 So, we'll assume you had a good weekend. All right. Well, I guess I got to stop chatting and start getting going here. Okay, good. You can hear me. Great. I'm glad that you can hear me. All right. I'm glad everybody has shown up. Thank you for coming today and welcome to our little mini lecture today about how you should start with SAS, SAS software. This is the last of my little lecture series on case studies and SAS integration. I'm a big SAS evangelist. I'm always getting in trouble with SAS because I'm always talking about them and they're not open source, right? So I'm like a free sales rep for them. Really, they don't pay me. In fact, I get in trouble with them for talking about it. But why do I talk about SAS? It's really ubiquitous in my field of public health. And so I'm happy that you showed up today and I can share some of my knowledge with you about SAS because I've been using, I've been a user a long time. So this is about how should you start with SAS? And I'm just going to be really philosophical with you. You can be born again in SAS. So I'm not really that young. Like I'm in my 50s and I've been using SAS for 20 years. But I feel born again with SAS and you'll see why. So there's some challenges in the future with SAS. I'm just not going to lie. When you're an old software company and your software has been around a long time, you're going to have challenges as things like the internet get invented. But there's like something I really see that's really awesome in the future for SAS that's different from the past. So you really want to sit and listen to what I found in my research. So I'm Monica Wahee and I'm a data scientist sort of in the healthcare space. A lot of you know me already, but I'm an epidemiologist and biostatistician. And so if you're new to learning SAS, so first of all, if you're not new to learning SAS, then you already know SAS. You kind of know what I'm going to be talking about, but you want to stick around to really see what everybody's saying about SAS now. Because if you've been using SAS, that means you've maybe been using it since the maybe 80s, maybe 90s, maybe 2000s. Some people have been using it that long and you have a really different view of it than when you're just starting with it. So when was I just starting with it? I was just starting with it in the early 2000s. If you're just starting with it now, you might be able to relate to the story. Okay. So when in about 2000, if you went to a College of Public Health and you went to learn SAS, this is what would happen. Okay, is you would take a biostatistics class that would meet in a classroom like Monday, Wednesday, and Friday. And then you'd have a lab that would meet on like Saturday or something for a long time and it would meet in the computer room at the school. When you sat down at that lab, there would be somebody like a teaching assistant at the front who's a biostatistician, and that person would teach you like how to log into the SAS server. So you'd have this command prompt, it would be kind of a black screen, and you'd be typing out, you know, there'd be a cursor and you'd go into SAS. And what you could do is you could create code and you could launch code. If you did that, you had to make sure that the code was mapping to the right data sets because the data sets would be on that server. So you'd be creating code on that server, running the SAS code against data sets on that server. And every time you did that, you know what you would have to do? You'd have to look at the log file, right, because you'd have to see if the code you executed worked. So you'd have to call that up. Remember this is all command prompt. And also, what about your results? You know, let's say you ran proc-free, you look for two-way frequencies. Well, you'd have to also type that in to call that up. Now this, if you, you know, say history programming, that's a very normal thing to have to do, you know. The problem is in the year 2000, we already had it like Microsoft Access and I was running Microsoft Access databases. So these are databases that kind of look like, you know, Microsoft Wordbooks, or Microsoft Excel, like today. When I was learning this and I was going, oh, you mean we don't have any windows or like, you know what I mean? Like because I had these expectations I got from Microsoft Access. And I was like, are we really going to keep doing this? I mean, isn't there going to be a more Microsoft Access-y like version of SAS? Well, the answer was kind of, yeah. I just needed to wait about a year or two when they came out with PC SAS. So PC SAS, you installed on your computer, you can even install it on pretty powerful laptop. And when you would run it, you'd run it as an application and it really kind of did look a little like Windows XP. Oh, I see someone in the chat. Oh, hi, Ebenezer's in the chat. Everybody say hi to Ebenezer. He is my fan and colleague and a great digital marketer. If you ever need digital marketing and science-based contact Ebenezer. All right, so PC SAS was like wonderful because I was so like done with the command prompt stuff because I wanted to look at the log file and the output and everything at the same time. I wanted to look like Access. And so PC SAS was all Windows-y and looked like that. And so I was really happy, right? But think about it. It's PC SAS. So what happened to the server and this whole networking idea I just was talking about. Remember when you go on the command prompt and your data is on the server? Like everything's on the server. It's this beautiful SAS environment. And so I was like, okay, my little SAS environment on my laptop, like how am I going to do that with like a cohort study? So that's what I was thinking back in 2000. So just to recap, there was this weird command prompt stuff going on. They gave me PC SAS at school. I loved it. It looked really Microsoft