the design team, when I started, was me and another guy sitting in a room. And, some, somehow, like the second hire they made, maybe the third hire-
Wow.
... was an information architect, rather than somebody who could do anything else. [Laughs.] Because I'm kind of a specialist, so I feel very lucky for that. Now, we've grown a huge amount and, I kind of unofficially lead a team of three. so we have people who specialize on kind of the front end side of the, front end, information architecture like really looking at user research, developing navigation, doing journey maps, all that kind of stuff. We have somebody who's really more on the content strategy and taxonomy space and then I most likely, like, herd cats.
We're in the Developer Relations organization, which publishes all of the documentation and a lot of learning content for, Microsoft's technical products. So, the way we explain it is that, like, if your dad is using Excel, that's not us. that's Support and that's really valuable but we don't do that. but if you're developing for anything, if you're administering anything, anything like that, that's all on our platform. And, I say that I do Microsoft Docs but it's about one of, maybe a dozen, 15 websites, that's in the, learning, like content learning documentation space that Microsoft has that, we run. Because Microsoft loves having lots of different websites. we're trying to fix that. We killed a lot of them.
so yeah. As far as I know, mine is the only real team of information architects at Microsoft. I keep trying to find other people. I know that there's one other that works with our Analytics organization but he doesn't get to do quite as much IA work. and I found a huge number of people who, have previously done IA and even previously kind of thought of themselves as IAs and had like really deep, theoretical background and a huge amount of experience who now are called something else. and so, a lot of my first couple of years there was finding these people and really building that coalition across groups. We could get stuff done with more than just me.
And I can't be the only one for whom it's a joyful discovery that there's somebody called that at Micro... Do you have that? Do people go like, "Oh my God, I'm so glad that you are an information, that we have that 'cause that's important."
Oh yeah. Absolutely. We have a very large, Content Development and Maintenance organization. and most of them have a, a big appreciation for that I think. And there's like, "Oh my God, somebody else will do this. Somebody else will like understand what I'm trying to do here." and then, I get that from outside too. People are like, "Wait a second, like you actually get to call yourself an IA? That never happens." And I feel very lucky for that.
Have, have you bumped into resistance in your environment there? I mean, we don't have to talk out of school or anything like that but with, and especially given who the primary user of Microsoft Docs is, when they hear architect, they think about something and then there's what you do and, there may not... So I'm curious if you bump into software architects, system architects in your day to day who, ask about what an information architect is?
interestingly, I've gotten way more push back from actual architects, thinking that it's really presumptuous for us to call us IAs. It's like, "Do you have to go to school for that? Are there tests?"
Oh no.
No, no. There are no requirements whatsoever, I make it up. [Laughs.]
[Laughs.]
Actually, the the technical architects and the software architects have been by far my biggest, allies, I think. our technical arch, our back end, former back end current lead technical architect recently got to like quit all of his other responsibilities and just focus on the technical architecture for Docs which is like his dream job. And he is my, like best colleague who works in a completely different organization. I think the mindset of the architectural approach is really common and, I feel like I've, I have learned a lot about how IA should work and especially how to do it at this scale from him. And he's really changed a lot of the way he thinks about how we build things based on the work that I've done there and, like we're actually in the process right now of overhauling the whole backend of this enormous system because it was built to house about 1300 documents four years ago and now we have 75 million.
Wow.
[Laughs.] And we haven't changed it at all. [Laughs.] It's still just like working the way it's working, just harder.
Whoa.
So he's, re... I might be, I think that number, that number is correct if you could count for all the localized content. But I could be wrong. but like it's just changed massively in scale and so, we're re-architecting all of the back end because we're doing this work for the Department of Defense. I don't know if you've heard about that. Mixed feelings about that.
But, there is a very interesting technical problem of making the whole site run, in a completely air gapped environment. So, it won't have any outside communication with the Internet. we get no analytics. We can't push anything through. We just have to make it run basically. It's like, it lives in a container. Like, like a physical container basically.
and so-
Now also, in a bottle. If it, if it-
No. Exactly.
If you need it in a bottle, we can bottle this.
So, we're re-architecting the whole backend and he's doing that. not completely very much informed by the, like user facing content modeling work that we've done. So that's just been an incredibly productive collaboration for me. I get a lot more PM, I get a lot more pushback from PMs and people who are much more interested in just like shipping something good right now. whereas the devs are like, "Wait a second. We have to have this thing forever." Like, we're going to build this and then, we're going to be stuck with those consequences. and even in, even in that case, it's mostly a case of, education and, communication, getting to talk to each other and like really thinking about what the incentives for behaving that way are and all of that. But, I've been pretty luck over all.
Wow. And, so sometimes I flip flop on, the answer to is it difference in scale or difference in kind? Like it, it, is there some sort of an axiom that you can use to say that, no, no. Once this is an order or two of magnitude bigger, so in your case, the document, the docu-verse the world was conceptualized with and then the real world of the docu-verse that exists. Is that difference in scale at Microsoft where you work, have you worked in smaller, info verses that had different dyna... I guess that's what my question is.
Yeah.
Sorry for dragging that out. But, what is my question here? Is it, is, is what you do, is the way that information and people and understanding works with littler things different than how it works at the scale you're at now?
Yeah. So, I consulted with Factor, working with Bram Wesland and Gary Carlson for like five years or something like that before I moved over here and we did a lot of projects that varied in scale. you know, we did some really little things, where, you know, it's, it's important that we had a big, a big audience but, you know, it was a couple of hundred pages or a couple thousand pages or something like that. I remember my interview for this, they were like, "Have you ever done anything to this scale?" I was like, "Oh yes, I did a side audit of 35,000 pages." And they were like, "That's great." [Laughs.]
[Laughs.]
Because I had looked at the website before I went to that interview. in my defense it's really hard to get a sense of the depth of the thing just from looking at a homepage.
Yeah.
so I had worked on things that I thought were big and then things that were pretty small. I don't think what makes good information architecture changes [inaudible 00:20:52] what people need from the experience really changes but I think what my job is and can be changes quite a bit.
Yeah.
Sarah Barrett: like, it's funny because you're like, "Oh, go to the website and you'll get to see it." And I was like, "Oh, please don't go to the website, it's still terrible." [Laughs.] Because I feel like anybody who knows anything about AI would go look at it and be like, "Who does this?" Like, somebody is actually doing this information architecture because this is terrible. 'cause it is in lots of ways. but it's better than it was three years ago and, there's just a huge amount of, kind of like sub-structure.
Well, to be fair, people look at what is the logo top left there and it's, and given that it's one of the biggest companies around, you just presume that they're shipping their org chart and so, if this is a disordered, if this Microsoft Docs site isn't good, it's not Sarah's teams fault, it's, it's Microsoft. They're-
That's makes me feel slightly better, thank you.
Dan Klyn: They're “too fail to big” as I like to say. so I'm curious about the, when you worked with Bram and, Gary, I presume, that stakeholder work, because you're an external consultant, there's always some like aligning with what the stakeholders want and there's some special chef kiss magic sprinkles on you as an outside consultant maybe because they resent your bill rate so much, that they listen to you, that you can get the deciders to align and, find some shared purpose and then you help them move forward.
And, and when I think about the difference between being an outie, being a consultant and being inside, that's the part that I wondered, do information architects still do a lot of that, you said cat herding, this is more like, lion taming. So I'm curious about the, the amount of stakeholder work that you do as an innie IA and, and how that's different from when you were an outie IA.
Sarah Barrett: Yeah. I think, that's one of the main reasons why I went in house, is that I could just even doing, you know, I really stand behind the work we did. I think, I did great work with Factor. I think they do great work. but even doing amazing work, I'm sure you understand, you can feel yourself making shelf ware sometimes. Where you're like, "Oh, this is-
[Laughs.] I've never thought about it that way. I just tensed up. Like, oh God. I'm just making something that's going to get filed that nobody's ever going to look at or use.
Yeah. And like, so, I know for a fact that some of the work that I was most proud of, and that represents like the most nuanced user insights and, you know, the, the most novel, you know, best AI ideas or whatever, ended up never getting implemented. And so, I really wanted to get a chance to follow things through and figure out how to make stuff actually happen. and even kind of knowing that, one of the things I realized after being here for a little while was that, I had never fully appreciated how much work those project sponsors did before we ever got in the room.
Dan Klyn: Isn't that amazing?
I knew, but I didn't know [laughs] how many years of work they did before I could come in and tell them something obvious. that would all go, "Okay." [Laughs.]
Dan Klyn: Well, and, and if they can hold that, oh man, yeah I know exactly what you're talking about. And-
Sarah Barrett: Yeah, I think, it took me, I would say about a year and a half of being in this job before I got to do what I thought was IA. The rest of it was building relationships and education and like the art of making things happen. Like bringing the organizational maturity up to a point where they could hear, like, "Hey, these are the things that you should put in the navigation." and I don't think this is an indictment of my organization. I think that's just like how long it takes. But like, for reference, I started in October of 20, God, 18, 17, something like that. Whatever the eclipse was, right then. [Laughs.] And, we shipped the like obvious persistent navigation that I put on a whiteboard back then, this week.
Dan Klyn: Whoa. That's that's one of those aircraft carrier turning things.
