In this episode, we talk about new content types on Stack Overflow, when it's appropriate for managers to make technical decisions for their ICs, and Paul's experience getting a new tool online with Google's Cloud Run.
EPISODE NOTES
To start things off, we talk about the launch of Articles, a new content type for Stack Overflow Teams that lets you write longer pieces. Sometimes it's best to share knowledge through Q&A, but other times you've got complicated, narrative, DevOps recipes or a policy paper and FAQ. Now your knowledge artifacts can all live in one place. "The FAQ is the great folk form of the internet" - quotable moments featuring Paul Ford. If you're interested in another cut at this old saw, Mailchimp.com/developer is Postlight's take on what developer docs should look like. Sara is convinced it's all about the left nav. Speaking of convictions, a conflict is tearing Sara's home apart. Ben and Paul step in to save her marriage. The question at hand: should managers of developers EVER make technical decisions? Finally, Paul talks about his experience using Google Cloud Run to build a fun little tool called Ephemeralist. It pulls in random images from public domain collections hosted by museums and archives. Use it to take a break from the negativity of social media or the news. Also, revel in the joy of Paul's neologism, the Browseulator. It recently brought me this little gem.
TRANSCRIPT
Ben Popper This fly has been landing on my head the entire podcast! Err!
[INTRO MUSIC]
BP Hello everybody and welcome back to the Stack Overflow podcast! I'm Ben Popper, Director of Content. Hiiii Sara! Hi Paul!
Paul Ford Hey!
Sara Chipps Hey Ben. Hey Paul.
PF Hello audience. Hello friends.
BP So we just published blog on the Stack Overflow blog because we have a new feature for Stack Overflow for Teams. It's called Articles.
PF Aaaarticles! Oh that's--wow!
BP As a former journalist, I approve. I don't really know what it does. But since it's called Articles, I like it.
PF I read about it. Here's what I think you've done. I think that you looked around and you saw social networks like Twitter and Facebook wall posts and also just sort of where you guys been going with questions and thought to yourself no one has ever just put a text box up [Ben laughs] where people could long post...
BP Mmm...yeah.
PF And so like, yeah, like almost like a like a log of web thoughts, like some sort of...
BP Hmm.. a web log.
PF Some sort of....blog. It seems like you've added this revolutionary new category of internet content called blog. Now, let's be clear, this is for Teams. Right?
BP It's not for the public site.
PF Let's separate that for the humans. Even though I am I'm a happy part of this marketing product. It actually is really good for people to know the difference. There's Stack Exchange and Stack Overflow, which are like the full blast...
SC Public.
PF Here's 20 billion questions. Publicly moderated creative commons, open to all.
SC Fire hose. Yep.
PF And now what and teams is like for me and my buddies at little places like Microsoft, I mean, I just like, like, where, who uses Teams? And how...
BP Yeah, yeah, so it's an internal version of Stack Overflow. Actually, funny story, one of the first people to come to us and ask us to build it was the FBI. We haven't written that story and put it on the blog yet, but it's gonna be a fun story for another day, but they don't believe they are currently a customer. But we had other folks approached us like Bloomberg, Bloomberg and Microsoft and say, you know, all of our engineers are using Stack Overflow. But there's lots of questions they have to ask and they can't ask in public because it's about proprietary code or code for something that is still, you know, it isn't hasn't been pushed live is not in prod. So.
SC Yeah, so, so a little bit a little bit of how the bread is made. Sometimes people forget that. And they post code from their internal code base on Stack Overflow. We get a lot of emails.
PF But this is, uh, this is normal. You're, you're like, I have 5000 engineers or 500. And they use this thing all day, we should have one inside, you know, who's one of the biggest users of the Wikipedia software is the CIA, right? Like they've got their own Wikipedia.
SC Wow, fascinating. Well, yeah, I think that's actually related to the Articles feature because one thing people are using their internal Stack Overflow for is a knowledge base. And sometimes that knowledge takes the form of q&a, but other times they're things like, you know, team updates or team policies that you might want to have in an article format.