Access-y, Microsoft Word-y and stuff. But I was wondering what's going to happen if you try to network this, okay? So if you kind of remember the evolution of the internet, you know, the internet was there in 2000, but you really couldn't put data online. You really couldn't do, like my friends had a startup and they were selling computer parts and they wanted to automate people like going to their webpage and looking up the inventory, like live. So when they sold stuff, it would update and they had the worst time. And this must have been before 2003 because I was living in Minnesota and this was happening in Minnesota. So then I moved to Florida, which was totally a mistake, but that's another story. But anyway, so, you know, data handling on the web was just not good. So people didn't really have databases on the web. So this early period of PC SAS, you know, I never really thought about SAS serving to the web. I was like, why, you know, it never even occurred to me. But now think about it, that's the big deal, like serving data to the web, interacting with the web and SAS and everything. So when I was back then, I was going, okay, what is the future of this? And I wasn't saying it like in a negative way. I was just saying, like, we clearly need the analytics, like SAS analytics is amazing. So how are we going to get that analytics by the data and by the team without a SAS server? Because I can't build a SAS server on my laptop, right? So now fast forward is 2023. Let's say you're newly learning SAS now. You might have like similar kind of thoughts because even though maybe you're learning SAS Viya or maybe using SAS ODA, the free version on the web. Let's just admit it, it really doesn't handle like modern programs, right? It really feels very legacy, you know? So how are we going to do this now? How are we like SAS Viya exists on the web, SAS ODA exists on the web, which is amazing. You'll hear me go on and on about that, how awesome that is. But like, okay, now what are we going to do? Like now SAS has met us, you know, met R and Python and SQL and stuff on the web. And now how do we work with it? Like, what do we do? Like SAS wouldn't use literally for many things for decades, okay? But some of those uses now, I'm so you'll, if you showed up to some of my earlier lectures, you'd find that there were components that were so amazing and SAS 20 years ago, that now if you adapted them, you're kind of like, oh my God, how do we get out of here? Because even though they're still awesome, it's like the world around SAS changed and now there's something like way better. And you'll see this when I go through the research I did for this presentation, what I mean by that. Well, before we continue, I just want to make sure you know about that I'm holding free online workshop and application basics for SAS integration. So what is it? You can see the sessions on the slide. It's three sessions. They're going to be like two to three hours because I don't know how many people show up. It might be longer, so you'll budget for three hours. So, and it's Monday, Wednesday and Friday next week. And it starts at noon Eastern time, because I think that's probably the best time for everybody internationally. So what's going to happen at the workshop? The workshop is based on an actual online course that I developed that's part of my data science mentoring program. So if you want to elevate your skills and data science and produce portfolio projects and actually do some real world applied projects, you're going to want to check out my mentoring program. But what this this workshop, it just takes one of the courses from that mentoring program because there's a foundation courses. And I'm actually going to teach it to you. So it's an online course. You can go through it on your own, but I'm going to teach it to you so you don't have to. And what I'm going to do is talk the courses exactly like mainly about applications. So if you're like me and you came through like master's public health or you learn biostatistics, whatever, you probably didn't learn about how computer applications are built or about business applications. And so it becomes really difficult then if you want to do application integration. Oh, that's great Michael that you're interested here. I'll follow up with you afterwards and we'll connect. That sounds good. Wonderful. Yeah, so, you know, we as we move forward in health analytics and public health. We the word application now we're going whoa somebody's using their diabetes application somebody else's, you know, saving their steps in their steps application. You know, how are we going to take that data. Of course we're going to want to put in SAS right eventually like if you want to make a model. How are we going to put SAS in that pipeline. And that's the it's sort of like the agony and ecstasy of SAS. So if your legacy SAS usually you're going to start crying. And if you're new SAS you're going to start cheering because I have bad news for old people new people unfortunately. But anyway, if you just want the news, please sign up for this online workshop. It'll be a blast because we'll all get to discuss the thing about making application pipelines. Yeah, there's a lot of technicality do it but it's really design it's really thinking. Like I'm an epidemiologist and if you've heard of like Bradford Hill causal criteria, you know, it's really doing that kind of head work into figuring out how you're going to make your application pipeline and make your dashboard saying or make your analytics gorgeous or whatever. Okay, so you'll have fun. I guarantee you'll have fun if you show up at this workshop. So, and please hurry up and sign up because it starts tomorrow. Alrighty, so now I'll get back to the regularly scheduled program. And I'm actually checking the chat I'm learning how to use zoom. So please feel free to ask