Yeah. And like, we actually started keeping track once we, once I expanded the team. my colleague Rachel Price really took that navigation work over and did, did most of the work around it and to like the final design and all of that. And she, we've been doing the match and we have about 50 hours of meetings convincing people to one hour design work. so it's mostly talking. It's mostly relationships, it's mostly making things happen. especially for the big things. You can get a lot of little things done. It's hard to get big visible things done.
Dan Klyn: This may be an unfair question but I'll ask it anyway. When you look back at you, think about you in high school, were you able to get, to make things happen, have you been a make things happen person, once you became autonomous as an actor in the world, is that, is that just something you're good at? Or is it, as oppose to something you needed to learn through lots of, contextual bruisings and,
Sarah Barrett: no. I was very much the, like I did programs who didn't do the reading. like that was kind of my whole thing. [Laughs.]
Dan Klyn: [Laughs.] Oh, that's great.
Sarah Barrett: like I, I was reflecting the other day because my best friend from college who I'm still very close with is the exact opposite of this. Where it's like, I didn't learn how to use a flashcard until I was probably like 28. this was very much a later process and like, the, the art of getting things done has been a, a thing of very deliberate study for me.
Of, a thing of very deliberate study for me. speci,- specifically while I was at Microsoft, where I was like, "There's clearly something here that I don't know how to do because I write."
[laughs]
Sarah Barrett: It's obvious. this isn't anything new. Like, I'm literally going into multiple meetings about why bread crumbs should exist and that's wild because-
[laughs]
Sarah Barrett: -this argument has happened. And like [laughs] you know, this isn't, this isn't trying to convince some- anybody of anything new. so like, why is this so difficult?
And so I really sought out people who were very good out this, and like information about doing this better. And we now have a very, now have a very deliberate way that we work on the information architect team. We have like a three page document that outlines what our operating principles are, how we have meetings, what kinds of deliverables we produce, like how we make stuff go.
there's this incredible article that I really recommend called Policy Entrepreneurship at the White House by Thomas Kalil. I think that's how you pronounce his last name. I'm not sure.
Dan Klyn: Spell it.
Sarah Barrett: K-L, K-A-L-I-L.
Dan Klyn: Okay.
Sarah Barrett: Policy Entrepreneurship at the White House. And basically he started up this team in the Amab- Obama administration whose job was to get other people to enact policies. And his team was made out of people who really knew a lot about things like technology and education or certain kinds of green energy or whatever it was. They were subject matter experts in some capacity. but they had no authority, which sounded very familiar to me. I have no authority to do anything. [laughs]
And so he has this incredible article, how they did it, and he has like photos of his white board that they used for onboarding and all the operating principles and that kind of thing. And it is like 100% in my mind how you get things done in big organizations. And there are some of those that we've really really taken to heart, we've really operationalized on the team.
like one of my favorite ones of his like, "What is the most important thing and why aren't you working on it?" so like if you had five minutes in an elevator with the President or whatever-
Dan Klyn: Yeah.
Sarah Barrett: What would you tell him was the most important thing? You haven't done anything on that in the past two weeks, have you? Because it's really true, in these big organizations you get caught up in like the art of the possible and just responding to fires.
And so, you know, it, it's happened to me a couple times where I've been in a meeting with some executive something and they've been like, "What should we do with navigation?"
And I'm like, "Show Jeff. I don't know." [laughs] Like, I can't do anything with navigation. I haven't even thought about it. [laughs] So, like it's this huge problem that I have, that I know I can't do anything about so like I haven't thought about it.
So we've set outside time where once a month the team and some like special guests comes together and we just have a full day where we do a workshop. No screens and be like, "Okay, what would we do if we could do anything?" So we all have the same answer next time that happens and we can all kind of inform it. We don't build that thing, but like we think about it. And lots of other things like that.
Dan Klyn: And, did your education in graduate school prepare you for any of this stuff? You went to the University of Washington, to the I ischool. Is that right?
I did. I have my MLIS. I was a medical librarian when I was in school. it's much less. It, I, it wasn't very practical in that sense. I think most of the people who go to library school are very diligent, detail oriented people.
Yeah.
in a way that I was not naturally but I think I have become as an adult. [laughs] I went straight from undergrad to grad school and didn't know anything about anything. but it was 2009 and I had to have somewhere to be [laughs] because it was not a great place to ... not a great time to be like looking for a job. but I found the theoretical background to be incredibly valuable. I, I feel pretty firmly that very little of what we are doing is new.
As if we were issued new brains when the internet came out or the iPhone was developed or whatever, and so the basics of information seeking behavior, the basics of retrieval, the basics of knowledge organization, all of those things are constant.
And the, the harder things get the more I go back to like my readings from grad school [laughs] and those kinds of things. Like I actually have the like Patrick Lamb's Taxonomy's book sitting on my night stand because I'm building an ontology for the first time and I don't know how to do it. [laughs] And all kind of things like that.
So it's, it's been very valuable because ... I saw a very good quote the other day about like how data can only tell you about what's happened in the past. It can't prepare you for something brand new. You need theory for that and that's that's really what I, I go back to that theory for.
Dan Klyn: So back to stakeholders then and this work you need to do to persuade people to, to find that shared purpose to get permission to change something that's more than ... That's, that's one of the ways that I've been talking about it when people are like, "Well, information designer, architect, what, aren't you just UXers?"
It's like, "Well, we make the most sense when somebody needs to make more than an incremental change." Most of the clients that we get to work with have developers, designers. they have all the things, but what they're ...
Oops, oh the space bar.
Sarah Barrett: [laughs] I thought that was me.
Dan Klyn: The space bar, what do I do? I, I started gesturing at my keyboard. the thing that they don't necessarily have is the ability to be led through something more than an incremental change.
And so that's one of the, one of the ways that I've been talking about it. And in order to do that, you either need to be persuasive or you need to have better research or better theory. So I'm just curious how you're operating and, and what, what resources you're, you're drawing from, what bags of tricks? So it sounds like grad school gives you some theory. consulting gives you some interpersonal mojo. What else does it take to be a successful information architect at your scale?
Sarah Barrett: That's interesting, because I, I, I'm not sure if I agree with the incremental change thing.
Okay. Yeah.
mostly just because ... That might be the biggest difference at this kind of scale. like, there's a lot of stuff that's very visible but honestly doesn't matter very much. Like it's a really big deal for us that we ship this persistent navigation. like we've been trying to do that for a long time. When people talk about, "Oh, what does an IA do?" "Oh, I design navigation." Like that's kind of what you expect.
Dan Klyn: Yeah.
It kind of doesn't matter for us. Relatively speaking very few users use it. Nobody necessarily expects it to be that great. it's important in terms of something like basic place making.
Yeah.
But we very effectively trained our users that are navigating, to know that our navigation is terrible and they don't click on it.
and, but that's like one of those big disruptive ... I don't mean disruptive but like it's one of those big overhauls. It's more than incremental change in some ways.
a lot of what I do is focus on how we do our work. and not just me and my team, but as a, as a whole organization, so that we are more, we make more sense in a year than we do right now.
[laughs] Oh wow.
which is really hard, because like-
Yeah.
-the natural state of things is entropy. [laughs]
Dan Klyn: And, and how do you, how do you know? How do you, how do you measure? Are you, do you have a ... Okay, is the objective to be more sensical, more understandable by interval whatever and then the result is, "We tested people and they were able to understand X,Y,Z better?" I'm curious how ya ...
Yeah, this is actually one of the arguments I have a lot at work where people are like, "Well, what, what's, what's the biggest bang for our buck? Like what should we do first? What would change things?"
And I'm like, "I don't know what will have the biggest impact but I can tell you that we're doing nothing. Like, we are not doing any of the things that you would normally do to make something make more sense." So like, "Let's do something," and it will be slightly better. I can almost guarantee it."
Yeah.
so, like, I spent part of the last week working with some of the content managers and the PM who runs our doc set creation process to develop a workflow for assigning base URLs different doc sets. Very boring. however, we have this ... I mean, I don't find it boring. I found it kind of fun [laughs] but it's not glamorous [laughs] and it's not high, it's visible, very.
But like right now we have two different sets of documentation of the site. One that is like WAC team and one that is WAC Microsoft teams. They both have content about Microsoft teams published by two different teams,-
Oh-oh.
-organization, because they're are so many new doc sets that get created that we don't realize that this has happened until it's been out there for a while and somebody goes, "Wait a second." So we're trying to make sure that doesn't happen anymore. That has a very minimal effect right now, but in a year things might make a little bit more sense than they do right now.
there are all kinds of ways we think about ... Like tweaking slightly the stuff that everybody is doing all the time to help better information architecture emerge.
Dan Klyn: Yeah.
Sarah Barrett: I can't go reorganize 75 million things. Like, I can't, I can't put them in a spread sheet. [laughs] I can't, I can't do any of the normal ... I don't have a site map for the site. It's too many things. I can't do any of the ...
I don't have like the normal levers that you would usually have. So what I can do is I can set up governance processes where somebody looks at it. And that doesn't mean everything goes through my team 'cause there aren't enough of us. It means that we make a little checklist that says what makes something good.
And then we put it out in the world and it gets slightly better. It means that we, like we do all kinds of little template and documentation and brown bags and like educational work so that everybody is doing better information architecture with all the choices that they make, if that makes sense.
Dan Klyn: Well, it does make sense and I want to live in that world, and-
Sarah Barrett: [laughs]
Dan Klyn: But I wonder, it sounds ... Am I wrong to guess that you choose to make it be like that, that nobody went around asking asking for you. "Can you help us institutionalize-"
Sarah Barrett: No.
