Cal Smith is a Senior Technical Architect and Associate Director for Cognizant. Cal has a long history with computers. He got his start in engineering school and has since done everything from animation to writing assistive software for the visually impaired. We discuss all of this (and more of Cal’s background) in our conversation today.

Throughout this episode, Cal also gives us a great overview of Health Cloud. We talk about its development and one of the unsung heroes of the platform: Omniscript. Tune in to hear all of this and more.

Show Highlights:

• How Cal got involved with Salesforce and creating communities.
• How he got started with Health Cloud.
• A typical use case for Health Cloud.
• What a validated machine means.
• How deletion of records is prevented within Health Cloud.
• What Omniscript is all about and what it does.
• Cal’s work and involvement with Dallas GiftForce.

Links:

• Cal on Twitter: http://twttiter.com/CloudMech
• Cal on LinkedIn: https://www.linkedin.com/in/Cloud-Mech/
• Cal on Github: https://github.com/crmprogdev

Episode Transcript

Cal Smith:
He knew how to do the declarative stuff, but he really didn’t know how to code or program. Well, I’ve written in so many different languages over the years.

Josh Birk:
That is Cal Smith, a senior technical architect and associate director over at Cognizant. I’m Josh Birk, your host of the Salesforce Developer Podcast. And here on the podcast, you’ll hear stories and insights from developers for developers. Today, we’re going to sit down and talk with Cal about his long history and computing. We’re going to go over an overview of Health Cloud and we’re going to talk about one of the unsung heroes of the platform, OmniScript. But we’ll start as we frequently do with Cal’s early years.

Cal Smith:
Well, I was always kind of mechanically inclined and was curious about the way things worked and a mechanical engineering degree is sort of like a liberal arts degree in anything mechanical, whether it involves gears or thermodynamics and those types of things.

Josh Birk:
Got you. I mean, you spent a long time in engineering, including you have a couple of patents under your belt, right?

Cal Smith:
Yeah, I have four US patents from my days at Ford Motor Company related to climate control, which is air conditioning systems and so forth on automobiles.

Josh Birk:
Interesting. So you went into engineering, which is kind of adjacent to computers and software engineering, what started to lead you into that territory?

Cal Smith:
Well, I took my first computer engineering class, I think it was in 1976. It was in FORTRAN using punched cards. This was before floppy discs had come around yet. And you’d put your job in to submit it and if you were lucky, it would come back to you within a day or two, because things ran on priority. And as a student, my job’s got the lowest priority so you couldn’t wait until the last minute to run your homework needless to say, especially if you had a compile error or something like that.

Cal Smith:
So we used computers all the time throughout my engineering program as I was a student, including doing 3D modeling and that type of thing, even in its crudest rudimentary forms, gosh, that’s hard to say rudimentary. But yeah, we used it for all kinds of simulations that we would do. I can remember in my early days at Ford’s creating simulations of automotive air conditioning compressors.

Josh Birk:
Interesting, interesting. Well, and kind of a side question to that, tell me what it means to be a member of A Bunch of Short Guys.

Cal Smith:
Oh, okay. Wow, you really did do your research.

Josh Birk:
I did a little bit of homework, it’s true.

Cal Smith:
A Bunch of Short Guys are guys that make short film animations. And I used to work with Autodesk Maya doing 3D modeling and animation. And these guys were some of the best of the local guys who did that genre, at one point Dallas here was a hub of animation and now all of that work has pretty much moved to Austin and parts far and beyond and elsewhere.

Josh Birk:
Got you, got you.

Cal Smith:
Yes.

Josh Birk:
Nice. Well, and I want to give another shout out to something that I found when I was doing the research. Tell me briefly, at least a little bit about because you were working for a project that was giving computers to the blind?

Cal Smith:
Yes, yes. Computers for the Blind. I was part of the team or actually led the team, I should say, from Dallas Developers who set that charity up with the nonprofit starter pack. And what they do is they take donations of old computers and they like when possible to get them from corporate corporations and they rebuild these and set them up with a new license from Microsoft and they add all kinds of visual assistive software to them and for a nominal fee, then they provide them to someone who’s visually impaired.