questions that I might even notice. Okay, so today, what if you are new to SAS so I just griped about how freaked out I was learning SAS 20 years ago when I was like going oh my God this is current and Brahms and I just am using access over here. You know, like I was, I don't know, you know how like sometimes young people have like really high expectations of technology I've got to come down from it now. But I said, you know, I need to refresh myself I need to figure out what it would be like if I were that today. So I went on Reddit and I put the link in the slide but it's really not important who is actually saying these things so I just kind of cover their names. What's important is the points they're making. And this is actually a two year old post, but you know SAS kind of moves slow so I think the stuff still applies. And so the question around I went on Reddit, you know Reddit is kind of the social media where people talk anonymously and very frankly they tend to talk very openly. I'm not a big Reddit user but I found this and I thought this is great so just to step back. Reddit why would I be using evidence from Reddit well part of the problem was what I started looking for was just reviews of SAS, like people just writing up like Gartner writing up oh SAS as this new component or whatever. And I actually just cannot really find anything like I found a lot of marketing material from SAS and case studies from SAS but the problem is they were not you know if you're in the open source community you're really used to seeing very technical marketing stuff. Even if they're not really selling anything like if you're a consulting company that sells like our integration right like you'll do services. When you talk about like our packages you use you talk about how awesome they are and it's not like you're really selling our you're selling kind of yourself right like that you can apply these packages, you know, and so it's sort of weird like how does SAS make a marketing message right and I think that they don't maybe don't realize that they have to get technical with us they have to start talking about their components and making diagrams so that's what I'm looking for that's what I really want to see. But, you know their private company so then how much do they want to share, you know, so I don't know what they're going to do about that but the bottom line was, I couldn't find like really good evidence based official just opinions of what they were doing or case studies or use cases. So I went on Reddit. Okay. So this is the question I picked and I think it's really good. The questioner says I just moved from Python to SAS for four months due to new job requirements. So I'm imagining this person's like a data scientist and maybe works on contracts and so Python. I don't know Python I know our but Python is open source, and it's pretty complex just like our so. So this person adopted SAS for four months for a project at a job. And they asked you so they had four months of experience with it and they asked I wonder how you think SAS compared with other languages in a future. So I'm going to sort of interpret why I think this person was asking this is because they probably had this experience that the SAS was really challenging, where to buy that. And we're like wondering should I now that I know four months of SAS, and it's hard. Should I throw myself in the SAS, or should I just back away from it's to her. And the short answer I would say to that person is depends on how you feel like if you really like SAS. If you had a good experience and you want to throw yourself in the SAS, I would say do it. If you really didn't enjoy it if it wasn't a joy experience and don't do it you won't enjoy it. You know, I mean that's as simple as I would say why, because there's a place for you. It doesn't matter which person you are either person throws yourself in the SAS or not. There's a place for you data science. So that's the short answer. But now I'm going to, I'm sort of going to frame it. Should you throw yourself into mastering a challenging data science programming language that has been around since the 1970s. So that's maybe what this person is asking. And so I, in 2000 that's still a good question, right. So, 20 years ago it didn't matter whether I said yes or no to that question because SAS is still here. We're all still using it and at the time it was the only game in town now it's a question right like we've got Python and everything. But like I said, the advice I give this questioner is if they enjoyed SAS, you know, they liked it and they saw a future for themselves and then, yeah, throw yourself into it because there's going to be a role for you. But if you really don't like it, I don't know if you're going to like having a whole career in SAS. But that's, but what's different about SAS today. And that's sort of what I'm going to say is, is in the olden days, you kind of had to be in for a penny in for a pile of SAS like you had to set up a SAS service shop, and then just keep living in the SAS environment, and not really integrate, not use SQL, not use other things. Or you had to just figure something else out, which there wasn't anything else like if you really need to do analytics and you just think about the SAS, right. But nowadays, SAS is different because we have Viya, right. And so imagine you have any applications that are online applications, like I was somebody contacting me from like Oracle NetSuite, who's I think a sales person. I don't know much about Oracle NetSuite, but just the word NetSuite made me think, oh, maybe this is like online Oracle. If it's online Oracle, just think how easy that would be to integrate with online SAS, like SAS Viya. So those of you who've been around SAS a long time are starting to see what I'm saying, which is, today, if you're a new startup, you can adopt SAS without