computer-programming

Transcript

[MUSIC: Oceanside Drive, Ethereal Delusions, used with permission of the artist.]

ROBERTSON: Hey folks, just a quick note to let you know that we are trying out a new format for Engineering Out Loud. Instead of pairing two related stories in a longer podcast we are creating shorter podcasts that feature just one story. We’d like to hear from you about what you think. You can email us at engineeringoutloud@oregonstate.edu or send us a note to our Facebook account. Now, onto the show.

[MUSIC: The Ether Bunny, Eyes Closed Audio, used with permissions of a Creative Commons Attribution License]

NARRATOR: From the College of Engineering at Oregon State University, this is Engineering out Loud.

ROBERTSON: When you think of engineering do you think of human and environmental health? Maybe not so much. So, for this season we are focusing on stories of how research in engineering at Oregon State can impact broad areas such as cancer treatment, food contamination and the detection of nuclear weapons tests. We are going to start with the topic of 3D modeling and learn how it can advance science in ways that you might not have imagined.

[AUDIO CLIP: from Toy Story]

Video games and movies are what we conjure up when we think of 3D modeling. But it’s also a tool that can help medical professional better target cancerous tumors. Today we’ll start by talking to Eugene Zhang, a professor of computer science at Oregon State University and an expert in computer graphics and data visualization. And, of course, I start with the most important question first.

ROBERTSON: As far as a good example of animation, what is your favorite?

EUGENE ZHANG: Well my favorite is the Toy Stories, as well as Star Wars, I like Star Wars because, you know, my kids like them so I started to really get into them as well.

ROBERTSON: The Star Wars that Eugene is talking about is the animated TV series The Clone Wars from Lucasfilm Animation that ran from 2008 to 2014.

[MUSIC: Oceanside Drive, Ethereal Delusions, used with permission of the artist.]

One of the planets in this show, called Mustafar, is a volcanic planet that has constant eruptions flowing into rivers of lava. You’ll see why I mention that in a minute. I asked Eugene what has changed to take us from very rudimentary 3D wire-frame graphics, such as the depiction of the Death Star plans in the first Star Wars movie in 1977, to the very sophisticated 3D images you see today.

E. ZHANG: Computer animation has matured as a field of research over the years and now we are talking about technologies that have been developed in the last 10-20 years with a strong focus on numerical simulation. Like simulation of fluids. Flows, like the lava flows you see in Star Wars. There have been a lot of techniques to speed up the simulation that would have been impossible before the GPU era.

ROBERTSON: The GPU era which stands for graphics processing unit that Eugene mentions had a breakthrough in 1995 when the first 3D add-in cards came out on the market.

E. ZHANG: With the increased computation capability we are now able to produce a movie at a much faster rate than say 30-40 years ago, for instance it might take 6 months to make another episode for Star Wars.

ROBERTSON: So, how is what you do with 3D modeling related to what people see in the movies?

E. ZHANG: So, one of the things that a lot of people do not necessarily realize is that even day one computer graphics animation, 3D modeling has been an integral part of that. People see these fascinating shapes, motions, but there has to be a way for them to be represented in the computer so that the artist can manipulate these shapes make them move, make them change the form and the more efficient the representation, the easier it is for the artist to work with them. And also for scientific simulation like the simulation of lava it involves very sophisticated mathematical modeling techniques that require very well-designed meshes.

ROBERTSON: Okay, so I’m going to jump in here and explain that a mesh made up of points in a 3D space called vertices, there are lines that connect these points, called edges. And the face is the area between the edges.

E. ZHANG: So, in fact if you think about a simple case. Let’s say you are talking about a cube. A cube has 8 corners, and these corners will be the so called vertices and there are 12 edges separating the 6 faces of the cube. So the surface of the cube consists of 6 faces would be exactly a mesh. And another example is to look at Spiderman. Spiderman has this very interesting sort of mesh like pattern on the cloth. You can see points and lines and these lines intersect at a right angle and they form these rectangular patterns on the cloth of the Spiderman. That’s a simple version of a mesh.

ROBERTSON: The reason we are spending some time with meshes here is that from what I understand from Eugene, the mesh is really, really important in 3D modeling. The mesh is the mathematical bones onto which everything else is applied. For applications like animations or art or architecture a good mesh will allow you to create more realistic images. But, perhaps, more importantly when 3D modeling is used for scientific research such as simulations of tornados, earthquakes and tsunamis the results will be more accurate.

E. ZHANG: In fact, it's known in the community that 90 percent of the time that simulation researchers spend on performing a simulation was spent on generating a really good mesh. And only 10 percent of the time was to actually run the simulation.

ROBERTSON: So, since Eugene works on perfecting these meshes, his work is fundamental to all 3D modeling and can be applied to any field. Even the modeling of complex internal organs. My next guest will explain why she and Eugene would want to do that.

YUE ZHANG: My name is Yue Zhang. My research area is in numerical simulations and scientific visualizations.

ROBERTSON: So, Yue’s background is in mathematics, and along the way she discovered that what she really enjoys is collaborating with others to find mathematical solutions to real world problems.

Y. ZHANG: Looking at fast algorithms and also complex mathematical modeling is very interesting, very challenging. On the other hand the real-life problems can enrich these models even further, because in real life there are always factors that need to be considered in the mathematical modeling.