PF Like, where would use, I would think so the first thing that would come to mind for me is like really complicated narrative, DevOps recipes like sort of, I've got it. So you got to do this thing. And you got to get one of these stood up, and it's not and there's a lot of decisions to be made. It's not something that can be automated. It's not just a bulleted list, like, what are the use cases for blogging? Let's just call it what it is inside of Stack Overflow for teams.
BP I think like one of the things that I learned a lot about when I came, which I didn't know about before was documentation, which has like a really rich history in the world of software engineering, how do you do documentation well, and then there's documentation, then there's the company Wiki, then there's then Stack Overflow came along and became sort of like this public corpus and this public brand. And so internally, as people get bigger and bigger with their Stack Overflow for teams, they want to have everything in one place. They don't want to have a confluence Wiki over here. And then you know, like, their own homebrew, you know, list of documents over here and then the question the question and answer database over here. They want to have one thing. And so this does in just a little bit more towards having like sort of CMS capabilities. When there's like different kinds of knowledge artifacts, you can create.
PF Those long answers. are often a little kind of out of place, and you kind of want a place to put stuff. That's a little bigger and a little broader.
BP And I liked what you said about like, if there's a narrative to it, like what if there's some subjectivity to and you want to walk people through the rationale, maybe this is a better place to do that then the q&a where we try to keep things very, you know, objective.
PF Look, the FAQ is the great folk form of the internet, like it's how Usenet discussion boards and certainly, asking questions and getting answers online is the best way to sort of meta moderate a community because it keeps people from asking the same question over and over and over again, and the stack tapped into that right. And then that, but there is also this need for kind of broader storytelling and institutional knowledge, which, yeah, it makes sense. It makes sense. I mean, I'm gonna continually mock your organization for adding a text box in 2020. [Ben & Sara laugh]
SC That's what we...it's gonna be on the podcast.
BP It took us longer to make than dark mode, Paul. I mean, this was some serious engineering that had to go into this.
PF I'm gonna tell you that seriously. Like, this is the thing none of it's that complicated. Like if you were making a 3d color picker that also allowed you to, like factor in, I don't know, like the phase of the moon, we'd all be like, oh, and then nobody would use it. But people are gonna put words in the textbox. And then the question is what happens next?
BP I just have one question, Sara, which is, as a former journalist, and now the director of content, one of the first things I said when I arrived was, why don't we do articles in the meta, like dev two, we have so many great people here that have these amazing discussions. And recently, we had even been bringing like interesting Q&A artifacts over from the software engineering Stack Exchange, where it sort of evolved into a discussion with the comments and multiple answers. We bring that over to the blog and sort of write a little bit of context around it, and people love having that discussion. So why wouldn't we just do articles on public StackOverflow?
SC Well, I think that's a good question. And it's something that we've talked about. I think something that we've to remember is that Stack Overflow as it exists is something that is that works, right? It's a community of people that really want to have the world's largest knowledge base of programming questions, and putting something very different, like articles into Stack Overflow would really upset that ecosystem. And the way that for example, you know, our moderators spend a lot of time moderating questions, and for them to moderate long articles or subjective writing, that puts a lot of work on their plate. So we wouldn't want to introduce something like articles without being really thoughtful. I think that there's definitely ways to scratch that itch that don't have to do with changing the format of main Stack Overflow. So, I think that's how we're thinking about it. We don't want to throw a bunch of extra works on our moderators.
BP Right. Right. That makes sense. Okay, thank you.
PF It's actually fascinating, right? Because this is scale and inaction in a medium size or large org. There's very low risk to putting that text box up there, and just go Like, hey, have at it, you know, if you've got something longer you need to write obviously put it here we'll give you some hints as to best prank drives as to how to organize it the tags work and so on at the scale of Stack in public, it's almost like you're saying, oh, we're also going to add a text box to Instagram where you can add your narrative like you're blowing up this like very clear mental model about the Q&A model and it really like when you're talking about hundreds of millions of people instead of a couple thousand Max and it really is just an order of magnitude differences. That's kind of where product happens is in that delta. And just saying it because it's sort of fascinating, right? Like, yes, of course. Oh my god, you drop the simple text box. You add blogging to Stack and kaboom!
