Raw Data By P3 Adaptive
Cloud Services Used to Be Really Dumb w/ Eric VigesaaListen Now:
In today’s episode, we welcome Eric Vigesaa of Microsoft to the show. This special guest has a decades-long relationship with our host. According to Rob, Eric can always be counted on as a partner in cynical observations. Rob describes Eric as one of the rare breeds that enjoy the excitement of building an emergency solution. Someone who prefers a shorter distance between the software and the problem. The closer he is to impact, the more satisfying the solution becomes to him. Sardonically, Eric is also like the groundhog of Microsoft, emerging from IT, looking around, seeing the shadow, and returning back to IT: empowered, passionate, and skilled. Eric describes his non-IT time as an evolutionary course that allowed him to explore and grow his aptitude.
Also in today’s episode, Rob and Eric describe the beginning phase of Microsoft’s shift to computing in the cloud and the project carnage in its wake. Promising projects suddenly found themselves on the road to nowhere and are now, nowhere to be found . . . with one popular exception: streamlined conditional formatting. That simplistic innovation not only made it out of the darkness to see the light of day but also became Rob’s gateway drug to audience participation. Other innovations weren’t so lucky. Eric and Rob describe what happened to a multitude of promising projects that were sliced, diced, julienned, and unable to survive cloud modernization at Microsoft. The history might pull you in, but it’s the dry wit that will keep you through the end.
As a bonus, today you also get at least one bad Star Wars joke . . . and don’t forget, if you enjoyed this episode or the joke, please leave us a review on your favorite podcast platform and be sure to subscribe to get new episodes delivered to your inbox.
Also in this Episode:
Rob Collie (00:00:00): Hello, friends. Today's guest is a very dear friend of mine named Eric Vigesaa. I've known Eric for about 20 years now. I promise you that every time I think about Eric, I smile. I moved away from Seattle 13 years ago, and if I moved back tomorrow, one of the first orders of business would be to get into a routine of seeing Eric again, so transparent, so warm, clever, and he knows the value of fun. He and I will both happily and actually even enthusiastically make observations about the world or confessions about ourselves that on the surface might come across as cynical or pessimistic, except while we're doing it, we're smiling the whole time because it's actually optimistic.
Rob Collie (00:00:49): We're acknowledging things about the world or ourselves in a very raw, unfiltered, non-sugar coated sense. But we do it in part as a means of improvement, identifying those rough edges that we'd like to sand down, but also as a way of saying, even though this is true, things are still going to be completely okay. For us, I think it's a way of defanging the negative power that these things can have in our lives. It's like shining sunlight on the vampire, and I'm pretty sure you'll get a sense of that vibe from our conversation. He and I worked together at a very, very interesting point in the history of the world of software, when the software industry had gotten the message that it needed to shift to cloud but didn't yet know at all what that meant.
Rob Collie (00:01:36): So, we lived through some very, very interesting false starts down that road at Microsoft. So, a lot of the things that Eric and I worked on together never actually really made it anywhere in terms of final product. But if you've ever used the conditional formatting feature in Excel, the more modern version starting in 2007, Eric had a lot to do with that and I had a little bit to do with it as well. So, not all is lost. He's still at Microsoft. He's in a very different role, and I dare say that he's had quite a bit to do with the actual transition to the cloud. So, come for the morph, stay for the fascinating history when we get into it.
Announcer (00:02:12): Ladies and gentlemen, may I have your attention, please?
Announcer (00:02:16): This is the Raw Data by P3 Adaptive Podcast with your host, Rob Collie, and your co-host, Thomas LaRock. Find out what the experts at P3 Adaptive can do for your business. Just go to p3adaptive.com. Raw Data by P3 Adaptive is data with the human element. Welcome to the show, Eric Vigesaa. How are you today, sir?
Eric Vigesaa (00:02:46): I am doing well. It's a privilege to be here. Ever since you looked me up, a flood of memories have come back, so I look forward to sharing some of those today.
Rob Collie (00:02:55): So I would say that memories that you and I would both record will skew so funny, at least to you and I, because who knows about everybody else's sense of humor.
Eric Vigesaa (00:03:06): Beyond that, who knows.
Rob Collie (00:03:08): I was thinking about this ahead of time, just like you. If you wanted to predict whether or not someone would get along with me, one of the tests you could run is to have them hang out with Eric for a little bit and maybe vice versa if I might be so bold. I think that the similarities between the way you and I view the world, I can't think of someone more on the same page as me as you. I didn't realize that until running the regression, going back over the number line, going, "Oh, holy hell yeah, this guy's as weird as I am."
Eric Vigesaa (00:03:47): I take that as a fine compliment. I feel like there are some differences and maybe in how we project. I feel like you have maybe tapped into your skills to be a little bit more externally dynamic than I, but as far as what we dig and how we like to communicate, I totally vibe with that. Yeah.