having to adopt the SAS server, you can just get the analytics, right. Which is what you wanted in the first place. So I'm kind of getting in my head of myself, but that's sort of what's beautiful and new. And then now I'm going to tell you what's ugly and old, okay. So one of the things that kind of disappointed me was I found sort of a theme that open source solutions work better and are cheaper right now. So, for example, let's say you have a problem, like you need to do principal component analysis. And you don't, you haven't done it. And these people suggest go look for an open source solution before you turn to SAS, even if you have a SAS job. So this one person said that their department got rid of SAS due to infrastructure issues and high costs, and that they converted all the SAS workflows into Python and saved a lot of money. The reason I put this up there is I wasn't sure you could do that. Like I didn't even, I wasn't sure you could do that. I wasn't sure that was even possible. So the fact that that person said they did it, and they said they did it two years ago, I'm like, okay, well, maybe not in every shop but it definitely was possible in that shop. And in the next shop where it says my company's doing the same thing. Now this is something really interesting. This person says they are porting everything over to Python and some stuff to nine. My understanding is nine is a platform you can do AI on and other automation, which is open source. Okay. So word porting. Okay, I have not heard anybody use that term, since I was like a teenager. So I'm like, okay, that we say migrating now, we don't really say porting but when I was little, you know, because we're stealing software we poured it over from one system to another. You know, I mean it's bad but you know but you couldn't even buy it that's the thing is you go to computer stores and they wouldn't even sell the software. So the problem is if today you've got a problem and you want to solve it with SAS, you're probably going to get it's easier for you to get a return on investment with an open source solution. Now just make, here's the fair comparison. Okay. SAS costs a lot of money but they give you a lot of support. They give you a product and support like they'll come and help you like SAS whenever I've worked somewhere where SAS we have SAS, they are like hands on helping me they're on the phone, you know. So I love that about them and when you get spoiled of that, and you're fighting with our Python, you miss it. Okay. So SAS is not just software it's support it's a whole bunch of other things. Okay, but still it is easier to get ROI with an open source solution, even if you dump a lot of effort into it, because of certain structural barriers basically because of that whole SAS environment thing. You know, if you're outside the SAS environment. It's really hard to get good IO in and out of the SAS environment so it's just. So this is, this is a big problem for SAS right now. You know what I mean, like that you can have open source solutions to some of the things that they sell to you, and the open source solutions are cheaper and easier to implement. It's not everything, certainly not everything, but that's some of the problem with some of what SAS is sold. Like, there are times in the past where those components were awesome like if you think about the output delivery system. Oh my God amazing. But now the output delivery system I'm sorry like it's old fashioned so if you're depending on it in your workflow, it's like what are you going to do now, you know. Even people in SAS shops are trying open source solutions first before looking at adding SAS components for all these reasons for the structural reasons for the fact that you actually have to pay SAS to the component. You know, then you have to change your workflow so I would keep recommending that but that's not going to work each time. Okay. Now this is kind of a problem. And I think SAS is undergoing an identity. I don't know if I want to call it crisis but an identity shift. I think they're really trying to change their image, because they don't have a very good image among like influencers. You know what I mean, because SAS is not very open, like it pretends to be open but then, like me like their content their legal wants to contact me because I'm telling her ready to adopt SAS. I'm like I have a book about SAS. I'm a LinkedIn learning author on SAS. Like, that doesn't pay me, they threaten me. So I'm like, you know, you can't go around doing stuff like that, and get people to love you and want to like your posts and stuff SAS, you know what I'm saying and so that's my rant about SAS. And this is that rant, if you look at it from a different point of view sounds like what these people are saying person one two and three. I put like you can download the slides and see their entire ran. But basically SAS is kind of spoiled because, like I said before, it was really cool to have that SAS environment if you had a SAS server you're very happy. And you could just add components you could fill it up with data that thing just ran like, beautiful, like it was beautiful. I love you. You know, I love going to that computer room using the SAS server just go so fast and it had everything you wanted. But the problem is, they didn't really innovate until it was too late like they kind of didn't get on and started realizing that are and Python were kind of competing with them. And so one of the people, these people are saying SAS is always trying to sell us some expensive add on that doesn't work as well as Python or our solutions. Well, 20 years ago, if you needed something and you were a SAS shop of course you would get a new component what else I mean there wasn't anything else. Now, if you need to make a dashboard or something why would you use SAS why not use Python there are and so. And if you come