Dan Klyn: -some processes that are more about structural integrity of meaning than they are about just shipping shit all the time. Like, it sounds like this is you thinking this.
Sarah Barrett: yeah. this is the thing that had kind of emerged for awhile as I kind of transitioned from doing project work on the outside to working inside. And actually in a weird way it's something that really crystallized in mind was on maternity leave. I had a baby last year [laughs] and just made me think a lot about maintenance.
Dan Klyn: Okay.
Sarah Barrett: because as anybody who has a baby knows, you-
Dan Klyn: Yes.
Sarah Barrett: -do the same God damn thing over and over and over. [laughs] and I also read Jenny Odell's book, "How to Do Nothing," which is very good, if you haven't read it. She talks a lot about a performance art. Not a lot, she talks a little bit about a performance artist named Mierle Laderman Ukeles, I think that's how you pronounce her last name, who coined or developed the idea of maintenance art. and she wrote this incredible thing in 1969 called The Manifesto for Maintenance Art 1969! Exclamation point. [laughs]. just like the most 1969 thing in the world. I love it.
And she talks about their being two basic systems in the world. There's development and maintenance. and actually I just brought it up because I love it so much, but she ... because it's like poetry, but she talks about maintenance being the sour ball of every revolution. After the revolution who's going to pick up the garbage on Monday morning?
Dan Klyn: Yeah.
Sarah Barrett: I think in tech we are very dedicated to development. Like, it is literally in the name of [laughs]-
Dan Klyn: [laughs]
Sarah Barrett: -our, our departments and that kind of thing. and she, she calls it like the death impulse, like it's the, the radical overhaul that gets rid of things that were and creates something brand new.
And like that's fine and that's what everybody wants to but while, when I kind of first came in house I was like, "Oh, I designed a lot of home pages without thinking about how that home page was going to stay good over time."
or like, you know, I created a little box in Axure because I'm now old enough that all the tools have changed [laughs], because like I created a little box, right? I still don't know how to use Sketch. [laughs]
Dan Klyn: Me neither, nor Sigma nor any of the things.
Sarah Barrett: I glad that some people are still using Axure. I think it's a better product but what do I know. I don't-
Dan Klyn: Oh, Joey Thompson is using Axure, so.
Sarah Barrett: Yeah,
Dan Klyn: You have people.
Sarah Barrett: I don't, I don't do page layouts anymore, so. but like, you know, you create a little box and you're like, "Oh, featured new content goes here." Or whatever the thing is and it's like, "Cool. That means that there's going to be content that is featured and good and new and that forever."
Dan Klyn: Mm-hmm [affirmative].
Sarah Barrett: Like, how's that happening? and that's, you know, a small example but for us it's, it's everything.
Dan Klyn: I think that's about as good of an example as you can get. I think that's, that's terrific. By what mechanism, is there something showing up there that's always going to be good?
Sarah Barrett: yeah, and so she talks about maintenance as "keeping the dust off the pure individual creation, preserve the [inaudible 00:40:55] change and protecting progress." And that, the way I see it is like our entire work. I can, I can put some things in a hierarchy and it can become some [inaudible 00:41:08] You can throw some terms and taxonomy, whatever the thing is, but it's all about doing it over time. and like, the, the over time and like sustaining it and distributing it among people so it can be maintained is that, that is the work for us.
Dan Klyn: Oh, I, yes, I want to live in that world.
Sarah Barrett: [laughs]
Dan Klyn: And I'm thinking again as you make the mouth words about how you see this, like, "Yeah, I want to ... Where she lives, I want to live there."
you mentioned the PMs who need you to ship it and that, that tension, I, I think part of the reason why I want to live there so badly is because it's such a rare occurrence when making things be good can cause a deadline to move. And I'm curious if that's if you have strategies for that or ...
Sarah Barrett: Yeah, and like the new stuff that ships is probably some of the most frustrating part of my work. And like some of it that ships I'm really proud of and it's great and some of it it's like, "It's out there. It's fine." [laughs] "We hope nobody clicks on it anyway. It's okay." [laughs]
Dan Klyn: [laughs]
Sarah Barrett: one of the things I ran into I think is that different ... It's a big organization who creates this site, these sites. There's several of them. and because it is a big organization different people and different groups have very different experiences of what this thing is. like I think it's the classic you know, blind men look at an elephant think it's four different things or whatever.
Dan Klyn: It is very differently performant for different audiences? Is is like bad ass for visual studio people and terrible for some other-[crosstalk 00:43:07]
Sarah Barrett: no, no, not even in that sense, but like, so the, the PMs work primarily on the new features and even then they're not a monolith. Like some of them work on the brand new things that we shipped a year ago and they worked pretty well. Some of them, it's their job to get the dot net reference auto generated from an API to show up on the site.
it's somebody else's job to figure out to put the Microsoft graph documentation, just literally about a graph and how it works, into a hierarchical table of contents which is just against laws of physics but it's all we have for them.
Dan Klyn: [crosstalk 00:43:46] Yep, yep.
Sarah Barrett: So like they have really different experiences of the site and how it works. Similarly, like the designers kind of see the new stuff we ship mostly which isn't saying we don't have the perspective but they, they don't have the, the luxury of looking at all at [laughs] one time because they're very busy-
Dan Klyn: Yeah.
Sarah Barrett: -working on the new stuff. The devs have this crazy old code base that was built to do 1300 things where like, "You can't touch that. That's a load bearing bug." [laughs] 'Cause like they don't know how it all works.
Dan Klyn: Oh my God, I've never heard that before. Is that a Microsoft phrase, "load bearing bug?"
Sarah Barrett: I think it's a joke that one of us made and I'm really-
Dan Klyn: Ohhh.
Sarah Barrett: I've really like glommed onto it. I quite like it.
Dan Klyn: I love that. Thank you.
Sarah Barrett: [laughs]
Dan Klyn: Just say you made it up until you-
Cool.
-[inaudible 00:44:27].
I, I'm sure I didn't, so.
[laughs]
Sarah Barrett: I think that is a joke someone else made. but yeah, like this, this complicated thing where you're like, "I don't know what's going to happen if we, if we touch that."
Dan Klyn: It's broken but we can't really work on that part because it's-
Sarah Barrett: Yeah, and so one of the biggest ... And then we have all of the content people who actually like live in this system and are trying to publish content in an ongoing basis and like deal with all of its idiosyncra- idiosyncrasies and all the problems this year. So like they all have very different perspectives of how well thing works and what parts of it are painful or not.
and I am very lucky in that the development organization and the content organization feel the maintenance need very acutely because they're, they're doing it. And so I have the luxury of working with them directly and they're like, "Yes. Incremental change. Yes, how do we make this little thing [laughs] this thing a little bit easier? How we just do our jobs a little bit differently so that something emerges."
we have to do both the big overhaul changes that get some get some eyes on them and you know, help people feel better about everything that we did. but a lot of the work is that, just helping everybody do everything a little bit better.
Dan Klyn: Fascinating. Well, and I love that two worlds way of explaining that duality. do you miss nuke and pave? Do you miss being able to just make shit up and [
Sarah Barrett: laughs]
Dan Klyn: -not have to have it be plugged into a real sustainable ecosystem?
Sarah Barrett: Oh, God, sometimes.
Dan Klyn: [laughs]
Sarah Barrett: that's actually one of the the like emotionally very satisfying things of doing those those days that I was talking about where we just like unplugged. we'd disconnect from the reality of it all [laughs] and just figure out what we would like to do is, it really gets some of those I don't know, that like consulting creativity. We're like, "We can do anything! I don't have to worry about-
Dan Klyn: Yeah.
Sarah Barrett: -implementing everything."
Dan Klyn: Yeah.
Sarah Barrett: I'm very big on, like op, like design principles and ground rules and that kind of thing. So for the IA team in general, one of our design principles is Execution is Everything. so we don't design anything if we can't do it at scale.
and that's one of the problems that we've had working with ... not problems but one of the [laughs] you know, things we've had with working with outside designer our outside [inaudible 00:46:59]. They're like, "You should have dates on things." And I'm like, "Cool. Which date?" Like, "On which things? How are we going to do that for everything?"
And they're like, "Oh, God, I thought that was a simple recommendation." [laughs] 'cause like very simple things can be very difficult with this many people, this big. So like in our real lives, execution's everything.
when we do these strategy days, like we don't worry about how we're going to implement it.
Dan Klyn: Yeah.
Sarah Barrett: Like we don't worry about anything realistic. And so like having those different kinds of, I think expli- expectations for different times you can like use different parts of your brain. It really does a lot emotionally, I think.
Dan Klyn: Do those, do those ground rules and or design principles, are any of those in variant? They are always part of how you are going to want to frame something? Are they often or always contextual? How do those work and who who ... Is that you cooking those up and formalizing them, operationalizing them?
Sarah Barrett: yeah, so the infrastructure and defining principles are for our team the three of us came up with them. and they, I think they're relative. They're in variant in terms of like how we do our work overall.
I, I use a lot of Laura Hogan's work about facilitating working groups and meetings and kind of thing. It's very good, if you're not all familiar with it. And she's a bit advocate of ground rules for individual meetings. And it's really the idea of like, "Okay, how are we all going to approach this specific social interaction? How are we going to think right now? What are the rules?"
and so like sometimes that's, "Hey, we are optimizing for transparency and recording this. We're going to send it out immediately." So, like, no secret [laughs]. Like keep in mind, like we're going to tell everybody everything."