ROBERTSON: So, that’s why she took the time to attend a mixer between researchers from Oregon State University and Oregon Health & Science University. The connections she and Eugene made there eventually led to a collaboration with Dr. Wolfram Laub in the Department of Radiation Medicine. The problem that Laub wanted help with is better targeting radiation therapy for tumors associated with prostate cancer. Yue explains why this is important.

Y. ZHANG: The difficulty is from the clinical side there is a safety margin because the radiation has toxicity and you can hurt the neighboring healthy organs. The larger the safety margin, the more harmful it is for the neighboring organs and to reduce this safety margin so that only the tumor portion is treated with the right amount of dosage is what we are trying to help with.

ROBERTSON: Next Yue tells us why targeting the tumor at the right location can be trickier than you might think.

Y. ZHANG: Because the organs are actually moving, so the organ shapes and positions are very important, are critical because the patients can have some movement the organs and the geometry can change.

ROBERTSON: So, people might not understand that, so why ... usually we think of our organs as fairly stable so why are they moving?

Y. ZHANG: Breathing, breathing and other bodily functions.

ROBERTSON: Think about that cup of coffee you had this morning. What’s going to happen to during a radiation therapy treatment that could take hours. Your bladder is going to fill up. And if there is a tumor next to it, the location of the tumor will change as the bladder pushes it out of the way. And so, to create a simulation of how the organs might move and change the location of the tumor, they first have to start with a 3D model constructed from the medical scans.

Y. ZHANG: The scans are 2D and they are taken at different slices on the human body so that we can construct a volume with a volume then we put mesh on this volume. Like what Eugene was saying we put node, edges and cells and faces on this volume. That now we have…

[record scratch]

ROBERTSON: Okay, I’m going to stop the tape there a second to bring in a conversation with Eugene where he talks about adding volume to the mesh.

E. ZHANG: We have an additional element called the cells, for instance in this case I would go back to the cube example, in addition to the 8 corners, 12 edges and 6 faces you can also think of the interior of the cube being a cell and if you put a number of cubes together you would get what is called a hexahedral mesh which means the mesh is made mostly of cubes, and in that case the number of cubes would also be an indicator how complicated the mesh is. For instance, for the simulation that Yue deals with there are usually millions of cells that were involved to generate a realistic simulation.

ROBERTSON: And now back to Yue

Y. ZHANG: Now we have tools to describe material properties on these cells, so if it’s an organ that doesn’t move much it has a little bit rigid than we describe one material property, but if it is something like the bladder, it’s very flexible, stretches a lot we use a different property to describe it.

[MUSIC: Oceanside Drive, Ethereal Delusions, used with permission of the artist.]

ROBERTSON: Now this is getting way more complex that a cool 3D animation. And it brings in another specialty of Eugene’s, called field processing which they are using to add information about the material properties of the organs.

E. ZHANG: Field processing is a very new subfield of geometry processing. Instead of modeling a shape, now we are modeling things that are on the shape. It's one thing to model the shape of the earth, to model the mountains and the ocean and so on and so forth, but it's another to model the magnetic field or the global ocean current flows on the earth and these are vector fields on the surface that can provide a lot of insight into things such as the air stability, pollution and climate change. And tensors are an extension of vector fields whereas a vector field has a direction a tensor field has multiple directions. So a tensor field is more complicated than the vector field and they can be only described mathematically by a matrix. So, you can sort of see the additional complexity here.

ROBERTSON: Indeed, so now you have millions of cells with an additional layer of information about the material properties of the organs. So, you can see why it might take a couple weeks for the simulation to run. But what field processing allows them to do is find the critical points where there is uncertainty in the model that can indicate change. Yue describes why this important for treatment.

Y. ZHANG: By looking at the simulation results the doctor can see a little bit more about what could be the changes to the prostate through the treatment period. Currently the doctors from the clinical side they scan the patient on the first day, so they have an initial scan which is a MRI -- very detailed. Then at that point they develop a treatment plan which includes the direction and the dosage of the radiation. But that's a static method. It's not dynamic it's not adaptive, so using our simulation we hope the doctors can have some predictive knowledge of where the organs could be and why the organs could be at one have one shape during the treatment. In addition, we like to track the how the material is changing through the radiation period.

ROBERTSON: Specifically, they would like to see if the tumor is changing… hopefully shrinking. And if other organs are being affected by the treatment.

Y. ZHANG: What we are hoping to achieve is we will get adaptive treatment plan and individualized for each patient. No two patients are the same. Cancer development varies from patient to patient, their ages, their health conditions, their family histories, all different. What we are trying to do here that is novel is we want to include bio mechanical modeling the simulations we want to include the tensor visualization on the material stress tensors.

E. ZHANG: We believe that tensor field visualization and analysis is key to the medical applications that we are talking about here as well as many other application going back to earthquake, tsunami analysis. This is a new direction for the graphics community but I would want to go bigger than that say it's actually something that faces the whole scientific community -- is to look at ways of modeling everything including the shape and the materials properties on the object. However, it is very challenging as Yue has mentioned, the fact that we are not doing biopsies on people then we only have the scans so we are sort of limited to extracting information about material properties through geometry information, like pixels the color of pixels and there is a lot of guessing work, so I'm hoping that there will be other ways, but non-invasive still to help us in order to model the material properties like the tumors.

[MUSIC: Oceanside Drive, Ethereal Delusions, used with permission of the artist.]

ROBERTSON: 3D modeling is inherently cool, and know you may know a little bit more about how complex it is. In conclusion I asked Eugene what it is about this project that sparked his interest.