SC Yeah, kaboom, yeah. [Ben laughs]
PF This just popped into my head and it's just it's just relevant to this world. If you go to MailChimp.com/developer. MailChimp got in touch with my company Postlight early in the year, late last year and said, Hey, we're going to relaunch our developer website, and there was a lot of product change happening,.They were bringing their transactional API for mail into this sort of MailChimp fold. And they wanted the API docs and all of the developer docs and all the developer SDKs and like, it was a kind of a fascinating project. Because docs in general tend to be kind of a lot of stuff. Like just like, we got our stuff here and hope you find your way. And in this case, it was very much like everything's on GitHub, you can you can put in a request. It's all very interactive. If you're in JavaScript, and you say, I want JavaScript code samples. It remembers that, all the way through as you're using it. So anyway, MailChimp.com/developer, we think it's a pretty good example of how developer docs should look and be.
SC Fascinating.
PF Yeah, I say that both to market the firm where I am co-founder but also like, I didn't do a thing you know, other people worked really hard mostly at MailChimp and with us in support.
SC Developer docs are hard. I think one thing I've learned to appreciate really strangely, is the left nav, in developer docs, it looks like you guys nailed the left nav. [Ben laughs]
PF Listen, nobody's fooling around on the left nav. So they were a great partner. And and really, we're very proud of that result. Because it's a lot of developers using a lot of code when it comes to MailChimp.
SC Docs are so hard. Yeah.
PF Ooof they are. And you know, they're a system they need to kind of light up and be dynamic. You can't just drop them in the CMS or hope that the swagger output is going to be enough if you have to, you have to think of stuff through now. Developers demanding.
SC That's not..that's off from my experince.
BP They're pretty laid back easygoing customers.
SC Speaking of which, I have a good, I have a good thing for us to talk about that is tearing my home apart.
PF Oh, no.
SC It is threatening our very life and the ability to coexist.
BP Sara, Let us save your marriage. Let us do it.
PF I'll take that cat.
SC No, it's not tea. So should managers of developers ever make technical decisions?
PF Oh god, this is classic developer thinking. Ever, ever is the magical word. This is like when this was like eight years ago when every developer on Hacker News decided that deadlines were impossible and must never become part of programmer culture. And then that one went away in like six months, because you know, you actually no one got anything done for six months, I would say 80% of developer time 20% is spent coding 80% is spent convincing their managers that they should use a lisp or that everything that we know about management is wrong. [Ben & Sara laugh]
SC Yeah, I think actually, we're not we're not--I don't think we can actually answer this question because none of us are ICs. It's not really fair. Yeah.
PF Here's what I've learned in 850,000 years of using computers is that it's a great question. And like every single engineering related question ever, it is proposed as a Boolean, and the answer is like a 5000 word text box.
SC Yeah, that's probably right.
PF It depends.
BP Yeah, it depends. I mean, I, this reminds me you know, it's like it's like Office Space. There are managers who come in and manage you who are lording it over you? Because they're your manager? And they don't actually have the knowledge? And then I'm sure Sarah, their managers, right, who were great ICs, who worked their way up, who might have engineering knowledge and could say, Hey, I think this is the right solution. In fact, I'm deciding for us so we can get some consensus, and they know what they're talking about.
SC Yeah, I mean, if there's something to be said, though, for being like really close to the metal, right for like being in the code every day, which is something you can't do as a manager, you know, it's sometimes not having that context you can lose the ability to kind of make those just not maybe the ability but if you don't have the understanding of the day to day just because you're not in it all the time.
BP Yeah, no, that makes sense.
PF Let's let's argue this from the coders point of view, which is you have, manager, once again swooped in and making decisions that are informed is not really what we're talking about. Right? What we're talking about is you've decided on a platform, you've decided on versions, you've decided on a front end, you brought in a vendor who've told us we have to use this instead of that we can't use the database we know. And we don't know why. And you're telling us..
SC Yeah it's because you had a nice dinner with them and they gave you some swag. And now you want us to incorporate it in the platform.
BP Right, you've already rubbed us the wrong way, five or six times. Now back off and let us do what we do.
