What did journalism teach you about data storytelling? | Sharon Machlis | Data Science Hangout

Welcome back to the Data Science Hangout, everybody. I'm Rachel. I lead customer marketing here at Posit and the host of the Data Science Hangout.

And I'm Libby. Oh, sorry. I'm Libby. I'm a community manager who works here at Posit with Rachel. And I'm also a Posit Academy mentor.

And we're so happy to have you joining us today. The Hangout is our open space to hear what's going on in the world of data across different industries, to chat about data science leadership, and connect with others who are facing similar things as you. And we get together here every Thursday at the same time, same place. So if you're watching this as a recording in the future, and you want to join us live, there'll be details to add it to your calendar below. We're all dedicated to keeping this the friendly and welcoming space that you all have made it over the years, and love to hear from you no matter your years of experience, titles, industry or languages that you work in.

And we are so happy to be joined today by our co-host Sharon Machlis, who is a longtime tech journalist and data professional who has recently retired. Everybody celebrate with Sharon that she's recently retired from Computer World slash IDG Communications, renamed Foundry. Sharon, would you like to introduce yourself?

About Sharon Machlis

Yeah, I am a longtime, as you said, longtime technology journalist. I started my career as a daily newspaper reporter, and when we get into unexpected career paths, I wanted to be a political... I love data. I actually was thinking about majoring in statistics, but at the time I went to college, you know, when dinosaurs roamed the earth, there was no R, no Python, no data journalism, and I took political stats. I'm like, wow, should I change my major? I really love this. And I'm like, well, what can I do with this? I don't want to teach. I don't want to be an actuary. No, I'll stick to what I was doing.

Two of us had wanted to be the statehouse reporter, and the other guy got that job because the editor said I was better with numbers, and I ended up as the business reporter. And that was when the Route 28 area outside of Boston was like a hotbed of many computer technology. And so I started covering tech, and I'm like, wow, I super love this. So, and that's where it all started.

I spent a long time working as a journalist, a tech journalist, writer, and editor at Computer World. And in 2017, I was asked to change roles to be a data professional because I was doing data journalism. And to keep my R skills sharp, I started working with our website analytics, and people liked those reports I was doing, and here we are.

And I just retired a month ago. So July 31st was my last full-time working day. So this is all pretty new. So thank you for having me, even though I'm now leading the life of leisure. But I do hope to still keep my hands in things and follow the R space, the generative AI space, because things are so cool and exciting and fun.

How did you decide to start using R?

Well, it's really funny because I was working with our federal government reporter on a data journalism project, and he got this really cool data set. And it had over 100,000 rows, and this was, you know, over 10 years ago. And I'm trying to open this in Excel. And then we're like, we find something odd here. And then the source for the data set is like, oh, no, there was a mistake. And then, of course, you know what happens. It's like, oh, my God, I have to like do all my formulas all over again. Shoot me now.

And I don't even remember where I found out about R, but I found it. I'm like, why did someone not tell me about this sooner? So this was like in 2012 or 2013. This was pre-Tidyverse. So I was learning base R. And then the Tidyverse happened, and dplyr happened. And I'm like, wow, this is the greatest thing ever. Because I had coded in a number of different languages before.

I mean, my graduate computer class, we learned Pascal just to let you know how long ago we're talking about. But I had really liked Perl. I liked PHP. I learned SQL. But somehow the syntax of R just spoke to me, particularly when we got to the Tidyverse. dplyr was like, oh, my God, this is just amazing. And that was it. And I switched from SQL and PHP to R.

I am trying to learn more Python now because I'm super interested in generative AI. And while you can do some good generative AI stuff with R, I mean, all the platforms, all the really good platforms like BlankChain and Haystack and LlamaIndex, they're all either JavaScript or Python, which is fine. And now I have time to learn.

What did journalism teach you about data storytelling?

Yeah. In my time as an editor mentoring younger writers, the big thing I always would tell them is you have to answer two questions. So what? Who cares? And that's with everything. That's really with everything. And the journalism things about you need to grab someone early. You have to have a good headline and a good lead paragraph because people have the attention spans of a gnat.

I started my career in the print days when you had page one of what we would call above the fold. It was sitting there on the newsstand or in the machine. And you needed to grab people if they wanted to take money out of their pocket. I mean, it's kind of the same online. It's not that different. You need a good headline and a good description.

And it's the same with data. I learned early, we would be in page one pitch meetings, the daily page one, and you'd have to make your pitch to the editor who's like the overall editor who was just not as immersed in this cool thing. And they call that an elevator pitch now or whatever. But I mean, you have to be, if you can't say it in three sentences about why this is cool and interesting, then go back and do it again. I mean, that's just kind of the things you learn as a writer.

And it is the same thing with telling stories with data. And I needed help with people to make that connection because somehow I was not doing that when I started doing data stuff. It's like, wow, look at all this cool data. Wouldn't everyone want to spend half an hour and dive into that? Yeah, they don't. So, you know.

So what? Who cares? And that's with everything. That's really with everything. But I mean, you have to be, if you can't say it in three sentences about why this is cool and interesting, then go back and do it again.