Rob Collie (00:04:08): Yeah. Another way to say is if I have some just sort crazy new theory on the world, I'm pretty sure sharing it with you. You might agree with it, or if you don't agree with, it's because you want to improve it. You are not going to dismiss it. We just did the Software Hall of Fame Podcast. At the end of it, because we had Connor, my friend Connor on, Connor and I, you probably got exposed to this. If you missed this completely, then I apologize. But Connor and I had refined the rules of what constitute a sport.
Eric Vigesaa (00:04:38): Oh, yes. Very, very familiar.
Rob Collie (00:04:40): Okay. All right. Okay. So, that's the thing that I know you'd be down for.
Eric Vigesaa (00:04:45): I want to say I came into it at a point in time where pretty much any question, objection, hypothetical, you had already thought deeply about that and had a well-reasoned answer. It was almost like, "Hey, what do you got for me? I've heard that one 18 times, and I have my answer down pat for that." I couldn't really find any chinks in your armor, find any points of doubt. I remember that very well.
Rob Collie (00:05:14): Yeah. You raised an objection and my first two words were always, "Ah, yes."
Eric Vigesaa (00:05:21): Oh, yes, the so and so objection. Yes, I'm very familiar with that.
Rob Collie (00:05:26): It's like in chess, right? Oh, you're using the Queen's gambit.
Eric Vigesaa (00:05:31): Exactly right. Yes. That's been conclusively debunked six years ago, but let me educate you.
Rob Collie (00:05:37): And while I'm speaking, you see all of the citation references from an academic paper appearing on screen and things like the words at all are being... All right. So, you and I, we crossed paths working on Excel in a very, very, very, very strange era. Microsoft would probably say, "No, that never actually happened. We never even had that release." It's not necessarily a shame, the release that we worked on. It's more like it just didn't do anything. There was this time where Microsoft had woken up to the idea that online services were the future. We were handed this mission from on high, and it was not to be questioned. Now Office, you are going to go build a release focused on online services. The classic example of good idea, tremendous foresight, terrible depth perception.
Eric Vigesaa (00:06:32): Is the world hailstorm coming to your mind?
Rob Collie (00:06:34): Yes. What were the different nodes, red and blue, or were they right red and green? I was reading a book about the making of the atomic bomb. It's really just a history of nuclear physics in a way and how they were talking about the different colors, of course. They're like, "Look, they're not actually colors. It's just the naming scheme, right?" It's the same thing with the hailstorm, the different properties that were either red or blue or blue or green or whatever.
Rob Collie (00:06:58): So, yeah, I'm reading a book about early 1900s physics and hailstorm pops into my head. It's like that's how seared. That was the release where Microsoft practically gutted the teams that were working on almost every major application in the Office suite. There wasn't going to be an Excel team. Don't you understand? We're cloud services now.
Eric Vigesaa (00:07:25): No work in the core apps was the mantra that I remember.
Rob Collie (00:07:29): Right, core apps. It was so derogatory.
Eric Vigesaa (00:07:34): Oh, yeah.
Rob Collie (00:07:34): Core apps.
Eric Vigesaa (00:07:35): Just dinosaur.
Rob Collie (00:07:36): Yeah. You don't want to work on the core apps. Into that void stepped a few youngsters who weren't thinking necessarily about their careers quite so aggressively. All the cool kids went off to do something else. We're not the cool kids, but I guess it's a reasonable consolation prize that, okay, I guess we'll just get to work on Excel.
Eric Vigesaa (00:07:59): Is that why we got a shot? I've always had these thoughts of, "How the heck did I get to work on this stuff? What qualifies me to influence a product at which is so broadly used?" Maybe you just hit on one of the factors is all the hot shots were looking to get onto this new wave.
Rob Collie (00:08:18): Yeah, I think that has a lot to do with it. You still had to have a pulse to be considered for such a thing, but that whole vacuum that was created, there were many, many spots ahead of us that would've been filled by people with longer running, stronger reputations than by us. I had done nothing in Richard McInnes org up until that point in time to truly distinguish me. I remember taking stock, I think there were seven program managers who had straight up left that org after the XP release.
Eric Vigesaa (00:08:51): Wow.
Rob Collie (00:08:51): I got seven battlefield promotions by default-
Eric Vigesaa (00:08:54): That's insane.
Rob Collie (00:08:54): ... in the order. So, again, you don't want to be overly self-deprecating and say, "Yeah, I mean they just scrounged around, scraped the bottom of the barrel, and found us." But we have to acknowledge the fact that that seismic shift created a lot of opportunity that wouldn't have otherwise been there.
Eric Vigesaa (00:09:11): Yeah. I think that's totally fair to say. Now what you did or what I did with that opportunity is more of our own making, right? Hopefully.