Sometimes it's, you know, Vegas rules. We're gonna, "Whatever we say here, stays here," and like that means we get to complain about everybody who's doing things wrong. We get to be really real about like why things aren't happening.
Dan Klyn: Yep.
Sarah Barrett: And like those are both totally valuable in completely different ways. And when you're very clear with people that, what the expectations are, you can get much better interaction.
Dan Klyn: Yeah.
Sarah Barrett: Some, similarly, like sometimes you're really optimizing for like deeply understanding a problem and like noodling on it. And sometimes you're like, good en-, "We're going for good enough." Like, what are we going to do in 45 minutes that's [laughs] going to improve our lives next week or whatever.
So, like, for the ground rules they're very situational and we try to make those clear for as many meetings as we can. Like, "Hey, how are we having this meeting?"
The design principles are really longer term things about how we do our work. So, things like, exe- execution is everything. one of our big design principles which sounds silly but we come back to all the time is that we do not lie to the user.
Dan Klyn: Doesn't sound silly to me. That's sounds really hard.
Sarah Barrett: It's very hard, and especially because like, Microsoft products don't make any sense.
[laughs]
We all know that. Like, they, they are not clear.
Dan Klyn: Yeah.
Sarah Barrett: it's very confusing how they all work and I frequently get people coming to us and they're like, "Okay. Here's the navigation we think we want to do."
And I'm like, "Cool, is that how the product actually works?"
Like, "Well, no, but it makes more sense this way."
I'm like, "Cool. We don't lie to people." so like that is a conversation we have over and over. We don't tell people things are simpler than they are because that's not serving the purpose of documentation. In other contexts I think you might want to simply things for people. but that's not what we do.
similarly, like what do we do? One of our others is that we we really work on using our structures to include. we have the idea of like being radically, radically inclusive in whatever structures we build because it's very easy, especially in a technical space to become very exclusive in the the language you use, how you define roles, and whether people are in or out of those roles. All different kinds of things like this. We really see ourselves as a place where we can say, "No, we're not going to organize things that way. We're going to do something that's more open to everyone."
Dan Klyn: And that, that sounds like accessibility and there, it, it touches maybe but it also sounds like if there was an organizational company-wide accessibility platform, that what you just described wouldn't maybe not be covered by accessibility as, as often construed by large enterprises, which has to do with low vision support and you know, adherence to testable standards about legibility and reading level and, and those kinds of things.
What you're talking about sounds like it could also be like philosophical, not just implementational. That's not a word, but
Sarah Barrett: Yeah, yeah, 'cause I mean, categories and, a, a lot of, a lot of IA is about creating structures, is about creating categories and human brains love categories. Like, we love saying, "Ah, this is a thing. I'll put stuff in it. Now, that thing's really real." [laughs]
and it, depending how you structure those things you can, you can tell somebody that this resource, this site, this, this job, this field is for them or it is not. and I think that's something we run into a lot. I ...
People I work with often really want to organize things by role. This is like one of the several we see this, because it seems very easy. Like if we can just say-
Dan Klyn: Yeah.
Sarah Barrett: "Hey, this is for the this for this kind person." Everyone will know they're that kind of person and then they will just find everything they need there and we won't have to do any other IA. [laughs] Like we won't have to, we won't have to describe things in any other way.
And I really push back on that most of the time because one of the things that we see over and over in user research is people being like, "I don't know. Like, I'm not like sure I do make some applications in dot net but I'm not a dot net developer."
Like, or, this, this doesn't seem like it's for me. This, I don't, I, am an end user? Like don't know. I'm, I'm, I'm an administrator at [inaudible 00:52:59] but like, this seems very complicated" Or whatever. And so like those people are still our customers.
Dan Klyn: Yeah.
Sarah Barrett: And those role categories are things that we have created that say like what kind of a person you are. And that's like a really wild thing to have done. and so dismantling that as much as we can and figuring out what we have to describe about our content and our world and all of that, so people can find their place in it is harder but I think really worth it.
Dan Klyn: Yeah. Wow. So, what's the, what's the future for people who care about information architecture in organizations like yours? It sounds like you are teaching a lot of people. You're facilitating a lot. You're framing. You're enabling them to do what they need to do. Do they think of that as you are providing IA, you think?
Are they going to want ... Do they want more of you?
Sarah Barrett: [laughs]
Dan Klyn: Are there more wrecks gonna happen in Microsoft land do you think, for IA?
Sarah Barrett: [laughs] That's my-
Dan Klyn: ... wreck's gonna happen in Microsoft land, do you think for IA?
Sarah Barrett: That's my goal. like I kind of jokingly say that I'm gonna be Dan Ramsden when I grow up, [laughs] and like my goal was really-
Dan Klyn: Don't we all?
Sarah Barrett: ... to build a team here, I know right?
Dan Klyn: Yeah.
Sarah Barrett: not so good the close-up magic yet, but, maybe one of these days. I means let's the quarantine will be for. [laughs] but, but yeah, my goal here is to really build a team, because there's a huge amount of work, and, if you're not super caught up in building the most, I don't know, the m- the most exciting sexy new thing, there's a huge amount of work here, and then also I think there's actually a huge future in, like online learning experiences, as we can all see, and that is ...
Dan Klyn: Yes.
Sarah Barrett: ... what we're doing. like this, the current goings on the world have been awful in general, but for the specific world in which I work, like docks usage hasn't gone down, like helps reporting publicly on how, how, how big a spike there has been in a lot of the different services that we support.
so like there's a future for this in the world, and there's a future for kind of innovative new learning experiences there. I am extremely lucky in that I feel like somehow nobody I work with has realized that like information architecture is not that big of a deal anywhere else.
Dan Klyn: [laughs].
Sarah Barrett: I don't know how it happened, [laughs] but they're all like, "Oh really, like there's, they don't have a team of your information architects?" And I'm like, no everybody has one and we're behind." [laughs] I'm like, "We, we need to get a ... we need to hire, come on guys." so, like I don't know how I quite leapt into that, but, everybody, everybody I work with in Microsoft is like, there are only three of you, and so everybody who hears about it outside is like, "Wait, you have a team of three IAs? That's [laughs] incredible."
[laughs].
So, like different, different expectations, for us, yeah, I think it's, it's growing and specializing, it's, getting to do a lot more with, the user research practice that we're building, we're working in building out like the ontology and knowledge graph to really make a lot of other things, there happen, we're deploying a site-wide content model, like all kinds of things, that can be really transformative, we do it right. It's a lot to land, but we'll see.
Dan Klyn: That is really exciting, congratulations for being able to, have the patience, and the, the heart to, to bring people along for real not, just, on paper, it sounds like you, really want them to be able to do things, not just talk about things.
Sarah Barrett: I can't do it all myself, like I did ... I don't have enough hours, the three of us don't have enough hours, so like if we want to, if we want to actually have anything get better, other people have to be able to do good IA with-
They have to be able to do it, yes.
... without us being there, without us looking over their shoulders.
Dan Klyn: Yep. Awesome. So, before we open it up to our, audience here for some, some questions if they have them for you, I hope they do, your toolkit, we've, we've talked about what's in there, but I'm curious if you, if ... is there a thing or two that are, you're doing more of, and/or are there a thing or two that you do less of, and I will, for instance, personas. so like just tools, what's, what are some, good ones, important ones, rising ones, new ones, and then are there things that you don't do anymore, that you used to do?
Sarah Barrett: yeah. One of our probably contentious, design principles for the IA team is that we do not design for personas.
Dan Klyn: Ooh.
Sarah Barrett: I know, right?
Dan Klyn: Whoa!
Sarah Barrett: We, we focus on making everything like the lowest common denominator of logic, like [laughs] is it, all logical? And that is what we try to go for, because, we support so many different kinds of users that there would be so many different kinds of personas, and like I, we can't get into that, so is it logical?
Dan Klyn: There are no Visual Studio Steve, no, ...
Sarah Barrett: No.
Dan Klyn: ... no, I'm not sure.
Sarah Barrett: None of that. in terms of like what I do personally, I do much less like actually designing the artifacts of information architecture as we think of it, I still do some concept modeling, we're really influenced by, Sophia Voychehovski's work, with object-oriented UX, but also builds upon a whole other bunch of things.
It's been very successful for us, because I do work with the developers so much, and they're like, "Oh, it's relational database, got it, we'll build it." [laughs] I'm like, "That was the concept memo conversation, the first I'm introduced on?"
Dan Klyn: Wow!
Sarah Barrett: Which was fabulous. so I do a lot of concept modeling, I do a lot more ontology, and taxonomy work, which is pretty new to me, I'm not great at it yet, but learning, and so we do a lot more of that. I do a lot more, data sciencey kinds of things, because, a lot of our arguments are only as good as they can be supported with analytics data, and that kind things have like been getting really into our, in that kind of thing recently, so I can actually look at that amount of data, I can scrape the website, I can query our, all of our tables about how things are going, and then like put it together in graphs and all of that.
Dan Klyn: Yeah. Cool.
Sarah Barrett: so it's been my recent thing, it's kind of fun, [laughs] to get into. yeah. So like less IA more making the case, looking at impact, and a lot more like management, how we do our work, like the, the work is the working kind of stuff.
