Wednesday 8 July 2020

Graphistania 2.0 - Episode 7 - The one after the Covid-19 lockdown

Yes! We were able to record and publish another episode of our Graphistania podcast. It's been an amazing and turbulent couple of months - but before the summer holiday season really takes off we wanted to get this to you.

Wishing you a fantastic and relaxing time - and in the mean time enjoy this episode!

Here's the transcript of our conversation:

RVB:00:00:00.000 Hello, everyone. My name is Rik, Rik Van Bruggen from Neo4j, and, yep, this has been a long time in the making. A couple of months have gone by, but we're back. We're going to record another episode of our Graphistania Neo4j podcast, and to do that, I have my dear friend and colleague Stefan Wendin on the other side of this Zoom call. Hi, Stefan.

SW:00:00:22.622 Hello, Rik. Nice to hear your voice again and being back in this Graphistania. I thought I'd lost my passport, but I find it again. So it's good to be back and see that customs are once again open.

RVB:00:00:37.496 I am very happy that we're back here, absolutely. Stefan, it's been quite an exciting couple of months, right? I mean, I think the last recording is from somewhere in February or March or something like that. So we've had a little bit of a silence, and that almost overlaps with one of the most exciting and strange and, yeah, bizarre episodes of my lifetime, the whole COVID-19 lockdown, obviously. I wonder how you lived through it? What was your personal experience like?

SW:00:01:13.382 Yeah, I think you summed it up with the word bizarre. It was almost like it was a parallel universe, I guess. I have the fortune to live-- or before, I was so fortunate to live in Sweden, and we weren't really doing that lockdown super hard. Of course, we tried to do a version where we should put the responsibility on people instead of just locking everybody, which is now, again, coming back to us, I guess, because now I can't travel because we didn't do the lockdown, so. But I think it was more busier than ever in some areas. You can see with some clients-- I mean, any government or agency sectors about life science research, those were kind of exploding. So that was more busy than ever. And then you can see some also struggling. So it was a little bit like this kind of paradox. It was more work than ever, but at the same time, you can also see previous colleagues or friends not working at all. So it has been a challenge, I guess. But with any challenge, you also, when reflecting, kind of grow as a person, I guess. So as we said before the call, I dropped eight kilos during this time. I don't know--

RVB:00:02:30.646 Unbelievable, yeah?

SW:00:02:32.048 Yeah. So I think I came out as a better person, trying to kind of not do so much of it has to be black or it has to be white, this kind of false dilemma kind of thing, but rather trying to see that it can be a couple of different scenarios and both persons can be right. What about you?

RVB:00:02:50.558 Yeah, no. I mean, for me as well. Obviously, I live in Antwerp, in Belgium, and we did have quite a harsh lockdown from the middle of March until early June, and that also meant a whole set of changes for me personally. I mean, I have three kids. Their school was cancelled. We had to kind of homeschool them/entertain them/make sure that we as a family didn't go completely crazy and get some work done in the meanwhile. So, yeah, I agree with you. It was a bizarre time, but it was a very busy time. Professionally, it's been busier than ever. And then the whole personal family life thing, it also came with a little bit of additional concerns. But I kept on thinking throughout this entire lockdown that we're the lucky ones. Do you know what I mean?

SW:00:03:52.064 Yeah, yeah, yeah.

RVB:00:03:52.525 I feel incredibly fortunate, and that's the feeling that really sticks with me, that I haven't been sick. I have been healthy all the time. I was able to ride my bike. I was able to eat well, live well, stay connected with my friends. For me, it wasn't really an issue. You know what I mean? I didn't go outside as much, but at the end of the day, it was very comfortable. I kept on thinking, too, for example, many of my boy's classmates who aren't as fortunate, they live in a small apartment here in Antwerp with three, four people, and they only have one computer, and they are not as fortunate as we are. And I do think that it was extremely heavy on those people, and I hope that, in some ways, we can help them out as a society because I don't want this to continue to weigh on them the way I fear it might. So my heart goes out to other people, sick, healthy, lonely, those types of people.