E. ZHANG: It's interesting to reflect why I'm very interested in the problem. I guess growing up I had always been sort of interested in abstract things like mathematics geometry. I have also been very interested in science, but I have always considered them separate and unconnected. And this project is one of the projects that I finally feel like the two sides of my interests have started to converge. Where we are doing mathematically motivated research but with real impact where we really want to help patients to survive, to overcome cancer. I'm really hopeful that the techniques we are developing or the tools we are building will useful not only to architects or artists in Disney or Pixar but also available to scientists, doctors that could actually save lives and overcome all these diseases including cancer, various forms of cancer, and AIDS.

ROBERTSON: Okay, my friends, that concludes our first episode on human and environmental health stay tuned for more episodes on how researchers in engineering are working to improve our lives. And remember we want your feedback on the new format for Engineering Out Loud. You can email us at engineeringoutloud@oregonstate.edu or send a message to our Facebook account.

[MUSIC: The Ether Bunny, Eyes Closed Audio, used with permissions of a Creative Commons Attribution License]

This episode was produced by me, Rachel Robertson, with additional editing by Miriah Reddington. Our intro music is The Ether Bunny by Eyes Closed Audio on SoundCloud and used with permission of a Creative Commons attribution license. The other music in this episode was Oceanside Drive by Ethereal Delusions (who is an Oregon State music production student) you can find him on Twitter @quixoticBPMoose. The sound effect was used with permission of Creative Commons. Links for additional materials can be found on our website. For more episodes visit engineeringoutloud.oregonstate.edu or subscribe by searching “Engineering Out Loud” on your favorite podcast app.

Transcript

[MUSIC: Sicut cervus by Giovanni Pierluigi da Palestrina, sung by OSU Chamber Choir]

ROBERTSON Ahhh…so nice. That is the sound of four Oregon State music students who are helping me introduce the topic of diversity. What, you ask, does this choral piece by Palestrina have to do with diversity? Let’s listen again.

[MUSIC: Sicut cervus by Giovanni Pierluigi da Palestrina, sung by OSU Chamber Choir]

One voice, two, three voices, four. Each additional voice adds a layer of complexity and depth — the separate melodic lines complementing each other, creating a beautiful polyphonic sound. It wouldn’t be the same if all the voices were the same.

[MUSIC: The Ether Bunny , by Eyes Closed Audio,  used with permissions of a Creative Commons license . ]

NARRATOR: From the College of Engineering at Oregon State University. This is Engineering Out Loud.

ROBERTSON: This season is on inclusivity in engineering. I think you might be surprised at some of our topics. We will be talking about the research here at Oregon State that is promoting inclusive design and accessibility, and our focus on creating a more equitable and diverse workforce.

[MUSIC: Sicut cervus by Giovanni Pierluigi da Palestrina, sung by OSU Chamber Choir]

To get some perspective on the importance of diversity in the workplace I talked to some people working in technology fields about why it is important for their companies. Here’s Brandon Greenley, an Oregon State alumnus who is now a general manager at Danaher Corporation, owned by Tektronix.

GREENLEY: Yeah there are so many angles I could take to answer that question because it has importance on so many different levels. But I would say from a product development standpoint ….because that's what I know and love the most as an engineer… but, when we develop products for our customers we are developing for a global environment and that global environment is very diverse in terms of their needs, their preferences and the things they are going to use our products to do.

ROBERTSON: Designing for a global market was mentioned by another alumnus, Nadia Payet, who is a software engineer for Google maps.

PAYET: A diverse workforce is a better work force. Diversity of backgrounds, diversity of opinions, of solutions that we can come up with. When I design something it's all in English because this is Google, programming is all in English but then I designed a product and then I'm like, ‘In French, this word …  I don't know how are going to translate this, but it's not going to fit on this tiny phone, so we have to find a different way.’ And oftentimes it has to be pretty creative to fit really long words. German is our canonical example of really long words. We are, like, ‘How would this look in German?’ And if you don't have people who think about these things, think about these problems ahead of time, then they will happen in the product. And it leads to some sub-optimal experience.

ROBERTSON: I think we can all relate to sub-optimal experiences when it comes to software. Which leads me to my next guest, Margaret Burnett, a professor of computer science at Oregon State, whose research is about creating optimal experience for everyone who is using software.

[MUSIC: Sicut cervus by Giovanni Pierluigi da Palestrina, sung by OSU Chamber Choir]

Can you give me a little bit of your background, so how you got into computer science?

BURNETT: Okay, I was actually a math major when I went to college at Miami of Ohio and I minored in computer science. At the time actually there was only one university in the country where you could major in computer science so that was a really long time ago. And I graduated and got a job at Proctor and Gamble as the first woman software developer that they had ever hired in the Proctor and Gamble Ivorydale division, so that was kind of a glass ceiling that I helped to break.

[glass shattering]

ROBERTSON: That wasn’t the only glass ceiling Margaret has broken. After working in industry for a while she went back to get her Ph.D. from the University of Kansas and became the second woman to graduate from that university with a degree in computer science. She and Cherri Pancake were the first women to be hired as tenure-track professors in computer science at Oregon State. And just last spring she became the first female faculty member in the College of Engineering to be named a distinguished professor, which is the highest rank the university can bestow.

[glass shattering]

Wow, quite a few glass ceilings! But that’s not what we are going to focus on today. Margaret does research to improve the gender inclusiveness of software. But first we’ll start more generally with what her research area is about. 