Josh Birk:
That’s awesome. That is awesome. So, well you mentioned NPSP, so then that must be predating. What was your first introduction into Salesforce itself?

Cal Smith:
Well, actually that’s pretty straightforward. A good friend of mine asked me if I could write HTML for him to create some PDF templates that he needed and he needed them pixel perfect. And that’s not the movie we’re talking about, but the templates had to be spot on and he didn’t have the patience or the knowledge to do it but he knew that I used to write HTML beginning with HTML three, and that was my introduction to Salesforce. Before too long, I was writing controllers and then triggers. It just kind of grew from there and best thing that ever happened to me.

Josh Birk:
Nice. I love how most people either get in, it’s either the controller or the trigger. It’s either you have to render a page or you have to respect some kind of business logic and that’s what gets you sunk into Salesforce development.

Cal Smith:
Yeah. Yeah, that’s essentially what happened with me. Next thing I knew I was working with him to create one of the first communities after communities were released.

Josh Birk:
Really?

Cal Smith:
Yeah.

Josh Birk:
Wow. Okay. How did you get involved into that?

Cal Smith:
Well like I said, maybe I didn’t say yet, he was a partner. And he essentially he knew how to do the declarative stuff, but he really didn’t know how to code or program. Well, I’ve written in so many different languages over the years C, C++, and various things, Assembler, you name it, that I didn’t have any trouble picking up Apex.

Josh Birk:
I was going to say Apex has to have so many similar design patterns based on what you’re already used to.

Cal Smith:
Yeah, exactly, so it came to me pretty easily. And he needed somebody that could focus full time. And we talked about the animation business having essentially gone south here in Dallas so I was eager for the work. And it was a great learning experience for me and was just ideal for me and really changed my life. And it was something I wound up really enjoying doing

Josh Birk:
Nice. Well, now these days you have your hands into Health Cloud, which is something a little adjacent to us but a lot of developers get used to, how did you get involved in Health Cloud itself?

Cal Smith:
I was working at Emphasis and it was something that we were trying to break into. And we had apparently lined up our first Health Cloud customer and so that year at Dreamforce, I got some Health Cloud training from Josh Hildebrandt and a few others while I was at Dreamforce. And then in January right afterward, because I think Dreamforce that year was right at the end of November, we started into a new project. And that was what kicked it off for me. And it seems like I’ve hardly worked on anything except Health Cloud since then. And I think that was 2019, something like that.

Josh Birk:
Got it, in the old days before. Walk me through a couple of like, what’s a traditional use case for a healthcare provider using Health Cloud?

Cal Smith:
Well, really in traditional Health Cloud, it isn’t so much health providers. You really have what are called your payers and your providers, so to speak. But providers are more like institutions in the sense of hospitals and medical clinics, providers are healthcare providers, which think of in terms of they can be doctors, any kind of physician, they can be nurses, people who do physical therapy. In other words, it’s a generic term, depending on what type of license you have to do some sort of medical work, a healthcare provider can be many different things, at least that’s the way the model works. So in essence, each provider has a specialty or area of expertise.

Josh Birk:
Got it. And Health Cloud is adding on the data layer that understands those variations?

Cal Smith:
Yes. Health Cloud sits on top of Service Cloud, and that’s really what enables things. And it extends a lot of the way Service Cloud works in a way that’s meaningful to people that are in the health industry. I have to tell you that you have to learn an entirely new vocabulary in order to work in Health Cloud because there are things called accumulators and all sorts of other things that you have to have knowledge of that’s very industry specific. You can’t just go implement Health Cloud without knowing how that industry works, which you can say the same about most industries, but there’s a lot of very specific knowledge when it comes to that industry, along with regulations that apply to it.

Josh Birk:
Right, regulations and I’m assuming a lot of just things that have to be accredited and validated. And speaking of, you’ve written about a Salesforce work functioning as a validated machine, what does a validated machine mean?