SW:00:05:13.183 And I think this is-- just to talk a little bit on that, I think the interesting part is, this is, basically, the first time we've had this kind of pandemic with a working, or a non-working, or-- internet, actually, is working, and the problem is that we most likely cannot really handle it, because of the amount of news spread or misinformation, and those kinds of things. So we have this kind of-- how it, at the same time, gets a little bit worse because we have misinformation, but we also, on the other hand, have this growth in empathy or caring for others. So it's really an interesting time, from studying people and behaviour, I think, which is what I'm kind of obsessed with.

RVB:00:05:57.914 Let's see what happens afterwards, right? And let's hope that the worst is behind us and that-- the good things that can come from that, like your 6:45 workout. Keep it up. Your bike riding, let's keep it up. Caring for other people, let's keep it up. You know what I mean? Let's try to keep the good stuff. And that seamlessly brings me with something I wanted to talk to you about, because I know that you've always been doing these innovation labs for Neo4j, right, helping people innovate with graphs. And you've been doing them remotely in the past couple of months, and I know that you've been using some really interesting and exciting new techniques and tools for that. Maybe it's useful for us to talk a little about that before we talk about some of the interesting use cases that we saw. What do you think?

SW:00:06:48.230 Yeah. Yeah. I think it can also be a really neat bridge, basically, into some of the use cases, I think. So as for those that don't know, I run the Innovation Lab here in the middle of Neo4j. Basically, what we do is kind of looked upon the knowledge that we as a company has had during 10 years of enterprise, at the option of this, and then we design processes to kind of speed up time for innovation and, basically, validation, picking the right thing. The problem with that process is that it was based, basically, on design thinking with some slight changes, and that is heavily in person, right? So seeing the pipe going completely empty was a bit of a shock. So then I looked up and was like, "Okay." But I'm not so much a fan of this new normal talk back and forth. I think there is only change, so we only have to adapt all of it to it, right? So I very early on study on, "How would I like to do a meeting? What are the things that are challenging to me as a person?" I get really annoyed when there's bad facilitations. So I could just see it's like, "When do I struggle in meetings?"

SW:00:08:02.962 So then, I kind of looked upon that as a design problem. So how can I design that to go away? So we started to use, of course, the breakout rooms in Zoom, which I think is an awesome feature. And you can use it very creatively. One of my favourites is this 1-2-4-All, where you basically go alone in a room. You go with a friend; that's the two. You take those two pairs, which is the four. And then all comes together to kind of structure the way you talk about things because very often in talks you have this kind of, "Oh wait. You go. You go. Can you see my screen?" and those kind of things, right? So we start to use also Miro, which I think is a great tool for online collaborations, for both putting together the business proposals or the pitch of the value proposition, basically, as of any idea-generating exercises. So there's a lot of that going on, which was neat, and in that, this is actually the most busiest time because we had a very, very healthy pipe because of the graph community literally exploding, right? But also, in this time, that went completely blank. But then only in a couple of weeks I kind of rebuilt that pipe. So it has been busy as I have never been busy before, to be honest. I've delivered these on back-to-back since COVID started. And one thing that really stood out in that is in the area of pretty much kind of life science or research, I think, anything from kind of patient journeys, drug discoveries. There is this interesting thing about the COVID graph, for example. Do you want to fill in something here, Rik?

RVB:00:09:48.724 Yeah, absolutely. Because, obviously, innovation is something that we all really care about, but it kind of brings us to the topic that we usually have in all these podcasts, which are the interesting use cases that we saw. And obviously, in this stressful time, there were a few use cases that popped out. And maybe we can talk a little bit about that here. Maybe I'll start. I saw some really interesting work being done in the healthcare domain, right? And I think you shared that with me a little bit, but things like, for example, the COVIDgraph, where friends of ours from the German Centre for Diabetes Research, together with other people in the community, together with our own staff, they started working together on this huge project that brought together all kinds of knowledge and information about COVID, right, and the coronavirus research. Amazing project. We did a podcast episode about that in March, I believe. Amazing story, but it led to a number of other really interesting cases.

RVB:00:11:12.445 For example, there was a fantastic blog article on the Neo4j developer blog about modelling patient journeys. What happens to a patient as they get treated for a particular illness or disease, all of the steps that they have to go through? And I used to work a lot in the healthcare domain, and I know that this type of journey analysis can be not only the difference between health and sickness but also the difference between efficiency and non-efficiency, you know what I mean? Between happiness and unhappiness. Because if the journey of a patient, if that starts going wrong, it can not only lead to a prolonged illness, but it can also lead to a lot of frustration, a lot of anxiety, exploding costs. It's one of those things that, looking at that journey, it's a fantastic thing to really think about and model. And guess what? Graphs are really interesting for that.