to SAS you're like we need to make a dashboard they'll try to sell you something that's not going to make you a dashboard as good as Python or art. So, then people get mad at SAS when they act like that, you know, then the other problem was it's hard to do remote work with SAS. But to be honest with you that only applies to legacy server setups. If you are in startup and you buy like SAS Viya and you are completely on that on the web. I'm sure remote work will be like no problem. The problem is, is if you had one of those beautiful SAS environments, those are physical environments right, and you can, you know, like tunnel into them, you know, like, tell that kind of stuff, but it's not really like, it's not like we're used to doing remote work like you're in the SAS environment like and you have to navigate it is really, I've done that and it's really like awkward and stuff is a little passion. It's like easier to just be on site at those SAS servers, and there's SAS consultants you can just hire them to come to your site and do stuff, you know, but then do you really want to like adopt server SAS today and the answer is no don't start a new SAS server. That's not a good idea today. The reason why anybody needed to put data on SAS servers was because you needed to serve it to the analytics component as efficiently as possible so you don't have IO problems. Now, you can serve it to SAS Viya from Oracle NetSuite or whatever. So you don't need to build a SAS server but if you have one. Like what are you going to do now, right? I guess hire consultants like need to come over and see if we can take care with your server. Person three, I'm sort of like paraphrasing what they said is now this is another problem. We can't find someone with the necessary SAS skills, but we already built everything in SAS so we were stuck. And so that's why if you happen to be the kind of person who's new to SAS and you're like, wow, this is a really interesting program. These products are really interesting. I'd love to automate stuff in macro language. I'd love to learn all of these components. I'd really like to make pipelines in SAS. If you're feeling that run after that feeling, throw yourself into SAS. And you probably, I mean, there's only so much you can do on your own outside of an organization, but if you can get yourself into a SAS using organization like most pharma organizations in that way, you can just have a blast. Like if that's where you're at, there are just so many solutions you can build in a SAS environment. And so, and you'll get paid a lot because like every time I look for the last 10 years, people are not becoming SAS users. They're not adopting SAS. New people are like, no thanks, I'll go to Python, I'll go to R. You can just download them. You can download R in Python and just use it. Of course you can use SAS ODA on the web, but it's not like there's a whole open source community behind SAS ODA telling you what to do. And although if you read my book, Mastering SAS Programming for Data Warehousing, I'll show you how to take my free online course. I'll show you how to do it. But yeah, so if you like SAS, the future is yours. You'll just keep getting more and more money in your salary. But another problem is that will only work. You'll keep getting more and more money in your salary if the SAS shop stays SAS shops, right? So one of the other problems that they were running into was that it actually from one of my earlier lectures, some SAS shops are just running out of money. They just can't pay. SAS keeps increasing their prices. Of course, it costs money to run a business. But people just can't pay, or I should say people, organizations can't pay. So what they start to do is sort of make, take some of their operations, like especially visualizations and stuff, and start like taking them out of SAS and moving them, migrating them to R, migrating them to Python, maybe migrating some data storage to SQL. They start just migrating some pieces out of the SAS environment, some functions too. But think about it. If that's going to happen, you still need to know how to use SAS really well. Because instead of operating in the environment, you're going to have to get stuff out of the environment, and you're going to have to know SAS and something else, because you're going to have to study the SAS version and then rebuild it in Python or R. So that's all going to be going on. So these are legacy SAS shops. That's activity that's going to be going on at all of them. So one person made this observation, or two people made this observation, that SAS, I quoted this, SAS being intertwined with education was a big deal. And I do want to emphasize that. That is why a lot of SAS users who learn SAS in college don't know much about applications in general, right? Because if, I mean, theoretically you could learn a lot of informatics in a few classes and a master's in public health or a master's in biostatistics. But the reality is SAS is so hard to use that you spend all your classes learning SAS, although I've been told that it's a little different now that they're trying to introduce R and Python into these course curricula. But how SAS got to be such a monopoly over education is to really start in the education space. And really, you need to have a university to run a SAS server. And, you know, if university are running a SAS server, they have all the components. And the idea was, you know, if I'm at the University of South Florida using their SAS server and I can use their geographic stuff, and I can use their sentiment analysis stuff, and I can play with SAS Enterprise Guide or whatever. When I leave and graduate, you know, they're going to cut me off. I can't use the USF server anymore. So I'm going to want to go to a workplace that has SAS, and I'm going to tell them I want this, you know, SAS Enterprise Guide and everything. And