Dan Klyn: Yeah. Yep. And then, how about the things you used to do more of,
Sarah Barrett: I haven't designed a page layout in like five years, [laughs] so the things that were kind of more UXey. There are people who are so much more talented at that than I am, that I just absolutely love for them to do it. th- eh, how anything looks on the site ...
Yeah.
... is, is very far out of my wheelhouse by now.
Dan Klyn: And then, one last question before we open it up, which is, creative practice, do you have something that feeds your creativity, and your imagination, like painting, or dancing, or basket weaving, some, some kind of a ...
Sarah Barrett: [laughs].
Dan Klyn: ... some kind of a thing we might think about, if we're, we're, our batteries are not charged, and we want to be able to do good work like this, do you have a ...
...
... secret source?
Sarah Barrett: [laughs] I worked as a seamstress for about 10 years, when I was in college, in grad school, I worked in costume shops making things, and, I cannot overestimate the importance of making things with your hands.
Yeah.
whatever it, it doesn't have to be clothes or whatever, that's what I do, but s- I also like knitting, cooking, all the domestic stuffs that dudes usually look down on is extremely worth doing very centering, the ability to do something well and create something in the world that was not there before, like ...
Dan Klyn: Yeah.
Sarah Barrett: ... it is a real thing, it works. you can tell whether it's good or not, that changes absolutely everything.
Dan Klyn: That's awesome. Do you, ah, okay, next time that I talk with you, that I get to talk with you I'm gonna ask you about the costume making thing, 'cause that, that seems-
Sarah Barrett: Sure.
Dan Klyn: ... really, interesting. So, how about, folks here, and I know Ivan, I'm not sure I know any of the rest of these folks, hey Ivan? Anybody have a, question, or an objection, or, some serious thing that you want to talk about? I think you should be able to unmic yourselves.
Ivan: Yes, yes, yes, yes, me. hear me?
Sarah Barrett: Yes.
Dan Klyn: Yes.
Ivan: Good, pretty good. Good. Yeah, I ... well, you work for Microsoft, but I might well, my job as a ... unfortunately I was laid off this week, but,
Dan Klyn: I'm so sorry.
Ivan: I know, I know, I know, it was a shock, but, but with stuff happening I don't ... but anyway, I used to work with a company that dealt with, congressional records, essentially, and I think, and you, I wonder in that, you know, you know, in terms of information, you know, part of the problem is we have these different data sources from all these places, and they're in all formats, I mean as a UX developer I had to go through, you know, sometimes, some stuff was in HTML, was in XML, and I do, [laughs] oh yeah, yeah, you feel my pain.
Sarah Barrett: Oh no.
Ivan: [laughs].
Sarah Barrett: Certainly.
Ivan: Yeah [laughs]. Yeah, so, you know, go through that, so, you know, I wonder, w- ha- did you have to go through w- that working with Microsoft? Because obviously they have, you know, they have history of supporting older products, and, yeah.
Sarah Barrett: Yeah. one of the good and bad things about Microsoft is that I think they approach everything as an engineering problem ...
Ivan: Okay.
Sarah Barrett: ... and so sometimes it's very frustrating [laughs] and sometimes it's great, so I think last year we completed the migration of MSDN to our platform ...
Ivan: Okay.
Sarah Barrett: ... so like I don't know if you guys remember MSDN, the Microsoft.
Ivan: Yeah.
Dan Klyn: I remember that, the s-
Sarah Barrett: Yeah, like a development network. it all still exists, it ... like back all the content that used to like get shipped to people on CDs, we had to move that on to our site.
Ivan: Wow!
Sarah Barrett: [laughs] so that was completed last year, and that was mostly done automatically-ish, [laughs] through some engineering work, but like, yeah that was a huge thing. So we have like all of this concept that came up from that, I had a really fascinating meeting, a couple weeks ago with the guy who manages all of our open specifications content, and was like, "What is this, and why is it this weird format that fits with nothing else on the site?" And it's because it's all the stuff that was mandated that Microsoft keep in a certain way in the '90s from the Elbe antitrust lawsuits ...
Ivan: Ah.
Sarah Barrett: ... and that kind of thing.
so like that stuff has to be on the site, the numbers have to be formatted in this way, [laughs] like it has to be exactly. So, yeah, like the, the diversity of the ways this thing has to be, is kind of wild.
For better or for worse, so Docs is all open sourced on Gab, they're very, very few repos, I think they're, I don't, actually don't think any of it's private, some of it is not, you can't make contributions, you know, pub-
Ivan: Okay.
Sarah Barrett: ... public contributions, but it is all open source, and every single document on Docs is a markdown file in a folder.
Ivan: Okay.
Sarah Barrett: That's it.
Ivan: Yeah.
Sarah Barrett: And so the developers have done a huge amount of work in getting, smooshing everything into a markdown file basically. there are still the after-effects of that having happened like ...
Ivan: Yeah.
Sarah Barrett: ... things don't have to be another format to be another format, if that makes sense? [laughs].
Ivan: Yeah, that all ... it makes sense, it makes sense. I mean there's this, it's, it's a, it's, it's once, it's a information problem because it's sort of like, you have this in f- you have something in this format, and something in this format, and you have to sort of map it, and obviously it's a headache and problems, especially not just for most probably techs, but for EBs, and packet developers, and it, yeah.
Sarah Barrett: Yeah. One of the reasons I think that we, like my team really bonded with our development team, is that, neither of us have the, like the luxury of just having exceptions.
Ivan: Okay.
Sarah Barrett: like that's one of f- like, like that's actually not one of our, our principles maybe it should be, is that we-
Dan Klyn: Well, I was gonna say that not one of your principle.
Sarah Barrett: ... invent something.
Yeah.
Yeah. Like we, if we do something it's got to work for everything, it can't work for like less this stuff, and then not that [laughs]. so, and like that's also how a technical system has to work, you don't just get to like have these things like, "Oh yeah, it breaks every time you find one of those, but we don't have those very often," whereas maybe this is unfair, but I think in design, and even for a lot of PMs there's a little bit over, over in there.
Ivan: Mm-hmm [affirmative].
Sarah Barrett: and so, so yeah, we can work up against that too.
Dan Klyn: When I heard Iv- Ivan's question I was thinking about what you'd said earlier Sarah about object-oriented UX, and-
Ivan: Oh.
Dan Klyn: ... thinking about the heterogeneity of whatever the hell is coming at us from the world, and that is the order that architects and designers like us make ought to be able to catch some new thing and like, "Yeah, I know where to put that, it goes here," is, is working at a more abstracted level of objects rather than file types, is that something that, that you, that you do, or? I'm curious how, how detailed your modeling world is, and, and how that rubs up against the orthogonal, "Here's a whole bunch of new Docs in a new form- format, how do you deal with them."?
Sarah Barrett: Are you asking me or Ivan?
Dan Klyn: I think I'm asking both of you [laughs].
Sarah Barrett: Okay [laughs].
Ivan: Excuse. Oh shoot.
...
Okay go ahead.
Sarah Barrett: I'll give you a second Ivan. For us, we kind of did that big content modeling exercise about this time last year, and are working at implementing it, and I remember when Sophia taught it, she was like, if you had six content types, or like content objects, like you probably had too many, it's getting too big. I think at last count we had like 32 ...
Ivan: Oh!
Sarah Barrett: ... or something like that, which is too many, but like ...
Ivan: [laughs].
Sarah Barrett: ... the- there's, the site does so many different things, so I was like, "All right." I actually had a ... we do, cares, we're just starting it, and so I have a carer to keep the concept model under 35 objects, [laughs] like for this, this quarter, and then like as we ship new things next quarter we'll see, but like it's huge, a lot of sites put together, I don't know.
but, but yeah keeping it at that means to, doesn't mean that we can say like, "Okay, these things they're different on the backend, they're different for authors, I'm gonna try to make them as same as possible for our users." and we can try to keep a little bit, a little bit there.
Dan Klyn: Yeah. How about for you Ivan, did you have to make abstractions of content types, and of, things in the world so that when new th- new, requirements-
Ivan: Yeah.
Dan Klyn: ... come around that, that your whole sense of order doesn't fall apart?
Ivan: Yeah. Well, I work as a UX developer, so I was so at the front-end of it, I, but as a member of the UX team I got to be at least speculative to all those arguments, and, and it, and yeah, I mean for what I did, ah, the clients, ah, I'm trying to describe it, ah, because we're dealing with, information, government information, like former scripts and, and bills of staff, I mean, our, our, I'm trying to think how to describe it. It, it, it was, it was a much, it was I wanna say structured, it was, we had a si- we already ha- w- we already had a s- a like, ...
Sarah Barrett: Bigger things.
Ivan: Yeah.
[laughs].
Ivan: Yeah.
Yeah, absolutely.
Ivan: Yeah. I mean, you know, we unlike you we ha- we did, we had personas, and we had different, you know, we had the, let's say, we had the lobbyist, we had the, CEO, we had the, I don't know how to say it, legal aide, that sort of thing. So it was a, I wanna say it was struct- I mean we could, you know, relate that, weigh that information through a much more structured way, but one of our problems was, was that, you know, again we had all these bits of information, and we had to sort of work it how each wou- would sort of fit into this little, to these certain, I will say boxes, but certain things.