BURNETT: Generally I am interested in where people and computers come together, especially when the people are trying to problem solve with the help of the computer. So, examples are trying to make their budgets balance, or trying to make some sort of business decision with the help of some kind of software, or debugging. Anything that has to do with problem solving. And a lot of that turns out to be software development because that’s a heavy problem solving activity, but not all of it. So it's kind of a blend of what we call human computer interaction in the computer science field, and programming language research and software engineering.

ROBERTSON: So, in this area of human computer interaction what you have been focusing on lately are gender inclusiveness issues, and so can you tell me a little bit about that?

BURNETT: Yes, the gender inclusiveness issues that I'm looking at are actually gender inclusiveness of software itself. Lately, I guess, there has been a lot of news in the popular press about things like bias inside software. And so that's one of the things that I'm interested in too and that's where this gender inclusiveness comes up in my research is whether software itself is — especially software that is used for problem solving — whether, in fact, that has gender biases in it and whether those gender biases really matter. Executive summary: yes, they matter.

ROBERTSON: So, is your past related to any way to why you chose to work on gender inclusiveness?

BURNETT: Well, no, not really. You would think it might have been, but actually it wasn't. I was working on this blend of human computer interaction and software engineering that I was telling you about relating to people problem solving with computers, but I didn't have a gender inclusiveness angle to it.  And then at some point one of my PhD students was getting ready to choose a topic, and she and I started brainstorming about something that she might be excited about. And so we kicked around a lot of ideas. And at that time, which was in about 2003ish, the community had started really talking a lot about the lack of women in computer science, and this was mostly being talked about at the education environment level. And so, my student — her name was Laura Beckwith — she and I said ‘Well, what about gender inclusiveness of software itself.’ And nobody had really thought about that before. And of course we were interested in it from a problem solving perspective not a, you know, pink versus blue or an aesthetics perspective — but is it going to be a level playing field to enable people to bring their different talents to problem solving. So Laura got very excited about it and so she just started reading from everything. She started reading from psychology, and from education and from computer gaming literature and feminist literature and the academic discipline of marketing, she was just reading everything. And these hypotheses about how that might play out in the way software is, just started dropping in her lap. And we took some of them into the lab to investigate whether they really did play out in software and, lo and behold, we kept getting these phenomenal piles of evidence and data about the lack of gender inclusiveness of a lot of software features. She ended up doing her dissertation on that work and as I said, it was really her work but by the time she left, she had me hooked.

ROBERTSON: So, normally we don't think about software being inclusive or exclusive can you explain how it is that software can be exclusive to a particular gender? What does that really mean?

BURNETT: Right, okay. Great question. So, let's get really concrete and think about tool tips. So, tool tips are those things that, when you are on a machine with a mouse like a laptop or a desktop, you hover over something and some little tiny bit of information comes up which you can read. So, tool tips are very nice information features for somebody who is doing what we call selective information processing. And in that style you get a tiny little bit of information like what you get in a tool tip and then you act upon it, knowing, that, of course, you don't have enough information, you might be wrong, but you are willing to try something. You might take one little action and then you might gather a little more information, say from another tool tip. Or you might say, ‘Oh that was a bad idea,’ and undo your way out of them and try some different thing. But it's characterized by a very tight iteration loop between gathering a tiny little bit of information and doing a tiny bit of action and evaluating whether you want to keep going down that path.Now, this style of information gathering is a style that is statistically favored by a lot more men than women and it's the style that most software supports, via things like tool tips.

On the other hand, a style that is favored by a lot more women than men is call the comprehensive style. And so in that style, somebody wants a lot more information before they start taking action and then they will take actions in sort of a batch. So, the iteration loop is much less tight, it's much more batchy. So, in that case this tool tip has a lot of disadvantages. First of all it had only a tiny bit of information. Second of all it disappears as soon as you move your mouse, and so you don't really have the presence of a lot of information at once.

But once somebody starts thinking about tool tips and ways to present information from these perspectives — the information processing style — then there are usually easy fixes. Like, so for example  for a tool tip you could make it pinnable, which might mean you could pin it to the screen and then go get another tool tip and bring that up too and pin that to the screen. So you could be processing a lot of information at the same time and go back and forth and look at it as you started a series of actions.

Anyway, so, this is one way where even though the software didn't put the feature in thinking about gender, and even though it's not like all women are one way and all men are the other way, statistically a lot more women are on the one side of the processing information style and a lot more men are on the other side. And so if software can support both styles then it is going to not be uninclusive anymore. And furthermore it’s going to serve everybody who has this processing style regardless if they are male or female.

ROBERTSON: So, Margaret happily worked on these problems with her students, and had several published papers, until one day she got a wake-up call. Okay, it didn’t sound exactly like that. It came from someone in industry who had a huge problem with his software. His company produced software for a branch of medicine that is practiced by mostly females, and unfortunately females hated his software. He asked Margaret to help him.

BURNETT: And so, at that point I realized, ‘Gee, I don't really have anything to give him, do I?’ All I have is a lot of academic papers about lab studies we've done, but what this person needs is some practical method to be able to figure out what is it about our software that so many people are running into problems with. And what can we do to not have gender inclusiveness problems?” Because he didn't solve that, his company wasn't going to make it. So, that's when GenderMag was born. It's a method for people like that person who contacted me. It stands for Gender Inclusive Magnifier. And so, it’s a method for ordinary software developers, software managers, user experience people — anybody who has a hand in helping to create the software. It's a method that enables them to find gender inclusiveness problems with their own software.