theoretically, that's kind of worked really well actually for SAS, because not only have they always been intertwined with education, they've also been intertwined with the government. Like, I was sort of curious about, like, I was reading some stuff. I don't know a lot about CDISC, you know, these different pharma standards, because they're sort of new for me, like I'm that old, right? And what I realized is that SAS had a lot, like SAS was part of writing those standards. Well, that actually answered a question in my mind. And that question was why about 10 years ago, there was kind of a kerfuffle in the R community that people were using R for clinical trials, and they were having trouble going through the FDA with their results. And I thought it was because the SAS engine is super amazingly awesome. Like that engine, when it calculates stuff, it calculates up to five zillion p-values, okay? And I don't think R does that. So that's what I thought they were talking about. But what really they were talking about is they just didn't trust people to rebuild the SAS macros and other programs and have them perform the same. Well, of course, even if you look at it as a black box, you can always demonstrate, you know, reliability and validity, even if you keep it a black box. But the blacker box would be SAS, right? And so now I better understand that. So I want to make sure you understand that SAS as a company is extremely powerful, and therefore they're involved in education, they're involved in the government, they're involved in these standards and stuff. So for better, for worse, they're not going away. And if you embrace them, you will get the benefits of knowing SAS because they're in everything, right? So that's the positive. The negative is you're going to have to know not just SAS, okay? Hopefully I've just convinced you of that. Even if you're a SAS enthusiast, you love SAS, and that's what you're going to do, big pharma, blah, blah, blah. The reality is encroaching upon SAS and you have to do integration. Like you're going to have to do some integration. Even if you're a SAS shop, even if you've got a server, you're going to have to do something like it's never, you can't just keep buying SAS components, it's not going to work. So I mean, even if you say, okay, well, I want to migrate to Viya, okay, that's a whole enterprise because what if you want to put your data from the SAS server into the cloud? Well, that's whole shebang right there. And so what I'm just saying is that even if you throw yourself and you're like, I'm going to be the SAS superstar that saves the day for all of these legacy server places that has the creative solutions that does all that. Well, the problem is you're going to have to get a whole lot of other training in informatics that you didn't get in your master's in public health, your master's in training. You know, that's not statistics. That's informatics. That's like business systems and stuff, which is why my free workshop. This is basically an advertisement for it. So another thing, like I'm in healthcare and epidemiology, I was actually surprised not surprised that people were saying that finance and banks are stuck in SAS. I know healthcare second SAS for these other structural reasons. But these people are saying that banks are second SAS. People in healthcare are not so negative about using SAS, like SAS is pretty cool. Like they don't go, oh, I hate it, you know, like, like, sometimes I'm in a bad mood about SAS, but mostly I like it when I'm using it, I like it. Well, they don't like it, these banking people. And the second person says banks got comfortable. And Nersha did the rest. Most banking models don't need more than proc reg anyway, due to regulations requiring interpretability. You know, and so it really looks to me like in the finance industry, SAS is really not the right solution. They don't even really need it. So I don't even know what's going to go on there. And it takes a while, like if even if you want to leave SAS and do something else, even just for a function, like, let's say that you have a, you have a set of reports, right. And what happens is the reports generated, and then you run a big proc tab to get the output. And let's say you decide, okay, well, let me just generate the report and instead hand it off to our Python to have them serve at the reports of the web. Oh my God, that is a huge, huge task because the first step is to break down whatever SAS was doing and super document it. And then the second thing is to go try to build it in our on the web or whatever. And then you got to do that validation thing is so much work. But if you save money and save time and your users are happy and you don't have to buy some component next year. Well done work because you can probably, as the web evolves, you can probably keep that report going. But I don't know if you can keep that report going in a SAS environment. So that's kind of the way I would see that. So finally, we have the SAS champion. So I've read to you a lot of negative stuff about SAS, but I don't see negative stuff as saying negative stuff. I see it as opportunity. There's a lot of opportunity in SAS right now, just for even what I just described, like breaking down macros and figuring out what they're doing. You know, and you can use open source software to help you with that. You can use SAS to help you with that too. And if you actually work at a place that has SAS, they have SAS licenses. You can call SAS. Like, I have to tell you that the people at SAS have been really helpful to me. So and you'll hear that they have a really good reputation for helping people. So I can't oversell that more. Like if you actually have SAS, you should really lean into it at a workplace because we don't have it at home, right? Okay, so this first person said, I just kind of summarized what they