And another problem that we were having is that, we, we were just getting, at the time, time we were just, most our clients were in the United States, but we were also having foreign clients, which ... 'cause that's even a, a bigger problem because now you have to deal with, and this is a problem not all like left can lay back, I was just released from but our previous company well occasionally that was bought over by, by, a foreign, foreign insurance agency where he got to deal with different, not just information, but also different ways of dealing with information, like, you know, like well, like insurance different companies have ways of, of le- you know, the legal, requirements for different countries regarding insurance are just different, you know? Different requirements, different things, so, yeah. It's, it's, [laughs] it can be messy, but it's, it is sort of, you have to go through certain ho- hoops to sort of map everything out.
Dan Klyn: Sarah, do you all support multiple languages with Microsoft Docs, and if so how many, and, I'm curious about the testing across languages.
Sarah Barrett: oh God I'm gonna get this wrong, we localized into like 32 languages or something, but it varies depending on the content set, because different people pay for a low, and there's different, and there's also like tears of human translation versus machine translation fallbacks and all kinds of stuff like that that I don't really understand because our look team's really good, and they kind of just ...
[laughs].
... but it is, is an issue. We also run into some things where almost all of our content is authored, first into primaril- is authored first in English, like En-Us and then, localized, but there's a small amount of content that's existent only in Chinese because there's like an Azure China that only you only get in China, and so the- there's some weird things like that, that I'm like just starting to dip into, because, that is a well-organized machine [laughs].
in terms of our testing, when we do user research, we were, user research is new to our organization in a lot of ways, but it's something that we're trying to really build up, my colleague Rachel Price has been doing, amazing work in that, and we try to, get as diverse of the user set as we can, because we do have users all over the world. it has a big impact for other things we do on the site, like, I remember a talk from our engineering manager talking about how our average download speed in the US is under a second for a page or something like that, which is great, but in India it's like seven seconds ...
Yeah.
... and we had a huge number of developers, or, you know, users in India, so like we have to think about how we, how we design the sites so people all over can actually use it properly, ...
Acting and not leaving anybody behind in a non ...
Yeah.
... exclusionary way of operating.
Yeah. we try to do that kind of thing for accessibility too, like we've actually gotten the opportunity to do user studies with, like low vision developers, finding low vision developers and that kind of thing, which is, which is great, you don't usually get the funding to do a whole study for that, and that's really informed me a lot about how I think about accessibility in IA, and, i- that's, it's a, it's a big priority for the team, which is a lot of fun.
Dan Klyn: That's awesome. Joey, you have your hand up, what would you like to, say, or ask, or object to?
Joey: [laughs] I wouldn't like to object to anything. first of all, thank you so much both Dan and Sarah for being so generous with your time, it's been really useful to me, I, I'm very new to the, you know, field of like IA just as a kind of UX designer, and t- team leader, so it's every single bit I can get is so useful.
I've got a couple of questions, so I'm gonna ask one and then we can circle around if it comes back to it. so Sarah, you mentioned two years of, like advocacy for your team, and building up trust, and I was wondering if you could, share some of your experiences, what worked well, and, [laughs] what didn't work so well.
Sarah Barrett: Yeah. a lot of that was just trying to, yeah, so it's ... I mean it was really about like relationship building, and, just trying to be as helpful as I could, okay, so a few things, so fai- the thing that Dan and I were talking about at the beginning of finding people who do this work, regardless of what their title is, that was a really big deal, and the big things I did for that were, I ran a lot of working groups, that honestly largely didn't actually accomplish a huge amount, but people who wanted to be in them were really good allies, [laughs] and then I like, I go- I had those people, and we could like get something done, or didn't, but then I had the people, and then we could like get better things done later.
I also do my office hours, so for two hours every Thursday, at first it was just me and now it's the three of us, we all dial into a meeting, and there's not a, an agenda, anybody who wants to conjoin and ask us anything, and just like bring us something to look at, that's one of the, ways to just kind of do some general education. There are a few people who dial in most weeks, so just kind of listen in, which is great, you're muted Dan, I see you're, think you're talking, [laughs] ...
Dan Klyn: I just said wow, I'm just saying wow so much I'm muting myself just to cut off ...
Sarah Barrett: [laughs].
... I mute too.
so that's really fun, people was like kind of talk about it, it was also really useful at the beginning, especially, because, it's a complicated technical platform that works very idiosyncratically and so people would be like, "Hey, like how do I solve this problem?" And be like, "God, I don't know." Like, "I have no idea how you would do this." And then somebody who'd been there for a lot longer would be like, would come on and explain technically how it work, and that gives them like general theoretical advice, and they'd be there you feel like, they really did a lot to kind of build the, build the feelings about it, build the relationships.
we did a lot of ... we also focused a lot for the first like 6-12 months probably on like really small incremental improvements that didn't make anybody feel bad about where the product was, because there was a lot to be done. so it was a lot of just going to meetings being like, "Oh, you know, all of the, all of the things in that header could ..." or like, "In that navigation could actually be links." Just an idea, if we wanted, [laughs] or, you know, "Hey, we could have breadcrumbs on most pages." [laughs] really going for the low-hanging fruit, which is not my, not my natural way of doing things.
[laughs] and if you kind of do enough education like that, eventually people are like, "Oh, you know, breadcrumbs should be on every page when you're not in a meeting." And you're like, "Oh my God, they've gotten it." [laughs] so focusing on the really, really small wins, and then, th- relationships. I think tho- those things help to build relationships, but like making yourself helpful rather than a barrier, is a huge part of it.
some people who are doing IA kind of work in other parts of the org were like, "I'm gonna set up, like there are gonna be a workflow, and I'm gonna get the leadership to sign off, so that whenever anybody wants to publish something, like I have to review it first." And I'm like, "One, that sounds like a terrible job once you do that, [laughs] and two everybody's gonna resent you and try to find ways to get around you, because you can't make anyone do anything ever." Like you can't make anybody do anything, no matter what.
so if you're helpful, and you can make their lives easier, you will be a, an asset, and, if, if the managers or various people are hearing like, "Well, you know, Sarah said like ..." From different parts, like it becomes real. in a very slight tactical way, one of my managers tried to do was talking about the IA team, even when it was only me for like a year and a half.
Dan Klyn: Oh, that's brilliant.
Sarah Barrett: and now there is an IA team [laughs] but I was like, "Yeah, you know, the, the, the IA team first really doesn't have the capacity." [laughs] so that all helped a lot, specific examples, one of my big things with developers is like when I go ... I d- this is actually from when I was a consultant and I didn't develop this here, but like going and being like, "Listen, I do my job right, you will never decide what goes into their drop-down again, like, those will all be done for you already, you will never have to do that again." [laughs] And they're like, "Hallelujah, thank you. Yes, let's go."
Dan Klyn: "You're the lady that promised that thing, aren't you?" [laughs].
Sarah Barrett: Yeah [laughs]. And I ... actually I need a team because now we can't do it, so yeah, that's one of my strategy, I don't know if that's helpful.
Joey: No, no that's great, thank you.
Dan Klyn: Well, what was your other question? You said you had more than one.
Joey: I just wanted to give other people a chance, I didn't wanna [laughs] like, monopolize.
Dan Klyn: Oh, it's just, this is just getting to see her, it's fine.
Joey: It's kind of like more of a comment than a question type of guy, but, if your team, has like enough buy-in to make meaningful changes, in the information architecture, or as, or no one else in the organization is that invested, or attached to any of the existing structures for best practices, what would be like, your recommendations for those kind of like low-hanging fruits of what we should attack first? I mean just for some context, like I work at The Economist, I lead the UX team there, and we've made a ton of changes to like the header and the navigation structure, it's even like impacting like content strategy, and there has been like extremely like minor pushback. So, I, I feel like we've got a kind of open right here, so I want to know ...
Sarah Barrett: Ah.
Joey: ... where you would start first [laughs].
Sarah Barrett: yeah. For me, this is another Sophia Voychehovski thing, but I think she drew a little like, a feedback loop between the content model, the navigation scenarios, I mean, I don't think she put taxonomy in there, but I would put taxonomy and metadata in there, and like looking at each one of those, and then seeing like, "Okay I'm gonna tweak something here, what does that mean for the next thing? What does that mean the next thing?"
So like nobody's fighting about navigation, that's amazing, 'cause usually it's incredibly political, even though I don't actually think it's that impactful, that's just my opinion, but [laughs] I don't know, you can look at your own analytics, but nobody clicks on ours, nobody clicks on a lot of versus navigation.
but, really solidifying that content model and figuring out what the things you have are in your system, and then making that make more sense with your scenarios, like those user flows, making sure people can do the things they need, do you have the metadata you need to ... is that reflecting in navigation kind of just like erating it o- out that way, that's usually how I go about it, I don't know if that's relevant for you guys though.
Joey: no it is. I mean, that's exactly the path we've been going into, like adding additional metadata to our articles, and different content types, so that we can actually understand the journeys there, so.
Sarah Barrett: Yeah. I mean, I think the other side of it is thinking about the, the, the maintenance, the production, the, how we've, how all these things get done over time, if you have an open road like that, sometimes it's really tempting to, over-design things a little bit, where it's like I can put all of the metadata in the world on these things, and then we'll know everything about it, [laughs] and like we'll never have to do anything again, and like that never works. and, and so like figuring out what the, the l-
and so like, figuring out what the- the least you can do at each moment, that actually gets [inaudible 01:21:11] is pretty good. I don't know. A very good, lesson or, I don't know, story from our technical architect, who I was talking, who I like so much, he was like okay, the first time you design anything, you design like a little crappy... or, like, as a- as a young architect, like you design this little crappy thing, and it doesn't work very well, and it breaks, and [inaudible 01:21:31], but it's good and whatever.