PF I mean, this is, in truly really not truly non technical organizations. This is very dangerous, right? Because the CIO or CTO is a vulnerable human being without a lot of time and people are constantly telling them what the right solution is. And then they manage by Fiat and they're just like, Okay, well, we're using Oracle. [Sara & Ben laugh] I'm tired of your guys.
BP Shout out to Oracle, great podcast sponsor.
PF I don't know why it's Paul Lynn from [inaudible]
BP Sara, in your home, I know, obviously, you're a manager and director of lots of folks is Satoshi an IC or a manager or what is the Satoshi do?
SC Yeah, yeah an IC.
BP Okay, got it.
PF Yeah, but let's be clear, it's Satoshi. Sara's significant other is kind of a super IC, like he's very experienced and knowledgeable and has built large platforms. Not that I know who he is or anything about him.
SC [Sara laughs] No.
BP Bitcoin is a very large platform. [Sara laughs]
PF God it is.
SC The thing is, though, there's something to it here, right? Because to that, especially because being an engineer, there's a path of you know, at some point, you need to decide, am I going to be a manager? There's very few companies actually, that offer a path for engineers that isn't management. That become more and more specialized as time goes on. So I think that's an interesting challenge, too. When you're a manager and someone on your team has a really deep technical knowledge, what is the value that you bring to that person?
PF I've seen this from a zillion different angles, because especially because my company works with a lot of different orgs. And so there's a couple I think smaller mid sized engineering groups for this specific product focus tend to be in a pretty good place, right? Like they have tools that they use. They are agile enough that they can get upgraded to a relatively recent version of the software, like the ones that are in a tough spot are the really big ones that are running like, you know, Red Hat Enterprise Linux from 1999. You know, and an old version of Java that has, you know, groovy on top of it, and they can't really get much done. And then then somebody comes in with a new solution and and swoops in. And it's just a lot of damage, right? Like, there's a typically tight coupling between technical decision makers and engineers in small to mid sized stores.
BP From my perspective, as somebody who used to be an IC, like I was a feature writer for a long time. So it was my job to generate the ideas to work on them for like weeks or even months, independently, and have to like, get to a place where you know, it was something that would be worth publishing, then I'd have to write it and then at the end, an editor would come in, and if the editor started rewriting stuff that I felt was more about the color and the texture, I would get very angry and eventually after a while, I would no longer work with that editor. But you know that like some editors I ended up having great relationships with the people who helped make it clear tighten the transitions, better the structure, and maybe they would even come in early. And if I had an idea they'd like work through the idea with me like Paul was sayin, like, what platform we're going to use, what vendor we're going to use, like the high level where what are the building blocks we're going to use to make this stuff. But yeah, once like, I can imagine, like you were saying, You're not in the metal, you're not day to day on the code. If they came in and tried to sort of like change the style of writing, which is like my day to day, then, you know, our relationship wouldn't last very long. So maybe, to say there's a spectrum of like a good manager to ICU relationship and a bad manager to ICU relationship. You know, there are people who approach it in a certain way that is productive and people approach in a way that's unproductive, right?
SC Yeah, I think that's right. I think what you're looking forward to is someone that is open for discussion and isn't swooping in with a technical decision and saying, here's what we're going to do, right because in this case, it's often helpful as maybe like a consensus builder or someone that can talk to folks on the team to figure out a good direction, but maybe not the most helpful to come in without context and say, here's the direction. I think we should go. Totally, I think that's it.
PF I'll give you a developer's point of view example. I was once working on a project 10 years ago for a very large cultural institution and the CTO went, Okay, look, I know how these things go. We're going to pick a content management and asset management platform. We're going to spend half a million dollars and it needs to probably be like this. And so she used that to narrow down the decision. And from there, they picked this thing that was an absolute boondoggle, and everybody received it and went, well, there goes my career, like no one cares about this. It's a really bad tool. It's hard to work with. It looks like Microsoft Windows 3.1 and I hate it. You really just wrecked like 10 people's lives and took away Their future with that decision? Yeah, if she said Drupal. Everybody be like, Alright, alright, I'll be alright. I can figure this out.