Cal Smith:
Okay, in life sciences, to introduce a new term here, life sciences really applies to pharmaceutical companies, companies that make artificial joints, implants, heart appliances, those kinds of things, biomechanics. Most everything they do is regulated by the FDA and there’s regulation, that’s CFR 2011.11, if I’m not mistaken. It’s very well known and understood. In essence, most of what they do when it comes to things that are put in digital form that apply to providing reports to the FDA, to running different types of programs have to be done on what’s referred to as a validated machine, meaning that if they have a server on premise, because at the time these regulations were written the cloud wasn’t a big thing, and most things were essentially servers of some sort that ran some sort of the operating system, whether it be Linux or Windows or whatever, you had to be able to show that you could put it into a validated state, that it was stable and everything was reproducible.

Cal Smith:
Essentially what we’re doing now that Salesforce is in the cloud, we’re using that same analogy and Salesforce is being treated as though it is a validated machine. And we all know that Salesforce sits in a pod located someplace around the world and we share resources with a whole bunch of other instances. It’s virtual and one moment it might be in Chicago and the next moment it might be-

Josh Birk:
It might not.

Cal Smith:
Yeah, it could be in New York, it could be in San Francisco. It could be wherever because it’s all virtual, as necessary because of fallback and other things. But the idea is that because of configuration and other things that we do, and it doesn’t matter what languages we’re using to run it on, we can treat your Salesforce org as a validated machine, because we can make certain that the configuration that it has is set up and is put into that configured state that we had it at when we started everything and that we can run a checks and balances to make certain that all the data that’s gone in, that’s gone out and we can check our logs and everything, just like a balance sheet so that we know that it is in a validated state because everything is accounted for, no one has deleted any records or everything can be accounted for. And that’s what you do when you put a machine into a validated state, there are no records that are unaccounted for. All your logs tell you where everything is at.

Josh Birk:
So it sounds like the goal is like a 100% level of predictability.

Cal Smith:
Yes, absolutely. That’s what you’re looking for. You want to know that when your machine works or an operation executes, that it is predictable, just like when you run your unit test, you’re predicting through your unit test with an assertion what will happen.

Josh Birk:
Because we all know that unit tests vary in their, shall we say, state of quality? There’s the ones-

Cal Smith:
Well, properly written unit test, let’s put it that away. One that has the appropriate amount of assertions in it will take you through all of the steps and will assert that the outcome is the correct outcome based on the inputs that it asserts and can assert those all the way along through the process.

Josh Birk:
So which part, where does the audit come in? Does the audit include like a code review to make sure that people aren’t cheating with a bunch of i++ iterators or anything like that?

Cal Smith:
Well, that’s part of the process of building a new org in that sense. But yes, you would have audits along the way. These orgs are highly secure, as if I really need to say that. So they are regularly audited for a variety of reasons. But one of the main reasons that they’re audited is to make certain that all the records that you’re expecting to see are there and that there aren’t records that are missing. It’s almost as if these orgs are an ideal use case for blockchain.

Josh Birk:
Interesting. Has anybody started moving in that direction?

Cal Smith:
Not that I’ve seen in Salesforce but I have in other use cases. I’ve seen that for signatures on, I think it’s Pegasus.

Josh Birk:
What is Pegasus?

Cal Smith:
Pegasus is part of what Salesforce is replacing on premise.

Josh Birk:
Sorry Pegasus. Got you. But to walk through that a little bit, the rationale behind that would be because one of the core concepts of blockchain is that the data is immutable, you can add an append, but you cannot remove and delete?

Cal Smith:
That’s correct. In other words, the transaction that happened before and the transaction that happened after cannot change the transaction that happened in between and you can give that piece of digital information to someone and they can see that it is what it is, and it can’t be changed. It’s like cryptocurrency in that sense.

Josh Birk:
Mm-hmm, mm-hmm. And so in normal Health Cloud, is there already, I want to say workflows, but I guess these days just flows, is there already a structure in place to try to enforce that kind of thing?

Cal Smith:
A structure in place to what? I’m sorry.

Josh Birk:
To enforce record transactions, to make sure that something can’t just be deleted and forgotten.