ROBERTSON: So, although Margaret is a computer scientist, what GenderMag is NOT is a computer program that automatically finds gender inclusiveness problems. Instead, she has created four fictional people — you can think of them as characters from a novel. These kinds of fictional people are called “personas.” Each persona has a different personality which is a combination of five problem-solving facets. One of these facets she mentioned before — the information processing style. As you might remember, more females tend to prefer a comprehensive information gathering style. So, one these characters, named Abby, embodies that style along with some other problem-solving facets like risk aversion in technology. The other characters Tim, Patrick and Patricia have a different combination of facet values. Together, the four characters cover a spectrum of values each of the facets can have. I should mention that all these facets were derived from a large corpus of data. Next Margaret explains how someone can use the GenderMag method.

BURNETT: So, the software developer picks one of the personas. And so this is some slice of the population that they would like to address, that they feel like they need to know more about. So, let's say they pick Abby. And they take their software and they say, ‘Okay, I want to see how Abby would do at’ … let's say, for example, ‘adding a new recurring meeting to her calendar.’ And so they say, ‘Well, the first thing we would want Abby to do is to click on this button.’

And so now I've talked about two of these three components of GenderMag. I've talked about the personas, and I've talked about … they all have facets. The third thing that they have is this process. And so it's a set of predefined questions that you ask every step of the way as you try to figure out how Abby would go about this task. Or really whether she would succeed at the task. And so, you say, ‘Well, the first thing I want Abby to do is push this button.” And then you ask one of the questions on our list, which is ‘Will Abby even have that goal?’

And so you go back to the facets of the persona and really absorb what Abby's motivations might be, (that's the second one of our facets), what her information processing style might be, what her risk aversion might be, what her learning style with new technologies might be, and what her self-efficacy (which is her particular self confidence in how well she learns new technologies), what that might be. And they take those things into account and then they say, either yes, no, or maybe. Yes, she will have that sub-goal, no she will not have that sub-goal, or maybe, she might.

And in all three of those cases, or in whatever the cases are, they write down the reasons. So, they might say, ‘Abby is risk averse and the label on that button is down-right scary.’ It doesn't say, ‘Start a New Meeting.’ It says, ‘Maintain Permanent Settings.’ It’s like, she's not going to want to do that! That's sounds like a dangerous thing. And so they write down this reason. So, they have identified a gender-inclusiveness bug. It's usability, of course, it would affect lots of people, but the thing that makes in a gender-inclusiveness bug is that one of those five facets that the value that statistically aligns with more females than males is one that really triggered their decision. And having written down that bug then they decide how important it is to them to fix, and they also have some idea about how to go about fixing it. ‘Well, no wonder she is not pushing the button. Who would if they were risk averse? So, let's change the label on the button.’

So, that's how it works. They just go through the task—everything that Abby is supposed to be pushing, or clicking on, or gesturing, or typing, or whatever—and evaluate that and they will also evaluate whether or not Abby will know she is making progress after they do that.

ROBERTSON: And so, what has the experience been of the people who have been using GenderMag to evaluate their software?

BURNETT: It's been really encouraging. And so, we've found that people can start using it right away. At first they aren’t quite as good at it as they are, you know, 20 minutes down the road, or whatever. But they get it right away. Another great thing is they all seem to really understand that it's just not about gender stereotyping. So, it’s not about aesthetics, it's all about different problem solving styles, no matter what gender those things land in.

But the reason it is about gender is because of the fact so many of the problem-solving styles that are so far being overlooked by software are the ones that are favored by a lot more females than males.

The third thing is, they do find a lot of gender-inclusiveness issues in their software. And this is both good news and bad news. So, it's great news that they have such an easy time finding those issues using GenderMag. That’s wonderful. The bad news is they are finding so many of them, we are starting to learn just how pervasive this problem really is.

So, I did a field study last summer. And in that field study I had two software development teams from a government agency and one from an east coast software development company and one from a west coast software development company. And some of those people were developers, some were user experience people, some were managers, the whole gamut. None of them had any background in gender. And the combined total of how many gender inclusiveness issues that they found was one out of four of every feature they looked at -- 25%.

And this is horribly high. And this was them evaluating their own software. We weren't doing it. It was them. They are evaluating software that they created, they are supposed to love. And they are finding that many problems with it. So, that's the bad news. But the good news is once they find them, as I've mentioned, they have a pretty easy time figuring out how they can fix them, and so that gives them the ammunition they need.

Since the time I did that field study which has actually been published (you can read about that if you want to), I’ve done… I've worked with a lot of other teams. And so, the bad news there is that it actually is even worse than one out of four. When I put everything together that I have the data from on these teams working with GenderMag, the average is now up to one out of three. So, this is over, oh my goodness, at least 20 different software products ranging from early prototypes to software that is still in use after 10 years, and everything in between. Ranging from desktop to mobile to smart watch to you-name-it.

ROBERTSON: So, how flexible is GenderMag? Are people using it as you intended? Or are they figuring out other ways to use it?

BURNETT: Great question. This is my change-the-world project. And so, I don't really care if people use it exactly the way I invented it. What I really care about is that people find a way to use it in a way that works for what their company — their team — needs to make their software better.

So, it's turned out that people shape it all different ways. But at the very core of GenderMag are these five facets. And so long as they don't mess with the 5 facets, it turns out that GenderMag deconstructs really well. So you can use just the five facets and still make a lot of progress. Just be talking about things like information processing style among your software development team helps.