And, the second time you design, you like discover design patterns, and you like discovered that there's this larger world outside of it, you're like I will code everything once, and I will build the- the architecture to end all architectures, and it will do everything perfectly, and then I will never have to code again, and it breaks immediately.
You're like okay, not that. [laughs] And then you go back, and like the truth is somewhere in the middle, and I think that's- that's what I've found, too. I definitely did like the key to all mythologies on one of my projects, as a- as a younger IA, and I still have like, oh, god, I messed that up, feelings about that. But, it was a useful learning experience, I guess.
Joey: That's really helpful. Thank you.
Dan Klyn: Other attendees, if you've got a question, put your hand up, or, I might just call on you, but I'll take a little crack at, at Joey's question as well, which is, the places where simplification is happening. I- I feel like whenever simplification is happening, value is being lost, and that, making the complex clearer is the alternative to making the complicated simple.
And, so, and- and, this is more of a prospector, point of view, maybe than an inside of an organization, but as a external consultant and trying to find, you know, where's the... where's the seam in- in what's going on here, we often talk to people at organizations who are like yeah, IA, you've made the sale, but this organization has not spent one dollar on any called information architecture before. So, you have to find something to work on with us that's going to demonstrate value that doesn't cost a million dollars that doesn't boil the ocean.
And, so, that's one of the things that I think about is, how, you know, what is the value proposition of information architecture? For me, it's making the complex clear, and why is that valuable? Because complexity is inherently valuable, and- and the value of complexity is lost through simplification. So, wherever-
Sarah Barrett: That's a hard sale, man.
Dan Klyn: Yeah, yeah. No, because they like it simple. And- and, to tell people... you can't tell people they can't have simple, but it's shifting that to like no, there's an aesthetic of simplicity, of sensical, that easy to use, yes, you get that, but there's some process that is dumbed down. There's some way that information is moving through the organization that somebody did because it was easy, not because it was good, and, finding those corner cuts, maybe it's the opposite of the UX thing about no, that's where you put the path is where the people cut through. It's like no, look for where the people cut through. Somebody who's making cement decisions is doing it wrong. for whatever that-
Sarah Barrett: I don't know if this is the same thing or exactly the opposite, but, just like in looking at- when you're inside, looking at places where people have workarounds, that might be kind of a... I don't know, but like what do people have in their One Notes? What are they saving in there? what are they saving in their email folders? What- what, what artifacts are they creating so they can hand to somebody else when they start their job? that's often a place where you can see a lot of, a lot of those seams.
Dan Klyn: Yeah. No, I think we're saying the same thing. It's- it's- sometimes, in contextual observation, it's a piece of paper they have on their wall that's like oh, my god, if- if this thing just made that tangible all the time to people, oh, it'd be great, and, only Susie has that piece of paper, and nobody else does. That kind of-
Sarah Barrett: Yeah. I think it's elsewhere [inaudible 01:25:25] a really good thing about how, complexity isn't the problem, ambiguity is, and simplicity isn't the opposite of ambiguity. Clarity is.
Dan Klyn: Uh-huh [affirmative].
Sarah Barrett: I've been using that because, the people I work have a very strong desire to make things simple, and it's not. It is- it is a complex, complicated space and it is... we're designing for people to do very complicated things, but we have also a lot of ambiguity, and we can solve that.
Dan Klyn: Amen. So say we all. [laughs] Other, [Morgaine 01:26:01]? [Vartia 01:26:01]? MB Phone? anybody else want a crack at, Sara or me before we, have a benediction?
Mitchell: Yeah, for sure. Hi. I'm MB Phone, apparently. I'm actually, my name is Mitchell, I'm a designer at IBM. essentially trying to, I guess you can say, address some of the main concerns of logic across every, product at IBM, indirectly. my question-
Sarah Barrett: Hello, friend.
Mitchell: [laughs] yeah, right.
Yeah, you all already are acquainted, it sounds like to me.
Mitchell: [laughs] I- my question would be more like, how do you strategize in trying to- to make these large structural changes, to the infrastructure of a single or multiple products, being that there are other priorities on the roadmap that might have already been placed before you realize this IA issue was an issue.
Sarah Barrett: Mm-hmm [affirmative]. okay. I think there are a couple of things at work here. One is that, IA is never going to be a priority. In- in my organization, probably in yours, I don't know. That's the thing that I had to come to terms with is that like it is never going to be anybody's priority, so given that, what are we going to do? And, so, for me, it was really about identifying where, a lack of structure or lack of clarity or something was preventing the things everybody did want to have happen from happening.
because what I was seeing a lot of when I came into the org was it was really hard to ship certain things, or there were certain kind of experience- there were certain parts of the experience that just didn't work, and nobody else can tell why, and I immediately could tell like oh, we don't actually have any metadata. That's usually what people do to make this kind of thing, or whatever. Like, that's a really simple- a lot of things like that.
It's like oh, we've taken no control over our [inaudible 01:28:04] navigation. We would be able to do this thing if everybody... so like see what those problems are. Sometimes, you have to let those things fail before people will go along with it, and I don't mean let them fail as in like stand back and be like I could have prevented that, but like, you know, you make the case, you try to convince people, it doesn't work, you let it not work, you come in and try to solve it.
and then, the other big thing, I think, is, really what we're trying to solve like with those strategies and that kind of thing where like have a big idea for where you're going and like what the huge... what the- what the big goal is. Write it down. Always have everything written down. and then make it easier for everybody to go in that direction than in any other.
so, like I do a huge- or, we all. It's not just me anymore. Like, we do a huge amount of prep work where like hey, if you wanna go in the direction of greater clarity, you're gonna have a taxonomy. You're gonna have a strategy doc written for you. You're gonna have all of these things, and like I will do all of this for you that will make it simpler. If you're gonna make it harder, gosh, I'll try, but like if you're gonna get- if you're gonna make it more ambiguous, if you're gonna go against the concept [inaudible 01:29:11] design, I'm like it's so much harder, and it's not gonna make anybody else's life easier, so like you have to really pave those paths for people. I don't know if that helps.
Mitchell: Great. Thank you. Yeah, definitely.
Sarah Barrett: have you read Switch? Behavior Change When Change is Hard, or something like that?
Mitchell: No, but I'm going to.
Sarah Barrett: Okay. It's by Chip and Dan Heath, who I believe are, brothers, which is funny to me, but, it's... Christina Wodtke turned me onto it. It's like her favorite book, and it is all about how you can make changes happen, either like for yourself or in big organizations, and they identify like kind of three... they use a metaphor of like, a- a very small rider on a huge elephant following a path, and you can act on like the rider, which is where everybody kinda thinks that they want to do, and like our- our higher selves with a bunch of aspirations, the elephant, which is what everybody want's to do when they are lazy, or, you know, just like the- the urges about what is kind of pleasurable or easy or the path itself, where you're- the way you're going.
And, so you can act on any of the three of those or all three of them simultaneously to get people to do things differently. So, like, you can inspire people. You can make it easier to do the thing that you want rather than anything else. Like, you can work on that in all kinds of different axis. I'm not doing a very good job of explaining it, but I highly recommend it.
I would also really recommend that Kalil article, Policy Entrepreneurship in the White House, about how he recommends getting- getting things done when you have responsibility without authority. and it's a lot about finding your doers, writing things down, keeping an eye on the big picture, all kinds of stuff like that. Using other people's schedules for you, that's a really big one that's been really useful for us, too. all kinds of things like that.
Mitchell: Great. Thank you so much. I'm definitely gonna look all those up. Appreciate it.
Dan Klyn: anybody else in our participant panel like to ask Sara a question or, pose a, dilemma?
Speaker 5: this is probably bigger than, you have time to talk about, but you- you spoke about like expanding the audiences with what you're doing, and, especially if you said research is a kind of nascent thing on your team, how do you decide, prioritize where to expand and like what your opportunities there are?
Sarah Barrett: That's a really good question. let me try to think. One of... I- I feel like I have enough experience in this- I feel like not nearly as much as most people who are on Dan's podcast, so, wild case of imposter syndrome there, because I like did not found [inaudible 01:32:13]
Dan Klyn: Nope.
No. [inaudible 01:32:15] Sara. This is-
Sarah Barrett: yeah, well like when everybody else was finding adaptive path, I was like memorizing all the lyrics to Waiting For Tonight. which has served me well. but-
Dan Klyn: The World's Waiting For Tonight? [laughs]
Sarah Barrett: [laughs] but, like, I think that's one of the best things about having experience on, let's be honest, also like projects that didn't go so well. It's one of the thing- reasons I'm very grateful that I, had those years in consulting, because like, if we're being real, oftentimes, things do not go very well, and you spin them really hard, like you tap dance and make sure everybody's like oh, we got so much value out of this, but like they do not always.
Sometimes, it's your fault, sometimes it's not. Often it's not, but, like those times when it didn't go well, being able to be like oh, that's why the wheels came off, and so looking at that product roadmap, for like what- what the higher ups want to achieve with the product, with the site, whatever, where people think this thing is going, and being like that's- that's not gonna work. Like, I can see right there that that's gonna be where things, where things start- start falling apart, and making plans really early, like contingency plans on cutoff and contingency plans.