BP Yeah. Sara, I liked what you said about like, right. Coming in and maybe asking questions about a decision. Like, why did why did you do it this way? If we did it this way, would it? Could it be faster or slower and more effective? Or, you know, getting to that consensus as opposed to coming and saying, No, don't like this, we're going to do it this way, or coming in and saying, this is the way we're going to do it. Right, like, let the IC volunteer some of the ideas or start to build some of the fabric and then come in and question it, then it's a dialogue, like you said.
SC Yeah, and I think treating them like the experts to like, Paul, it sounds like in that situation, there wasn't a lot of conversations happening.
PF Oh, no, no vendors all over the place. And you know, just just exactly the perfect storm for incompetence to thrive.
SC Yeah, exactly. And just I think all these things, too, are definitely a conversation, because sometimes people do know better.
PF So really the answer when we go back around, should managers be involved in technical decisions is tightly knit communities with good listeners on both sides. Yes, what we really mean? Sorry, sorry if that disappoints you.
SC Thanks for solving it.
[MUSIC]
BP Love coding and want to learn new skills? Join the weekly AWS series Developers Let's Code live on Twitch. With hands-on demos and virtual sessions, you'll learn new core concepts on some of the hottest topics in cloud technology. Subscribe today at codewithaws.com.
[MUSIC]
BP Paul, I know you did a fun little cloud deploy last week. Is that right? You want to tell us about that? And then we'll wrap the episode.
PF Oh, absolutely. So in my weekend evenings are time for me and my computer as opposed to weekdays and afternoons mornings and regular evenings.
BP Paul, you should move your computer more than three feet away from your bed. I'm gonna be honest, yeah, me that I was troubled. I was troubled. I don't think workstation is the bedroom is the way to go.
PF I live in a small condo, there's not too much you can do. So a while ago, I don't know if anybody's ever used the website, Twitter, but it'll mess your head up. I it's very depressing. Anyone ever had that experience?
BP I've heard of that site, but I try not to look at it. It's like, yeah, your brain to mush.
SC I mean, that's 100% true.
PF Yeah. But I, I sometimes use it. And I found that maybe I want to use it a little bit less. So anyway, I made a really silly tool. And I'm not we'll put the URL in the show notes, because it's one of those autogenerated cloud URLs. But so the tool, all it did was take a lot of JSON data off of different public resources, like the Smithsonian publishes all of their, they publish 11 million different object records in JSON onto GitHub. And that's like, you know, multiple than multiple million museums. There's 2 million pictures of herbs alone. Archive.org carries a lot of old books. And so that's a wonderful resource like, so, taxpayer funded, public. are not for profit orgs that have these wonderful archives, but they're kind of hard to get to, like, you gotta go and search and know what you're looking for. So I just kind of spidered them a little bit and put those public resources into a SQL lite database. There's like a million things in the database, and then it pulls out 10 random images whenever you hit the page, simple thing, like anything when you're dealing with messy metadata and lots of it right, you know, took a couple evenings of messing around, posted some screenshots of it, and people were like, are you gonna put this up? And I'm like, Alright, well, I guess I better and so I was like, I'm gonna use Google Cloud run.
SC Why? Yeah, why?
PF I looked at a lot of others. So the problem I had, my preference would have been Glitch, but my database was already the paid plan for Glitch is like 400 megs of storage, and I need at least a gig.
SC Okay, of course.
PF That would be my preference. It's so easy to deploy on that thing. So I was like, all right, I should learn Google. I've never used Google Cloud, even in anger. [Sara laughs] And so I should, I should at least try it. Like you got to know I've never done Asure I've done lots of AWS stuff. So I was like, Alright, two hours, I'm gonna get a Docker container up and upload it to Google Cloud run, which basically is like put your Docker container over here. And we will listen on a port, and it will work just fine. But what I learned, I learned an important thing, which is that the way that actually runs, so you know, what Lambda functions are on Amazon, right? [yes] they're just functions like pure functions, you give them some input, and you get some output. And like they're not maintaining accounts or anything. And so Google Cloud run is basically that, but it's a container in the middle instead of a function. So now really a full fledged server. It's really just like one port, here you go. But it also is a full fledged web server, because we live in this weird world. So what I learned is that it has, what you get is to one or two gigs of memory, and everything runs in the memory, including all your data. So I made my container which Docker is really confusing. I don't know why people do that to themselves, but boy, you can bundle a bunch of stuff up and I put it up, and it took about two hours and everything broke, everything. I couldn't understand anything but Then you just kind of find everybody's script, you hit Stack. People are like, here's how you deploy. And so by copying and pasting from the internet, I was able to get it up my observation just with all these cloud services, I mean, first of all, they're all just pretending that they're not really just a bunch of servers or abstractions over servers and..