said, SAS works, and though our in Python are better at something, SAS will still be around for a while. And I would say where you really want to look at SAS is as a competitor to, and this is my opinion, and I don't know if I'm off on the deep end, but a competitor is something like AWS or like big analytics. So imagine you have something like a snowflake setup, like a cloud storage setup. And how does your data get in there? It gets in there from like a sales, a big sales, like database, okay? So now you've got, you've got everything sort of in the SQL format, relational database and snowflake, which is cloud storage. And then you've also got like your front end, you're basically your production database, which has all your sales transactions. Maybe you're selling like apps or something or selling, I don't know, something online. Okay. And then you want to ask business questions of your snowflake data, but your snowflake data maybe is relational. So maybe you want to change it shape into like a warehouse or something. And then now what? You can just lay that right on top. And now you're doing the big data thing, and you don't need a server. And so that's sort of like kind of what I'm, you know, thinking about is that SAS will be around for a while, but maybe we need to get creative with how to use it. So person two said SAS works, but only big companies can afford to keep paying for it and supporting it. And there's a doing the labor pool. So what this means is if you're a small business, so I just want to say I don't know how much it costs to adopt just SAS via your small business. I hope they have good pricing for small business because they did not have good pricing for small businesses for PC SAS 20 years ago. It was really accessible. I really hope that they have. But remember you have SAS ODA, which is free. So always play around with that. If you have public data and you want to try something, definitely use that. But then the other side of what person two says is that you're going to really need SAS users forever. And they're going to need to be really good at everything. They're going to need to be good at macros, reporting, enterprise guide, enterprise minor. You know, all of that data management data steps, you know, because look at all the problems we have, right? We have people migrating on the SAS. We have new people adopting SAS via like so there's all this activity in the in the future near future. It's just weird activity. It's like some like legacy people will be like exiting a lot of SAS and new people will be adopting SAS in a new way. So it's just going to be like really chaotic in my opinion. Oh, and Quentin's adds that he has a long back. So Quentin's in our local Boston SAS user group. And he says very true. SAS support is amazing. With SAS software you get validated software backed up by free technical support. And again, what's especially awesome about that is when you're trying to build pipeline solutions in your shop, you can literally call SAS and they'll be they'll collaborate with you like they'll help you because they're trying to sell you components and stuff, but they're not just trying to sell they're trying to sell you solutions. So one of the biggest problems I saw with SAS shops is that the SAS shop itself was lazy. Like they were lazy. And it was it was the shops fall because, you know, I'd go in and I'd be like, gosh, this is really slow or your servers like what's up with your I.O. And they were like, oh, I guess we have a whole bunch of data sets on and I'm like, well inventory them and take them off, you know, like do something, you know, or tune it or, you know, you can do thread threading. And so part of it is that like though that banking thing I'll bet those people in the ID department there just don't even care. And so, but on the other hand, if you that's what I say if you want to lean into SAS they were so happy to help you. I mean, if you are paying customer, they will help you because they know you're paying a lot and they do want to use their components. And it's not necessarily their components are not necessarily better than other things. But if you're in a SAS environment, usually their components are going to work better for you as a first pass. I don't know how SAS is about demoing things but I would say if you're in a server environment and they're trying to sell you a component demo it see if you can demo it because or if it's in Viya, you know, it might run better than an open source solution. Like you just don't know. So, so don't, don't make assumptions. That's one of the things I learned at SAS is, first of all, SAS has an awesome log file. Let me just shout out to the log file. It's very easy to collect data about how your SAS code runs. So do it, you know, try different environments actually do scientific studies because remember if you're running five zillion reports and you can shorten that each report by like one second just think about how much that matters right. So this inertia is the main reason why companies use SAS today, instead of migrating. And I would say that that is true about, maybe not all of SAS but like the components you probably shouldn't be having in your SAS environment anymore. This person over here on the right side of the slide says something really hilarious. When SAS has a problem they just pretend it is a new way of doing analysis. They used to be kind of guilty of that. But I think now they're getting, they're coming around to being more realistic. Like they understand, you know, SAS didn't have any competition for a very long time. Like I like to clarify. I learned SQL like in the early 2000s. And I already knew SAS. So I kind of knew access and I learned SAS and I was analyzing stuff and then I learned SQL. And so I was sitting down I was doing a SQL query. And if you know SQL, like you can do a count query where you