Some software development teams love personas. And so, it turns out you can use just the personas without the rest of the method because the personas encapsulate the facets. And so some teams do that and it works really well for them.

Other teams, they absolutely hate personas. That's fine. They can just use the questions with the facets. And get a lot of mileage that way. So, because of the fact that it's turned out to be so deconstructable, you can use bits and pieces of it and still get a lot of mileage.

ROBERTSON: What are your goals for GenderMag?

BURNETT: As I've mentioned, this is my change-the-world mission. And so I would like GenderMag to be used by everybody to figure out how to make their software more gender-inclusive and more inclusive in general. It's all downloadable, it's all free. I hope people try it out. I hope people will let me know how it goes for them because I'm still collecting data. We improve it about every 2-3 months, so they are welcome to download it as often as they want and it's freely available to everyone.

ROBERTSON: Inclusive design is one way that researchers in the College of Engineering are promoting diversity. Improving accessibility is another. The next story is about someone who is working on improving the lives of people who are blind or who have low vision.

RECTOR: My name is Kyle Rector and I’m an assistant professor in computer science at the University of Iowa.

ROBERTSON: So, you are probably wondering why I would be featuring someone from the University of Iowa. Kyle is actually a graduate of Oregon State University. She got her undergraduate here in both electrical and computer engineering and computer science before going on to get her PhD at University of Washington. During her time here, she was a student researcher with Margaret Burnett.

RECTOR: Yeah, I actually worked with her for four and a half years. I’m very lucky that I got to work with her. It was a research solicitation and it had to do with gender and technology. And so I was interested in learning more because I noticed there were so few females in my electrical engineering classes. And actually it was Margaret Burnett who convinced me and encouraged me to join computer science and also to go into research.

ROBERTSON: So, here’s the cool thing. Not only does Margaret’s research have a direct impact on improving software, but her students go out into the world and amplify those impacts. On top of everything else Margaret is an award winning mentor for undergraduate and graduate students. Kyle is just one example.

RECTOR: My area of research is in human-computer interaction, so similar to Margaret Burnett. But more specifically I’ve gone into accessibility, so trying to design technologies to help people with disabilities. And mostly I work with people who are blind or visually impaired.

ROBERTSON: Have you ever thought about how someone who is blind could learn yoga? Probably not. But Kyle did.

RECTOR: It’s called eyes-free yoga because there is no screen involved. So, the entire system is operated with your voice and also by listening to the audio and verbal cues.

EYES-FREE YOGA: Rotate your shoulders right. Rotate your shoulders right. Ding. Lean forward. Ding.

RECTOR: So, what I did is I built a system using the Microsoft Kinect, which, in case you don’t know what that is, that’s a depth camera. So, in other words you can place it on your mantel or a desk, or about 4 feet off the ground. And the idea is that this camera can assess your body posture. And so, with that information I can now give a very detailed verbal instructions for a yoga workout, but not just tell you what to do, but then I can assess your body in three dimensions and figure out if you are actually holding the posture properly or if there are things you can improve. And so then the system actually provides detailed verbal feedback so you can adjust your body. And this based on what you are currently doing. So, it is personalized and in real time.

EYES-FREE YOGA: Your core is good. Your legs are good. Your arms are good. Good job!

ROBERTSON: As you can hear the program also gives you positive feedback. Kyle did some testing to find out how her system with the feedback compared to a system that did not have feedback, like a blind yoga audio CD. It turned out the participants preferred having the feedback.

RECTOR: They really liked hearing how they were doing and not wondering that. Actually hearing some information that is concrete and understandable can make a difference. And then I converted that prototype into a full work-out system so now Eyes-Free Yoga has four full work-outs. All the standing yoga postures have custom feedback. And then I added some fun gamification techniques as well to try and persuade them to continue exercising.

ROBERTSON: Kyle’s system is available to download from her website which we will link to in the show notes for this episode. Or just search eyes-free yoga.  And keep an eye on Kyle. She is looking to expand her research beyond yoga.

RECTOR: What about faster paced exercises like aerobics? Or maybe you don’t want to exercise inside of your house but out and about. How do you increase accessibility of gyms and running tracks and swimming pools and other spaces like that. So, that’s where I’m seeing myself going in the next several years.

Working in the field of accessibility it’s all about trying to make technology and aspects of quality of life more accessible to people with disabilities. There are even researchers focused on computer science education, so how do you make computer science more inclusive of people with disabilities, and just all sorts of different aspects: employment, recreation, connecting with others. I think it’s a great field to be in for the aspect of diversity. 

ROBERTSON: That’s it, my friends. I hope this episode expanded your view of what engineering can do for society. Thanks much to the awesome Oregon State music students Emma Nissen, Blair Bowmer, Eldon Dela Cuz, and Jordan Mitts (I hope I said all your names right) who provided the music for this episode. And special thanks to Professor Steven Zielke, director of choral studies here at Oregon State. The music you are hearing would not have been possible without his help.

This episode was produced by me, Rachel Robertson, with additional editing by Miriah Reddington. Our intro music is “The Ether Bunny” by “Eyes Closed Audio” on Soundcloud and used with permission via a Creative Commons Attribution license. A final shout out to Eric Gleske who has been our patient and informative technical advisor.For more episodes visit engineeringoutloud.oregonstate.edu.