RVB:00:12:20.740 And then I spent quite some time also in April. I remember it was the Easter holiday here in Belgium, and kids were even less busy from school than they were because of the lockdown, but I spent quite a bit of time thinking and writing about contact tracing. Right? It's one of those use cases that has also really stood out for me as an interesting application for graphs, for us as a society. And it's all about understanding how people are in contact with one another, right? I mean, we had to lock down entire countries because we didn't know any more how the virus was spreading through our societies. And, of course, if we would be able to securely and in a privacy-conscious way track how people are potentially spreading the virus, then, well, we might be able to avoid it, and we might be able to avoid having to lock down entire societies. So I thought that was a fascinating piece of use cases and of material that I was able to gather there. Really, really cool stuff, also, for graphing data science. I really enjoyed that. But maybe you could talk about some other stuff that you've seen.

SW:00:13:50.660 Yeah. No, no, but I just have a reflection on your reflection, which is like a meta-reflection. No, but I think it's like, when you talked about this patient journey-- and as you know, I've injured my back way, way back. I was very, very struggling, and after coming out of that, I worked with one of the big companies, and I will not name it, treating cancer, so we looked upon modelling patient journeys and predicting where it would be frustrating and hard and which are the painful stuff. A super graph-use case, right, as we all know. But remembering back then, people were like, "Yeah, I can see that this has clearly value, but this is something Apple should do," was the comment of the CEO of this global company. And I was like, "You see the money. You see that it helps people." But I think the struggle was here that he could just not see himself doing it because they were so stuck in the mental model of tables, treatments, and stuff, and not having this flexible approach.

SW:00:14:50.446 I think this is also something becoming to me very clear now with all the different-- we did several labs in the data science sector, right? And seeing a company or organisation after another understanding the power of not only the graphs but using the Graph Data Science Library and algorithms to kind of iterate and kind of segment and understand this big data. So I think this is super cool, to see that this is finally happening. And I think, in one sense, all of this almost in a sense helped this to happen, I think. Or we didn't have time to ignore it any more. Now it was real, and we all had to work on it. So I think that was very, very cool.

SW:00:15:33.504 But it was a lot of other things. One thing that I was thinking also, on the topic of this kind of Apple should do this kind of thing, I stumbled upon, when reading the articles, this Neo4j Commander, which seemed super cool. Basically to, without cypher, in a dashboard kind of sense, editing your database, and it reminded me of a story when I was at the age of 30 or something, last year, basically. I'm young and fresh, right [laughter]? Okay. That was a slight lie to all the listeners. I'm 46, I think. So I was moving to New York. We were starting an agency. It was me and another Swede starting a agency with American employees, Japanese owners. You can imagine that cluster. It was a mess. At the same time, it was the release of the iPhone. And I remember that time. I was super excited. I loved this idea of good UX and good user experience. And when you put time into design, not only visual design, which is of course, what they also did in the iPhone.

SW:00:16:47.635 But I remember that meeting. I was sitting in a corner office in Park Avenue in New York, on the 27th floor. And then the client said, "Before the rest of the board arrives, Stefan, would you mind put the toy inside of your pocket?" And I was like, "What? But you have your phone on the table." He was like, "Yes, but this is a working phone." I was like, "Yeah, but I use this for work as well." "No. No, no. This is a toy." And I was like, "Why is it a toy?" I couldn't really understand it. But then he said it, and I think these words and the sentence kind of resonates with me a lot. "No, but everything is so simple on it. This cannot be serious, and it cannot be work." And I was like, "Does work have to be boring and hard?" And I think this was my immediate thinking. I haven't tried this in the everyday Commander, so I don't even know if it actually works. This is on my list to do for the vacations. But I saw this idea.

SW:00:17:45.478 When we're entering new paradigms with technology, it's very often that we default back to the old way of doing it. It was harder. We learned it, so now it's easier for us. Basically, this is the same as riding a car using a stick, right? And manual gears, right? It's like, "Why the hell would you do that?" "Oh, no. But I need it because I have an aggressive style of driving." It's like, "Yeah, there's something called sports mode. It's been around for 15 years. Get the grip, man." So what I think, this is also an interesting idea, of this idea of mental models and change. And when we're forced into change, and all of a sudden, it feels like the most normal. I mean, nobody would ever say iPhone would be a toy now. Now, it's just boring. It become basically status quo nowadays, right?