SC Or that they're not AWS actually.
PF Yeah, exactly. Exactly.
BP Deep down. It's all AWS.
PF When you look at Google Cloud, and you look at that left nav bar of all the different services, and it just goes and goes, they just breed new services. And it's tough, right? It's, there's none of that Apple, like, imagine the apple cloud, where they were just like, we're gonna have three products. That's it, and you're gonna want to use them. That's about it. And you'll be like, why I need a web server, and they'd be like, nad, you don't really need that. So anyway, we'll put the link in. It's fun to look at lots of pictures of random stuff. It's still really hard to host stuff in 2020. It's hard to host a website
SC Isn't that funny? It is. Yeah. And so you've you've put it up and you shared it.
PF I put it up. I shared it out. seems to work fine. It's silly. It's a nice distraction.
BP One of my favorite things in life is the cabinet of curios. Are you guys familiar with this? You know, if you were a noble person back in like the 16th 17th 18th century, when science was becoming a thing, you, you would have a room in your house that was just full of amazing like Paul was saying natural artifacts. These are some herbs I've dried, this is a fossil. This is you know, like a fine Falcon or whatever. Just like things that were amazing about the world and about nature. public records of art are kind of like that. And so I always try to put into my feed, my Reddit and Twitter and Instagram feed, like you were saying, Paul, I'll follow some chemistry accounts where they're just like, this is what iron looks like magnified, you know, 10,000 times. This is a video of a crocodile trying to eat a turtle, but it can't because the turtle has a shell and the turtle gets away and like, those things just break up the angry politics and negative tech is monopoly stuff that is also part of my feed.
SC Wow. So what is this pulling in?
PF Images from the Smithsonian image server API.
SC And it's like whatever people are calling on the API. It's not like random, is it random images or ones that people are calling?
PF Totally random, totally random.
BP This is a beautiful bowl. It makes me just makes me happy to look at this bowl.
PF Pictures from the Metropolitan Museum of Art, pages at random from the Internet Archive from books.
SC Military diving.
BP Ancient bracelet from the first century BC. This is what I need. Yes, a fossil, Hmm, that's what I need.
PF You know, where it's really good is you just sit there and you can click on them and they go back to the sources. It's nice when you're on the phone in the in the era of the pandemic, to be able to kind of just browse through some art while you're looking into the screen and people are talking because you can still stay focused, but look at stuff.
BP Another thing that I had done like this, which I'd recommend for just sort of like mental a little bit of peace. I signed up for Artsy, do you do you know that surface at all? It's like an art recommendation service, you just click on three or four or five things that you like, and then you refresh it once a day and it just serves you new art and if you like something you like it, and then over time it like learns certain tastes. And just like, every day, I would just like check it once or twice a day. And then I would save my favorite ones and make them background on my computer and then have it rotate through the backgrounds, you know, 50 times a day or something. So you're always getting this like nice, fresh look.
PF See? That's what you need lots of variety and images and...
SC Images! Random images from the internet. What could go wrong?
PF Well, no, that's the actual that for me. It was part of the point, right, which is I'm going to use big cultural institutions.
SC That makes a lot sense. They have their own filter.
PF Yeah and just monitor my sources a little bit, right. So at one point, when I was messing around with this, I imported all the zines from the Internet Archive collection, which is like a lot of UFO stuff, and a lot of white supremacy. And I'm like, you know, I just don't want that in my world. I have a choice. I have a choice here. And then I'm like, Oh, you know what you can actually elevate the American Museum of African American History, you can elevate that you can randomly bring that up. So you see more of it. And so I'm like, wow, you know, I could shape a little world that I would like to see. And spend time there. Instead of watching people yell about the same thing over and over and over again, which God bless. I tend to I tend to agree with people in the world, but I'm just like, I can't see it anymore. I know what's gonna happen. I know, we're all going to go vote in three months. So I gotta just, we're just got to get through it.