take like a, like a categorical variable and you can count it, right. And I remember I was sitting in a SQL classroom and I was like, oh, look, I did a count query. And then I was like, then I said to the teacher, I was like, oh, how do you do crosstabs? It's like a crosstabs. I'm like, yeah, like this is like gender by like education group. And he's like, I don't know if you can do that in SQL. And that's when it started to hit me. Like, Brock Freak is amazing. Like you can't, I mean, you could do it in SQL. And I even once met this guy who was trying to impress me. He was at a statistics conference and he was saying, guess what, I programmed an ANOVA and a linear regression in SQL. And it didn't impress me. I was like, whoa, you did that? Like, and he's like, yeah, it took like all night to run. And then I was thinking, oh my gosh, like that's SAS, like that's SAS analytics is right there. That's the difference between SAS and SQL. But on the other hand, in SQL, you don't need a million commands to just do a query, right? Like it's not data stepping. So that's the kind of issue, right, is like data storage versus analytics. So you can still get the SAS tools to work for you today. But besides the real analytics, the hardware analytics, it's kind of hard and expensive. And they're generally better open source tools. So this is the challenge. If you've got a legacy shop, you're going to have to just relook at your shop and just see what to do right going forward. And also, businesses changed, right? Like print shops? Like what do print shops do now? So these are business changes that need to be reflected in your data system. And that's just the pain that we all have to go through, I think. But the take home message I would leave for everybody is that SAS is the cobalt of health data science, and that's not an insult. It's just a reality. So my mom was a cobalt programmer. And by the way, cobalt is a really elegant program, she tells me. Although keep a lot of documentation, which I would say about SAS too. And during COVID-19, you may have noticed a lot of unemployment systems around the US just broke. Like they couldn't print unemployment checks. They couldn't do stuff. Well, my mom said a lot of those systems are actually in cobalt. And there was just nobody to program that cobalt. And my mom didn't want to do it. She's done with that. So I mean, I told her you can make a lot of money, but she's not into that. But if you are into making a lot of money, you could do that. Like you could prepare for that inevitability. Like where you're the SAS expert and there aren't really any. Because it's kind of going fast. Like the real SAS experts, like when I was at the Army, I was there working there from 2008 to 2011. And we had the super, super duper SAS expert. And she actually was retiring. Like if I had kept working there, we would have had to replace her. And she's like an indicative of that generation of SAS programmers. Those people just retiring. And the new people are not like throwing themselves into SAS at high rates that same way. So part of what I'm trying to tell you is if you're new to SAS and you're liking it and you're seeing this vision that I'm describing and you want to be on that train, this is what you could be doing. Like in a lot of like future times, you might be saving people's like data. You might be saving their research projects the way cobalt programmers could have been saving people's unemployment. So to be clear, SAS alone is not the future. There's no future in setting up a SAS server today, a physical one. There's no future in setting up today. I mean, it's just we're on the internet at a different time. But the future is probably in SAS analytics because you can imagine probably pretty easily a situation where somebody goes online and they go to do sales transaction and they're overpricing it. You know, like their pricing, like for surge pricing, like tickets, right? And that request gets handed off to SAS analytics to run a little AI program and come back with a price. That's the highest price I'm going to pay for that ticket given the features of my variables. You can see that handoff going back and forth to buy us. So somebody's going to have to actually build that, right? Somebody's going to have to design that somebody's going to have to figure out the AI or even, you know, what I think is so funny about AI is we always had framing ham. I mean, that's just like risk scores and stuff we could be doing that and it's a lot simpler than AI. I guess that's explainable AI. But anyway, I mean, this is what SAS is for and we can do it now. We can really, we don't have to sit in a doctor's office filling out a little risk score thing. We can literally hand off real-time questions. Real-time questions to SAS Viya have it run the algorithm handed back to us like it's a dream, but it's not the dream we were originally having. So now the future looks more like, to me, is trying to get data out of SAS servers and into the cloud so we can do the thing I was just describing, right? And so there's a big opening in the future for SAS users. And so I would just say if you're SAS users, you should go and take advantage of it. All right. Well, thank you for putting up with me talking so long. I'm not talking this long. But anyway, I guess there's a lot to say about SAS. I just want to remind you I'm Monica Wahee and I'm a SAS user, but I use a lot of other things that you could tell. And I want to encourage you to sign up for my online workshop. It starts tomorrow, so hurry up and sign up if you want it. It's free. And especially if any of the topics I said today really stimulate your thinking, just sign up and come to it and we'll be discussing exactly those topics. Well, thank you so much for showing up today and I hope you have a good rest of your Sunday. I'll see you using the link in the description.