TRANSCRIPT
[MUSIC: Eyes Closed Audio, The Ether Bunny, used with permissions of a Creative Commons license.]

NARRATOR: From the College of Engineering at Oregon State University, this is Engineering Out Loud.

MOVIE CLIP (2001: A Space Odyssey): Hello Hal, do you read me? Do you read me, Hal? Affirmative, Dave.

RACHEL ROBERTSON: Hello podcast listeners. Do you read me? This is Rachel Robertson from the College of Engineering at Oregon State University. Welcome to the first episode of Engineering Out Loud. Today I have Jens Odegaard with me in the studio. Jens, can you tell us what you do?

JENS ODEGAARD: Yeah, I work in marketing and communications in the College of Engineering here at Oregon State and I kind of focus on nuclear engineering as well as radiation health physics.

ROBERTSON: Very good. And I’m also a communicator for the College of Engineering but I focus on electrical engineering and computer science. So, our listeners are probably wondering what that clip 2001: A Space Odyssey has to do with engineering. It sounds pretty cool but it also a purpose which will be revealed later in the podcast. First I wanted everyone to hear from you, Jens since it was your idea to have this podcast, maybe you could just tell us a little bit about why you had this idea, what was the purpose?

ODEGAARD: So I come from a marketing and communications background but I’m talking to engineers here all the time on campus and they have really interesting research, really cool things that they produce and are building here, but most of the time it’s published in technical journals and in places where the general public wouldn’t run across it. So, I was thinking if we made a podcast to tell stories about their research and innovation that they are doing here it would open it up to a more broad audience and also help break it down from the technical details to a place that everyday folks can understand, like myself.

ROBERTSON: We should also mention that there is a whole team of us who have been working on this podcast including Krista Klinkhammer, Steve Frandzel, and Johanna Carson and you’ll be hearing from them all later on in the season. And our behind the scenes folks Mitch Lea, Jack Forkey, and Megan Kilgore. So, we should also tell our listeners a bit about the format.

ODEGAARD: Yeah, so the format of the podcast will be three seasons per year: fall, winter and spring to coincide with our academic terms here at Oregon State, and in each season we’ll explore one big topic and we’ll have six episodes per season that each tie back to that main topic from a different angle. So for this first season we are going to explore data science and engineering and now I’m going to turn it back over to you, Rachel, to introduce data science and engineering and tell us what that even means.

[Music: Eyes Closed Audio, The Ether Bunny, used with permissions of a Creative Commons license.]

ROBERTSON: All right! So, here we go. So data science and engineering is an area of research that holds great promise for improving our lives, but also has some potential pitfalls.

MOVIE CLIP (2001: A Space Odyssey): Open the pod bay doors, Hal. I’m sorry, Dave. I’m afraid I can’t do that.

ROBERTSON: Now we are back with Hal in a fictional 2001. But in reality 2016 we have not yet experienced a robot take over, instead the age of Big Data has ushered in a host of concerns we could not imagine in 1968 when the film came out. So, to help us understand this topic of data science and engineering I’ve invited Tom Dietterich to talk with us. Tom is a distinguished professor here, which is the highest rank the university can give a professor. He’s also just really fun to talk to. But more relevantly he is considered to be one of the founders of machine learning, a branch of artificial intelligence.

TOM DIETTERICH: So I’m very interested in how we can use data to build computer programs that do interesting things. And I do a lot of applied research in problems in sustainability and ecosystems and lots of the things that people on this campus are concerned about.

ROBERTSON: As we began putting together the episodes for this season we realized that data science and engineering can be tricky to define, but one thing I’ve figured out for sure is that it usually involves Big Data, extremely large data sets that can be analyzed for any number of applications. So, my first question for Tom was how is Big Data related to data science and engineering.

DIETTERICH: Well, so, when we talk about Big Data we could be talking about multiple things. On the one hand you have big data that is coming from scientific instruments — for example, the Hubble Space Telescope or any of these other huge telescopes that NASA has, or satellites that are sensing the Earth. They generate tons of data every day. But the other kind of Big Data that we think about a lot today is sometimes called our digital exhaust which is when we use our cell phones, when we search on the web, even now when we drive around in our car with our Google driving directions enabled, we are leaving a trail of the locations we’ve been, the websites we’ve visited. So that, for example, now Google can tell you for a restaurant or a stadium – when does it tend to be busy? And that’s based just on tracking how many phones were there at different times of the day. So, that’s a kind of Big Data…sort of as a side effect of other things we are doing.

ROBERTSON: So, now that we have a handle on Big Data, the next question is how is it related to date science and engineering?

DIETTERICH: So, the field of data science is really the marriage of statistics and computer science. In the past, data sets were small enough that statisticians could analyze them sort of interactively and manually. But with these massive data sets we need much more automation to find the interesting galaxies or (if you are analyzing the telescope data) or to learn to detect where accidents might be happening because you are seeing some sort of bunching up of the cell phone traffic. And so we can’t afford to have people do all that manually. There’s just too much data and too many questions. And so that’s when …so data science then brings in automated advanced algorithms from computer science to deal with the huge amounts of data for those problems. Data engineering. Well, that means different things to different people, some people think of it as engineering the data, but I think normally at Oregon State we think of that as doing engineering using data.

ROBERTSON: In fact, in this season of our podcast you will hear from several researchers who are using data for engineering in a broad range of topics: Geoff Hollinger is teaching underwater robots to incorporate human preferences in their decisions about where to travel to gather data; Xiaoli Fern is using machine learning to identify birds by their song from recordings in the woods to help biologists track bird populations; and Haizhon Wang models various evacuation scenarios – such as walking or driving – in response to tsunamis to help improve evacuation plans. But data engineering is not all for science.