BP Yeah, thank you for building this and sharing your adventure with the Google Cloud run.
SC Scale of 1 to 10. If I'm building a project, how much would you recommend?
PF I gotta say, for deploying a container. You know, once you get a few of the magic words, right? To me, the hard part is Docker. I find Docker really confusing. I don't find containers confusing. I find Docker confusing. I'm just like, oh, an image container. Because I if you use it every day, I think it's like Git where you're just like, it's fine. But when you don't use it every day, it just the interface to containers that Docker provides, always requires a lot of wrapping your brain around and the other thing every cloud service says this problem but my god, where am I logs? Where are my logs? Where do I put the things in the folder so that they appear where you say they will appear? And where is my stuff? Is is like the foundational question, every single new thing around containerization hosting or Cloud Platform should just have a single document called where's my stuff? And where do I see it? Like environment variables, all of it?
SC That's great. Yeah, I think that's right.
BP You built this thing. I'm seeing ephemeralist, you know, dash x four seven p, but then in my browser tab, it says browsalator, which maybe is a word you made up I love it.
PF Browsalator is a word I made up. Ephemeralist is I think what I'll call this thing.
BP Yeah. And Ephemeralist tab is called a browsulater.
PF That's right. Browse you later!
[MUSIC]
BP Alright, y'all, it's that time of the week we're gonna do our lifeboat. This one just has for me as the former English major, a nice ring to it. Calculating draw down in pandas.
PF You're going to be soooo disapointed.
BP How do we know how the pandas doing and how do you make the panda feel better right, no? Okay.
PF No, no. Let me share this link and we can jump in a little bit, I guess for those who are not aware, to what is a panda and why would you draw it down? Well, we can say lots of funny things.
PG Well, I mean, okay, pandas is a scientific and numerical processing library in Python. And it's the question gets to it. it's like data frames. Like it's really good for dealing when scientists need to do spreadsheet things. Pandas one of the things they reach for. Right. Gotcha.
BP So they're calculating the end the heat death of the world here, there's a drawdown. It goes in a sequence, they want to figure how to calculate it somehow. Okay. I gotcha. It says profit cumulative. Okay. So they're looking at a data frame here for some kind of earnings report or something like that.
SC Ooh fun.
BP The question was asked two years ago by user 8491020. And we have an answer from the same user. So there you go.
SC It happens.
PF They answered their own question!
SC Yeah. It means no one else could help, but they finally figured it out. Hopefully they can help someone else. Other people did answer but the user who asked the question and said this, the simplest solution I found apparently had the best solution. But the knowledge was shared. It has been seen by 6000 people and given 21 upvotes, so the important part...
SC Wow so he got his own life boat?! I wonder if this is a first for us.
PF I mean, listen, save yourself.
BP I think this is a first on our lifeboat. Yeah, you saved yourself. Exactly.
PF Doctor, heal thy self. Programmer, solve your own problem.
BP Alright, user 8491020 Congratulations on asking a question answering question and getting a lifeboat, shout out to you. Alright. It's been a great episode we learned a lot. If you're interested in Articles, you can check that out on the blog if you're interested in Paul's ephemeral stuff we'll that in the show notes. If you're interested in the debate over managers and ICs throw us a line on Twitter...
SC Or yell at your manager!
BP Yeah, yell at your manager! Podcast@stackoverflow.com, have them listen to this episode. I'm Ben Popper, Director of content here at Stack Overflow. Thank you for listening. You can find me on Twitter @BenPopper.
SC And I'm Sara Chipps. You can find me on GitHub @SaraJo.
PF And I am Paul Ford, co-founder of Postlight. Check out mailchimp.com/developer, it's work we help them with that we're very proud of and it's relevant to this audience.
SC Good docs!
BP Right.
PF Yeah!
[OUTRO MUSIC]