Finding the data community on Mastodon

Yeah, so I mean, not to get all political, but I'm just, I mean, I loved Twitter until the ownership changed and I got unhappy. But a lot of the art community has actually moved to Mastodon. Mastodon is not quite as easy to figure out because you have to pick the server you want. And then like you may not see everybody if you're on a small server.

I do, I am collecting people who post about R, using R, of course, because Mastodon does have an API. And I'm querying some of the largest servers to see who's posting with the R stats hashtag. There is no algorithm. So, so like things won't be surfaced for you. You have to be your own algorithm. But if you, if you post, if you follow a few people who we post a lot, you'll, you'll get there. It's, it's a little more work, but it's yours, you know, it's yours, which, which I really like.

And, and a lot of the art community is there, you know, the political community is not, mostly not there. The news community is not mostly there, but R is there. So, so it's a good place to be, I think.

And kind of what I see there is that the, if you like the nice kind of collegiate, you know, keeping each other informed and sharing tips and tricks, that side has come across from the kind of old Twitter, but without the noise and the, the other nastiness, you know, it seems to be quite a nice place to be if you can get hooped into it.

Yeah, I think it is. I mean, there's a, it's, it's a very, it's a very inclusive community. And, and I mean, also like there are a lot of cultural norms there. Like if you post an image, you must have all text there or, or people will complain to you. And I, I, you know, I have to admit that I wasn't really good on about that on some of the social networks. And now I've been really training to do that so much on Mastodon that I do it everywhere else also.

Keeping up with a fast-moving field

That's a really, really, really good question because it's, it's completely overwhelming. You know, I remember when I started covering technology, I was covering the mini computer industry for those of us who remember that, you know, it was like digital equipment, data general and prime, computer, and maybe Wang. And, you know, once every nine months, you know, they, they announced like a new, a new model and it was like a big deal. And the pace of change and innovation from when I started doing technology journalism is just amazing.

So I would say there are a few things, you know, I try to follow people who, you know, I respect and who are, who are really good at it. You know, some of the package developers, you know, when I was doing this for a living, I would try to think about, well, you know, like it was a little different than just like what was good for me. It was what was good for my audience. So there were certain things that are useful in the art, really useful in the art community that were not useful to my audience.

And I would say, you know, for yourself and your work to first, you know, to, to, to look at things and say, you know, how could I use this? Might I use this? You know, you go to GitHub and you see what, I, I follow people on GitHub and I look to see, you know, what's bubbling up. I, I will sometimes look at the R, you know, just R stuff on GitHub to see, you know, occasionally I'll see, is anything new popping up?

And you have to be, you have to be willing to waste half an hour and say, you know, wow, I tried this thing and it's not really going to be useful for me at all. And that's, was a lot easier for me to do. And this was my job as opposed to like, most of the rest of you are trying to fit this into your like real job and your real lives.

And you will miss things and you have to be okay with that. But if something, eventually, if something is like super amazing, like you will find out about it, even if you just casually follow social media, because it'll bubble up. LinkedIn is another, I know there are some people who don't like LinkedIn because it's a little bit self-promotional. But LinkedIn has a lot of good R people also. And it has an algorithm. So things will bubble up if a lot of other people like it. I'm on both Mastodon and LinkedIn very regularly.

Domain expertise and data science

It weighs in really heavily, both with generative AI and also with data science, because I, data journalism, because I, it is just, if you're a data scientist, and you're working on a journalism project, and you're not a domain expert, you need to partner with a domain expert. I mean, you just have to, or just you're going to make terrible, terrible mistakes.

Because I understood our media business so well, I understood all the weird quirks that we had in our Google Analytics implementations. I understood, like, oh, like, you think this is important, but actually there's, it's not necessarily as important as you think, because blah, blah, blah. I mean, because I had the domain expertise, it made me a much better data professional on my job, even though there are a million people who could code better than me, who have, who have more training about data science than I do, but they couldn't compete with me and my knowledge of the business.

I mean, because I had the domain expertise, it made me a much better data professional on my job, even though there are a million people who could code better than me, who have, who have more training about data science than I do, but they couldn't compete with me and my knowledge of the business.

And where we are, where that was, then we are with generative AI now. I mean, yes, there are a lot of people who've been working on machine learning for 15 years and they're experts, but this whole thing of how you use generative AI, well, almost everyone is at the same place. I think if you're young or mid-career, and you're interested in, in that, especially if you're somewhere else and planning to move to that, this is your moment, in my opinion, get really up to speed as much as you can and go for it. Because I just really feel that's where we are now, where I was in my career in 2013, 2016.

Tips for making data stories more interesting

You know, going back to my days as a business writer, and you know, it's really hard when you love data to not want to like, chuck all the interesting numbers, like right at the top of your story. And I kind of learned that you really kind of have to spread it out. And you can't put, you know, three data facts in your lead, and you can't put a ton of numbers in your lead, because sadly, everyone is not like us.