Cal Smith:
Not out of the box so to speak, that’s really a matter of the SI, systems integrator for those who aren’t familiar with the term, would configure or use set up some sort of validation rules and that type of thing or through permission would prevent deletion of records.

Josh Birk:
That was going to be my follow-up question. It sounds like this is a very specific use case where I would assume the ability to delete certain records would be held only by a very small group of people?

Cal Smith:
That’s correct, yes. What you’ll find in most organizations who have data quality problems, there are too many people that have permissions to do things that they shouldn’t be able to do.

Josh Birk:
Yes. I will quote the will have been out, but is upcoming interview I just did with Megan Broki, where she got a new job and she walked in and she was like, “There were 204 users and there were 204 system administrators. And then the next day there were 204 users and there were two system administrators and I was the most hated person in the company.”

Cal Smith:
Yeah, I can believe it. But that’s exactly the problem is that you have people who should be standard users that are doing their work with elevated permissions where they have the ability to delete records or modify records that they shouldn’t be working with those kind of permissions enabled.

Josh Birk:
Right, and when we’re talking about things like the life sciences, those kind of permissions, am I being overly dramatic by saying potentially has life and death consequences?

Cal Smith:
That’s correct. You’re not being overly dramatic when you say that, I don’t think, at all.

Josh Birk:
Got you. But this level of scrutiny and this level of predictability, do you think that there are lessons that other industries could learn from that? What’s gospel across the board?

Cal Smith:
Oh, absolutely I think there are lessons to be learned. Do I think every organization needs to go to that level? No, I won’t say that. But what I will say is that, imagine that you’re a billion dollar organization and you are making decisions based on your data that is being used to generate reports. Now imagine that 20% of your data is simply garbage.

Josh Birk:
Right, which is a very honest ballpark. I feel like.

Cal Smith:
Yeah, it is. It’s reasonable based on a lot of orgs that I’ve been in. Imagine I’ve seen orgs that had more than 20% duplicate accounts.

Josh Birk:
Not great, not great. Now, another thing that you have gotten your hands on, I believe that a lot of developers have not, is OmniScript, which officially became part of the platform when we did the acquisition of Velocity. Can you give me the elevator for OmniScript and what it kind of does?

Cal Smith:
Yeah, sure, I’m happy to, I love that tool, absolutely. I learned it when it was still Velocity. The best way that I can describe OmniScripts is that they act as a layer that exists between your integration, your Apex, and your database and your UI. So what they allow you to do is manipulate and transform that data, whether it comes in via an integration or you’ve queried it from the database, or somebody has input it via the UI, they can manipulate it in any way that they need to and display it and work with it using transforms and various other methods before it’s ever written to the database, or simply goes back out via integration.

Cal Smith:
So imagine that you’re in a huge enterprise organization and you have a lot of the stuff siloed, say you’re a big health insurance company, for example, and you have an application where you create all your policies that’s in another org, it’s not in your Salesforce org. It may be an internal SAP type of org. All right? So then you have this Salesforce org that’s dedicated to doing nothing but sales and selling policies. So when people are calling in or you’re calling out to try to sell policies, you’re collecting their information and you’re inputting it, and your integration is bringing in all the information that’s needed to collect it, to price these customers with policies. All right?

Cal Smith:
And all of this is on the fly. You can generate all of this without having to write anything to the Salesforce database. And you can give them a price and a quote, you can save it locally to the database if you need to. But more importantly, you can write it back to your external database without ever having to touch your local database as a quote that you have created for this customer, that goes to some other centralized database that you have on your website where they can pick up their quote from.

Josh Birk:
Got it. So it gives you kind of like a transformative layer that you can play with the data, whether that data’s third party or first party, and you don’t have to bother the first party or third party, if you don’t need to?

Cal Smith:
Correct, and you don’t have to imagine trying to do that with visual force or with lightning components.

Josh Birk:
Right, because those are such first party citizens on the platform that statement don’t talk to the first party databases, pretty difficult to do sometimes.