I had lunch with one of our designers who got hired, last summer, I think, right after I got back from maternity leave, and he was like yeah, you've been, because he was designing the persistent navigation. He was like yeah, you've been working on this for a while. I think I read one of your docs from like 2018, and I was like yes. I have been working on this for a very long time, but like, you think about what's gonna happen in the future, like, write everything down, and then try to plan for that big stuff.
That's probably not very helpful, because it's like know what's going to happen in the future, but like we organize our practice into three main things, concept model, navigation, and metadata. for me, that is what information architecture is. It can be lots of different things also, but like for us, in this case, those are the three things it is.
the person who does navigation also does a lot of interfacing with the rest of research and UX and that kind of thing, partially because that is our person's interest and skillset, but also because I think that's where you get the most into those kind of UX-ey parts of IA.
our- our metadata person also has some... that's Dana [inaudible 01:34:32], also has some expertise in, conflict strategy, but I think that also goes on, like, those- those meld together quite naturally for us.
And, so, when we look at expanding the team, we kind of do it in a fractal pattern where there are... there are three of us, there are three core disciplines. Each one of those three could become two, could become three themselves, and could build out into little teams. and, in the same way that we kind of iterate on how we think about the site with like oh, the concept model, the navigation, the- the user flows, the metadata. Like, you can iterate on how you build the team out that way.
I think relatively few organizations have some like massive need for one thing and not the other. that couldn't- that could not always be true. the one place that breaks down a little bit, I think, is in metadata, because you get to a certain size of organization, a certain size of content set, and it's just maintenance, like you're just have a certain number of metadata changes happening every week, and then somebody has to do those every week, or it all falls-
Dan Klyn: Oop. Looks like we've got a little freeze going on. Yes, I see a chat from Joey, and, Seattle is, delayed. Let's hope it's momentarily.
Speaker 4: Yeah, the internet in London isn't doing that great, either.
Sarah Barrett: Oh, I'm sorry. I think it might have been mine.
Dan Klyn: Welcome back.
Sarah Barrett: I apologize.
Dan Klyn: Apology refused. You had no way to control that, Sara. [laughs]
I've- I've learned so much. thank you for, everybody, for being here, especially Sara. I don't wanna belabor this if there aren't further, questions or objections or, what have you. I- I love the objections, the
Sarah Barrett: Yeah, you're really looking for one. [laughs]
Dan Klyn: I'm- I'm- I'm looking. So- so maybe somebody'll bring me a good one. Morgaine, or Morgan, depending on how we say your name, any objections before we, get back to- to lockdown and to isolation and social distancing? Come on. Help us out.
Morgaine: I just wanted to mention, because I had connected late there, but I've been enjoying listening to all the questions, so I really enjoyed everyone's questions. Just wanted to mention this, so thank you.
Dan Klyn: That is very kind. I was not fishing for a compliment, but, that is very [inaudible 01:37:11]. [laughs] Well, maybe I will ask Sara one more question while the rest of you figure out if you, if you're good or if you- you also have- have a last question.
I was born and raised in West Michigan, and we have West Michigan Nice, which is something that people, Herman Miller's is a global furniture manufacturer many people have heard of. They're headquartered here, and, we get to do some work with them, and I've talked with folks from... usually, it's people from the New York, New Jersey area who come out of a meeting like, goddamn it, why is everybody so nice? I have on idea where anybody sits relative to what that hell we were supposed to be talking about, and they're gonna stab me in the back. You know, what people really think, your friends stab you in the front, and these- these West Michigan nice people are gonna stab you in the back.
And, so- so, we knew about that. We- we've done a little bit of work in Seattle, and we were not prepared for the decision-making culture in Seattle, which is certainly not a people being nice when they aren't supposed to be nice, and we looked it up, and there's a thing called the Seattle Process, which, we- our- our facilitation with deciders really failed early on with the project with the Seattle-based company, and we were just like what the hell? We're used to being smart people who- who facilitate well, and sort of researching decision-making culture in Seattle, it- it's a thing that there is-
Sarah Barrett: I don't know anything about this.
Dan Klyn: That there is- yeah, that there is a... it's not like a circular firing squad, but there is a, a way of making decisions where people say stuff, they do stuff, then they need to stop doing that and talk about it, and then do something else, and that's a really, ham fisted, bad summary, but look it up on Wikipedia, Seattle Process.
So, you're a fish. Are you from Seattle? Because, I was gonna ask, do you- is this-
Sarah Barrett: No.
Dan Klyn: Seattle Process, does that?
Sarah Barrett: I have been here for, god, about 10 years.
Dan Klyn: Kay.
Sarah Barrett: maybe a little bit longer. I went to college in Portland, so I, I'm a kind of adopted Pacific Northwesterner. My husband's born and bred here, so like I think we're here, but I'm not from here. My family's all from L.A., and I grew up moving around a lot, and I went to high school, middle school and high school in Italy.
Dan Klyn: Oh, wow. Cool.
Sarah Barrett: so, like different, and then, came back to the U.S. for college and like had never been in a 7-11 by myself. and that kind of thing, so it was like a little bit of... little bit of culture shock.
let me... I'm gonna read this real quick. No strict definition. Refers to the pervasively slow process of dialogue, deliberation, participation and municipal introspection before making any decision and the time it takes to enact any policy.
That's very interesting. Okay. [
Dan Klyn: laughs]
Sarah Barrett: I don't know whether this is a Seattle thing, though your experiences suggest that it might be. I do think that there is... it's very easy to have nothing happen, and I think a lot of our, like just thinking about it now, a lot of the way that we work and like this- this process that I've defined for my team and that kind of thing might be designed to combat that.
the first big conference talk I ever did, which was not very good in retrospect, but like I keep coming back to, was basically about how, how you structure deliverables and engagement so that people actually understand and agree rather than just saying they agree, because then they will come around... they will come back and like no decision will ever stay made. And,
Dan Klyn: That's this. That's-
Sarah Barrett: Yeah, okay. the, the- the thing that I came up with for this, which I don't know if this is the right answer, the thing I came up with is making very small decisions incrementally so everybody understands why they were made, and you really have to go like smaller to deliverables more frequently so things really happen.
I think especially as a consultant, there's really the temptation to kind of like Don Draper your clients. I know that I really felt it where, like, you know, you get in there and you like really tap dance hard, and you're like, you know, you're oh, yeah, we're doing this, we're doing this, and it's gonna be so great. You know, that's great feedback, and the way we're gonna, like, you know, and you- you do that whole thing.
Dan Klyn: I've been that guy.
I'm trying- I'm trying not to be that guy, but I've been that guy.
Sarah Barrett: But, you have to, because like you're there, and your adrenaline's up, and you don't wanna... you can't... you can't do another change order, and you can't afford to revise this deliverable again, and like you've got to do it, and you get in the elevator-
Dan Klyn: We've gotta make you love this fucking thing, and accept-
Sarah Barrett: Yeah.
Dan Klyn: Yeah.
Sarah Barrett: Yeah, and you get in the elevator, and you're like we won, we did it, and like if you do that, it's all over. Like, if that's your feeling, you've got nothing done.
Dan Klyn: Yep.
Sarah Barrett: and so, like, it's largely about, I think, those small steps, the, consensus, like really- you have to really ferret out people to disagree. a big, I feel like my main, my main facilitation technique here is playing the know nothing. just being like oh, explain to me how that works. Okay. So, is it like this? Oh, no, right. Cool. And, I think that like-
Dan Klyn: Yeah, no- that is also, galaxy brain way to deal with Seattle Process is to have it come out of- of them, not out of you.
Sarah Barrett: Yeah, and this is one of the places where being a 20-something woman doing this was really valuable, because people love interrupting you and telling you you're wrong, and if you can just like use that, they'll tell you everything you need to know.
Dan Klyn: Oh, my god, that is a terrible acknowledgement of how the world often works, and what a beautiful way to flip that. That's great.
Sarah Barrett: And, I think the other thing about that is, like, so, playing the know nothing, eliciting all of that feedback from people, and then, when they tell you something, or like they make a decision or whatever, don't let it be general. Make it really concrete as soon as possible.
John Hodgman and his... Judge John Hodgman Podcast loves saying that specificity is the soul of narrative, so like do not let them tell you how it usually goes. Make them tell you how it happened last time. Like, make them tell you a story about it, or like you give a really specific example, because like [inaudible 01:43:58] here, we're gonna handle it this way. Okay, so like the next time this happens, it's gonna be this, and I'm gonna do this. Is that okay?
Dan Klyn: Oh, yes.
Sarah Barrett: that makes everybody... because, like, you really have to make people engage with things concretely in order for them to really think about it, because nobody ever wants to think about anything.
Dan Klyn: Yeah.
Sarah Barrett: so, yeah, that's my idea.
Dan Klyn: Well, thank god you do, because I'm- I'm- man, I'm- I've got- I've got a lot of different ways to think about, this work after talking with you for just a little while today. Thank you so much.
Sarah Barrett: Thank you.
Dan Klyn: Please also thank your child and husband and everybody else for- for letting us have, so much of your time today. It's really helped.
Sarah Barrett: It was very fun. Thank you so much.
Dan Klyn: Thank you. And, everybody else, if you have questions, Sara and I are easy to get at on the internet-
Sarah Barrett: Yeah.
Dan Klyn: And, if you send us a question, we will certainly answer it for you. So-
Sarah Barrett: Yeah. Find me on Twitter.
Dan Klyn: Thanks, everybody.
Sarah Barrett: All right. Bye.
Dan Klyn: See you next time.