DIETTERICH: Facebook encourages you to label your friends and then they use a machine learning system to …for the computer to analyze all of those images and the person’s name and try to figure out what kinds of patterns in the images predict that this is Tom Dietterich’s face versus someone else’s face. So, that’s doing engineering with data. There are many other examples. Self-driving cars have to recognize pedestrians and pets and dangerous conditions and so on and all of that is collected again by using data and then applying machine learning techniques to create the software to do those recognition tasks.

ROBERTSON: During this season will talk about the many benefits of data science and engineering, but first we wanted to talk more about what the dangers are. One of the obvious dangers are the privacy concerns we open ourselves up to when we use applications and devices that can be tracked such as Facebook, Google and our cell phones.

DIETTERICH: We have generally had an expectation of privacy in the U.S. even in the public sphere where we are walking around on the streets. I mean, people say one of the attractions of living in a big city is the kind of anonymity you have, well, you don’t have that anonymity anymore. Presumably we will be able to have things like, you know, you can look up and ask where your friends are and if they have allowed you to access that information it will tell you they are in the subway at 25th Street and heading in your direction or something. And you can see how that might be useful but you can also see how that might also really quickly lead to dangerous things. It’s an ideal tool for a stalker or a terrorist or a criminal. So we currently are relying primarily on the fact that companies have a tremendous interest in not having privacy be violated because people will stop using their products, but we don’t really have strong rules and regulations and so many companies have been compromised in one way or another. Data sets compromised, I mean, you know, our social security numbers seem to be pretty accessible, credit card numbers, health records and so that’s I think is the big danger is that without really good care in computer security all this information has the potential to become public.

MOVIE CLIP (2001: A Space Odyssey): This mission is too important for me to allow you to jeopardize it.

ROBERTSON: So, back to Hal and Dave from the beginning of our podcast. Our fear of an evil artificial intelligence has been around for a while and it resurfaced recently when Bill Gates, Stephen Hawking and Elon Musk spoke publicly about the dangers of artificial intelligence. If fact, Musk declared it was “our biggest existential threat.” Tom Dietterich has been the voice of the academic perspective in this debate. So, we’ve talked about what you think the dangers are of Big Data, but what are they not?

DIETTERICH: Well, Hollywood loves the story of the robot that develops its own will and its own goals and those goals conflict with humans and so it decides it needs to kill the humans to achieve its goals. Perhaps the most realistic scenario might be from 2001 where the computer system on the spacecraft was programmed to prioritize the mission goals above things like keeping the crew alive and so it made a decision that the crew was a threat to those goals so it kills the crew. And of course then we have an epic man versus machine competition and man wins! And it’s a great story. I think that the real threats are more likely to be software bugs. We have a lot of trouble in making software that works correctly. As we all know, aps on our phones crash, our phones have to be rebooted and things like this, and you certainly wouldn’t want a computer system in a safety critical application like a self-driving car that says, ‘oh, you need to pull over to the side of the road because I have to reboot.’ This is just …a car is not a phone and so developing software that is reliable and robust in these kinds of high-risk applications, that’s a huge challenge for computer science because it’s always been a challenge in avionics and spacecraft and even with the very best software engineering techniques mistakes still happen in those settings.

ROBERTSON: So, there are some obvious dangers to big data, but what are the benefits of it.

DIETTERICH: Well, I think there are tremendous potential benefits particularly in health care, for example. So, right now when someone is studying the effectiveness of a drug or looking for potential interactions between multiple drugs that might be adverse they really have to guess in advance what those might be and then include them in a controlled clinic trail. But there has been recent work at Microsoft that says we can detect adverse drug interactions just by looking at people’s web searching behavior. They might look up one drug and then another and then might also be looking up some symptom that they’re having. Other benefits, well, we already have a lot of benefits like I was mentioning before, I can now look at my phone and ask well, ‘When is this restaurant busiest?’ Or I can look on the Google map and say what time do I need to leave Corvallis in order to make sure I get to the Portland airport by 9:30 in the morning and it can tell me well, it usually takes between an hour and a half and two hours on a Thursday morning to get there. How does it know that? Well, because it’s got this information from people’s phones in their cars, plus some very clever algorithms. So, I benefit from other people sharing their phone data when they are driving. So, there is big benefits like medical and there are convenience benefits like driving. A lot of the benefits, it’s hard to even imagine what they are right now. I like to tell the story that when I was a young scientist we had just developed the internet and we thought the purpose of the internet was to allow computers to communicate with each other and so when the internet was rolled out and everything as working around 1985, we said, ‘okay, we’re done.’ We had no idea that the main purpose of the internet would be to put people communicating with other people and that this would be the revolutionary thing. So, I think it’s very difficult to forecast what will be the ultimate ways in which we use these new technologies. But they are likely to be interesting!

ROBERTSON: To find out about the interesting things we are already doing with data science and engineering, keep listening to this season of Engineering Out Loud from the College of Engineering at Oregon State. This episode produced by me, Rachel Robertson, with additional editing by Mitch Lea. Our intro music is The Ether Bunny by Eyes Closed Audio on SoundCloud and used with permission via a Creative Commons 3.0 license. For more episodes visit engineeringoutloud.oregonstate.edu.