Cal Smith:
Yeah, trying to bring that in and hold it in a case or anything else to then be able to manipulate it with all these variables, your heap gets really big.

Josh Birk:
Mm-hmm. And so it’s a whole custom object data layer that you don’t actually have to create. So is this why it seems like it’s very industry focused because it’s like, it’s these very specific integration layers, like the one you’re describing there where it’s a tentative price quote that we don’t need to care about until somebody’s actually going to do a sale.

Cal Smith:
No, the reason it’s industry focused is that there are pre-built scripts that are conducive or built for certain industries that we know certain industries are going to need. For example, in the health industry, you have to validate someone’s identity when they call in to talk about their insurance policy and their health information. So that type of basic script is already there and exists out of the box.

Josh Birk:
Got it. Very nice, very nice.

Cal Smith:
All right, and there are similar ones that exist for other industries that I couldn’t tell you about because I haven’t worked with them, but I know they exist.

Josh Birk:
Yeah, no, I know people have had difficulties talking about them, because it’s like, it’s not a vertical, it’s many verticals, with a lot of very different tailored solutions to it.

Cal Smith:
Yeah, with these transforms, that’s how you do CPQ because you can do matrices on this data while it’s in there without having to save it to a database or anything like that.

Josh Birk:
Got it.

Cal Smith:
I mean, it’s incredibly powerful. As I looked at it, as I was going through the learning, when I was on my learning journey for certification, it was boggling my mind all the power that the stuff had. And it made sense to me why people were going to Velocity or Salesforce Industries for CPQ instead of going to the Salesforce CPQ version. Which was made by, I can’t remember who now, but-

Josh Birk:
Several acquisitions ago, I think.

Cal Smith:
Yeah, we both know what I’m talking about.

Josh Birk:
Yes, yeah, yeah. Nice, nice. Well, to change course a little bit, I want to talk a little bit more about the stuff that you do in the community. Tell me a little bit more about, or tell me about, is it Dallas Gifts Force?

Cal Smith:
Yeah, it was something that a number of us got together and created, gosh, that’s been at least five years ago with the idea that it would be self-sustaining. And also, I guess it’s a little bit ironic in that we now have the new BA certification. A lot of the developers that were members didn’t really know how to do discovery or be a business administrator. So the commitment I made to them was that, all right, we’ll set up Dallas Gift Force and I’ll help you do it because I believe in it and I want to see this thing live beyond me and not need me per se. And in the process, I will help train the first five BA so that you’ll be able to do your own discovery. And then on an ongoing basis, you guys can choose worthy nonprofits that otherwise could not afford to implement Salesforce nonprofit.

Cal Smith:
And that’s how it came about. We had a core group that went through setting this up and how we were going to organize it and so forth, and that was the essence of why we formed Dallas Gift Force was one as something that was independent of Dallas Developer Group that could exist on its own and that everybody in the community could use to help the nonprofits. Because there are events that happen here, one of them that’s sponsored by Slalom each year in particular, where we would show up and they would say, “You guys brought too many people.”

Josh Birk:
It’s a good problem to have, but it’s still a problem.

Cal Smith:
Right, yeah, they couldn’t accommodate us. And we said, “Okay, we need to set ourselves up for success and pick what groups that we want to do,” and that was part of the motivation. And some good things came out of that, and I like doing good things

Josh Birk:
And that’s our show, now before we go, I did ask after Cal’s favorite non-technical hobby and, well, it’s actually one I used to do a lot more before I got a rowing machine.

Cal Smith:
Would riding bicycles count or are they too technical?

Josh Birk:
Nope. Nope. Riding bicycles is great. I love it. Is that your pandemic get out and exercise thing?

Cal Smith:
Well, it’s what I used to do as a hobby until I tore myself up a little bit from racing and crashing a few times too many, but I still love the sport and I get out when I can.

Josh Birk:
I want to thank Cal for the great conversation and information. As always, I want to thank you for listening. Now, if you want to learn more about this show, head on over to developersalesforce.com/podcast, where you can hear old episodes, see the show notes at links to your favorite podcast service. Thanks again, everybody, and I’ll talk to you next week.