RVB:00:18:33.679 I agree. So what are really interesting topics that we should, maybe briefly, mention here to our listeners? I mean, obviously, there was a big product change in Neo4j. We actually brought out quite a bit of new product functionality with the Graph Data Science Library new version that came out, but also Neo4j 4.1 that came out. Some really exciting stuff there. And then, I think these are all things that are incrementally helping people in the graph community become more productive, right? Lots of good stuff there. And then the last one that I wanted to mention I was obviously the best [laughter]--

SW:00:19:18.775 Of course the moment we have been all waiting for.

RVB:00:19:20.278 The best [laughter]. There was a great article that I-- actually, it just makes me smile because it's such a simple but great example of graph power. There was a great article of using collaborative filtering on--

SW:00:19:39.352 Dru dum dum dum, bshh--

RVB:00:19:41.164 Drum roll [laughter]. Beer recommendations. Yep. Yeah, absolutely.

SW:00:19:46.053 Yep. There it was.

RVB:00:19:46.769 There it was. Yep. So I'm still smiling when I read that.

SW:00:19:52.190 Yeah. And I think it's also kind of cool because it's-- as I've mentioned before, when joining Neo, I was of course googling, "What the hell has happened during the last 10 years?" and then of course, I stumble upon Rik's very famous, you must say now, or you can't get rid of being the original beer graph, right? So my only thought when I saw this, "Oh. I wonder who will be the great persons joining Neo now, after this beer graph, because that's the reason why I joined." So--

RVB:00:20:23.153 Yes [laughter].

SW:00:20:23.370 --it makes perfect sense, which is super cool, which comes back to one of the last things that I want to mention. Because I'm kind of always interested in these weird patterns of people, right? And how we kind of say that we do one thing, but then when nobody watches, we do a completely other thing. And so, very often when I do [beta?] or prediction, I put on the hat of an anthropologist. And very closely related to this is, of course, what you could see in cyberSW 1.0, which is basically putting a graph of geology data, mapping that using heat maps and stuff. And I think, this is, again, one of those beautiful examples of that graphs are basically everywhere, and how it's such a flexible structure dealing with complexity and showing how a network, and an ecosystem, do exist over time. So this is also something that I'm super interested. I sent it to some friend in Uppsala University, and we agreed on, "This is a perfect thing of sitting in the hammock during vacation and drinking [foreign]," which is-- I don't know. I don't even know how to translate it. It's one of those flower and one of the juices that you make in summer, so it's super summery. But we're going to spend time doing this. I think it's super neat. They keep reminding me how graphs can actually push insights, especially using, again, Graph Data Science Library; I really loved that. So yeah--

RVB:00:22:02.627 So we [laughter]--

SW:00:22:03.256 --keep it going, I guess.

RVB:00:22:06.317 Absolutely. So I think it'll probably be a couple of weeks or months before we do another episode, because the holiday plans, right? I know what I'm going to do. I'm going to I ride my bike quite a bit, I know, with my boys. You've got any massive plans as well, Stefan? I hope so.

SW:00:22:28.813 No, I'm waiting for you to ride your bike to Stockholm and bring me one of those Tripel beers that you have. No--

RVB:00:22:36.426 It's a matter of time, man [laughter].

SW:00:22:38.610 It's a matter of time. That actually is a good phrase. As we all know, time is relative, right?

RVB:00:22:43.828 Exactly. Cool. Hey, Stefan, thanks so much for taking the time again.

SW:00:22:45.481 No, [crosstalk] take a vacation, yeah?

SW:00:22:49.465 Yeah. Super nice. Yeah. Have a great vacation, then, to all of the listeners. And if you find any cool and interesting graph use cases, why don't you let us know?

RVB:00:23:02.573 Please do. Have a great vacation. I'll talk to you after summer. Have a nice day.

SW:00:23:08.400 Have a nice day.

RVB:00:23:10.095 Bye.

Subscribing to the podcast is easy: just add the rss feed or add us in iTunes! Hope you'll enjoy it!

All the best


No comments:

Post a Comment