I just have to keep reminding myself that it's like, if you're talking to people who are not data people, and you, you know, you're sending them like an email or a memo or like a dashboard or something, you really have to limit the amount of numbers that you put in at the very beginning, you need it's like your lead, you have to draw them in, you have to tell them this, you have to start telling them this story with one compelling thing, not like five data points that they have to sort of use their cognitive energy to figure out, well, what's more important here.

You know, that's why the graphs where they, where they, you know, highlight the important things and gray out everything else are, are, you know, Financial Times does that a lot. It's, it's really compelling, because people don't, people who are not data people don't want to use their cognitive energy to figure it out. It's like, you know, we love that. That's like the fun, like, you have to, like, oh, my God, I have this big data set, like, what can I see in it? What can I do with it? Let's go play with it. And oddly enough, so many other people are not like that.

Career setbacks and unexpected paths

It's the one that I talked about at the beginning when I talked about, I mean, I really, I mean, my dream was to be a political writer and, you know, I studied political science and journalism in school. And when I found out that the other guy got to be, you know, the Washington reporter, and then the other guy got to be the state house reporter, and I didn't, I was really crushed. I was crushed.

And then it turned out, like, I started covering business. And that was just at the rise of the mini computers in the Route 128 belt in Boston, you have to be really old enough to know that that was a huge, you know, world center of technology in the 80s. I'm like, this is really cool. I'm having a lot of fun with this. So, you know, my, my whole career ended up turning on this, like, sort of weird decision by one guy who's like, and I mean, I will tell you, in the 1980s, for an edit, for a male editor to say, I want the woman to be the business editor, because she's better in numbers, and the guy can be the Paul. I mean, that was like, super not usual.

But it, you just never know. It turned out to be a great thing. And I had a career that I really loved.

Deciding to retire

It was really interesting because, you know, my friends who retired said like, when you're ready, you'll know. And I'm like, how do you know? It's like, you'll just know.

And I was thinking of retiring next year, next spring. Uh, and then, well, for those of you who are, who know about web analytics and Google analytics has their new version, Google analytics for, and that was kind of a very high burnout project dealing with that. And so I was, I was, I was talking to my boss about like, after the GA4 project is done, like, I'd like to take, can I take maybe like four weeks off to rest and recharge and then like get ready to, you know, like in August and then get ready to, and he was looking into that with HR and then it was Memorial Day weekend. And I was thinking of, okay, like if I, if I rest and recharge, like how do I want to reshape my job? And like, what do I want to look like? And then I'm like, I don't, I'm kind of done.

So like, you know, Thursday of Memorial Day weekend, I was like thinking about how I'm going to reshape my job. And by Monday I'm like, yeah, I'm retiring. You know, to, to say it's, people laugh. It's like, like, how do you know you're in love? Like, you just know, it's like, there's not like a, but you do have to know that you're going to be ready to give up your professional status.

Projects and plans in retirement

There definitely, there definitely are. And I mean, I want, I do want to spend more time getting serious about learning generative AI stuff because I just think it's, it's fascinating and super cool. And I would, you know, I have some coding ideas in mind about local, local projects, like dealing with local government information, what you can do with that, whether it's, it's government meetings and agendas, whether it's, it's campaign finance.

I, I want to spend a little more time on my, I have a like tiny little neighborhood blog that right now is, is mostly, I mean, like not my city, but like my city council district. I've done some stuff, it's stuff with events. So like, I, I like, I've been like web scraping events to like put in a, in a searchable, you know, Shiny calendar and sort of stuff like that.

My new role was supposed to be like 75 data, 25 writing or 80, 20, something like that. And I did do some writing, but as the Google analytics project came up, it's like, I just didn't really have time to write anymore, which made me sad.

But, you know, to, to have the luxury of, of just learning something and then being able to write a how to about it. You know, I actually, I, I spent about six months learning on my own time, taking copious notes. I mean, this was like 2012, 2013. I mean, there weren't the kind of resources available that there are now. And I, I ended up with this like, you know, 60 pages of my own notes. And then I'm like, I have all these notes. Like, why am I not writing this for Computer World? And that was my Computer World beginnings guide to R which ended up being like one of the most popular things we had in terms of people being willing to register to download the free PDF.

Career advice

I mean, I, I would, you know, people always, this is very counterintuitive, but, you know, people would say like, you know, imagine where you want to be in five years. I'm like, really? You know, five years ago, I had no idea that there would be such a thing as generative AI, you know, it's just, I don't think in this field you can, you can structurally envision so concretely where you want to be in five years.

I, I would suggest, think of the broad things you want, you know, I want to be doing cool things with data that interest me and also X, you know, and then work from there. But, you know, I mean, the, the pace of change is just remarkable and, but I think if you, if you have a basic foundation of, of what you think you want and you're good with change and you like learning new things, you'll be great.

I, I would suggest, think of the broad things you want, you know, I want to be doing cool things with data that interest me and also X, you know, and then work from there. But I think if you, if you have a basic foundation of, of what you think you want and you're good with change and you like learning new things, you'll be great.

Thank you. And thank you so much, Sharon, for taking the time to join us today. I really, really appreciate it. And congratulations to you on your retirement as well.