Rob Collie (00:09:17): Yeah, that's true. That's true. It was a very strange environment, right? Okay. You're going to have to work on the core apps, but you and I were not assigned to work on the core apps in the maintenance sense. Do you remember there was another team? Eric Patterson was on a different team shepherding the core applications, but me, you, Chad, we were working on integrating the idea of cloud services and getting Excel ready for this brave new world in a way that was our job because no one was paying attention to us. Somehow we got to define what that looked like. Are you crazy?
Eric Vigesaa (00:09:55): Seriously. Chad played a big part in that. There was a big piece that I ended up doing a lot of spec work on that ultimately got cut, didn't ship, but it was pretty interesting as a thought exercise to imagine that. The iconic example was bringing structured data from some internet based data source into Excel and mapping it in to Excel in some way. We imagined data repositories out there that could be subscription based or could be free. Somehow you've got this live connection into Excel. That was a big scenario, which we were trying to get our head around and make real.
Rob Collie (00:10:31): Yeah, I think this was the first time in my career where I started the distinction between good idea, good strategic thinking, and still building something that wasn't worth anything. A lot of these good ideas just keep happening. So, the thing you were talking about, we're thinking about this idea of a marketplace where people could even sell premium data feeds. We were talking about this in the early 2000s, right? We're like in 2001 and we're talking about this. Okay. So, then years later, around 2010 or maybe 2011, suddenly, there's this thing called Azure Data Marketplace and it's exactly what you and I were working on, but it's better funded, it's better staffed.
Rob Collie (00:11:18): The world has gone on another decade farther. So, there's the potential for demand for it. They've got a whole team working on it. Guess what? Few years later, even that gets the plug pulled, down it goes. So, we weren't the first failed effort in Microsoft and this space and we were also not the last. You know what? Maybe we were the first.
Eric Vigesaa (00:11:39): We could be the first. That's true. That would be an honor. Yeah. Yeah.
Rob Collie (00:11:43): We failed before Azure Data Marketplace failed and we failed with fewer resources I'd like to add, right?
Eric Vigesaa (00:11:48): Yes. It was a very efficient fail.
Rob Collie (00:11:50): Very efficient. Look at almost everything we did work on there. I was obsessed with this thing called data merge. This idea of data transformation as data flowed from point A to point B, just like the thing you were working on, the marketplace thing. I could never get our VP to believe in it, so that died. Years later, we get to see this thing called Power Query come out, and it doesn't have exactly the same vision. They've turned it into data flows and things like that, but this idea of end user ETL was at the core of my "data merge idea" that I was so obsessed with. Now that I see what the language looks like, what Power Query looks like, and what the M language looks like, I'm like, "Oh, my God. I am so glad no one let me pursue that."
Eric Vigesaa (00:12:44): Oh, interesting. Because if you'd had, you would've made so many mistakes.
Rob Collie (00:12:49): I would've underestimated the problem and come up with something that was just useful enough that customers would adopt it and then complain about it. And then we'd be chasing that inelegant original solution forever. I wiped my brow and went, "Phew, what do you know? Richard was right."
Eric Vigesaa (00:13:08): It's interesting though, how in the beginning of these planning periods for a new release, it's all pie in the sky. People who get too practical and realistic too soon are poo-pooed. You're supposed to think big. Blue skies, we used to say. Yet that process from getting there to something that we actually think is viable and we actually are ready to invest in, it feels very non-scientific to me. At least it did back then. It really still does today.
Rob Collie (00:13:37): Oh, yeah. Looking back, that whole era is just one of vain hubris. Do you remember the joke, which was many, many, many ideas for functionality and features of Office would start out with someone saying, "Wouldn't it be cool if..."
Eric Vigesaa (00:13:56): Totally. It was literally a personal "Wouldn't it be cool?", rather than some data driven, market research driven, customer feedback driven initial thought.
Rob Collie (00:14:08): Whether or not you got traction on this idea, it had everything to do with whether or not the other people in the room respected the nerd chops of your idea, the coolness of it. It had to be cool. Practical? Oh, no, no, no. It had to be cool. So, yeah, the whole thing was just dumb. I mean, everything about that era shows what it's like when an organization comprised of computer scientist types finally starts to crest and crash on reality. Now, what? As the capacity and capability of computing has increased, what it actually does in the practical real world has become almost anything is possible, right?
Eric Vigesaa (00:14:46): Yeah, totally agree. I might get the timeline a bit wrong here, but you had come off an experience where you had something cut at the last minute.
Rob Collie (00:14:57): That's right.
Eric Vigesaa (00:14:57): Now this was very different. This was cut and a lot of these things were cut in an earlier phase. So, maybe that didn't feel so traumatic to you as it did to me. I think anytime you work on something, the first time you get cut, it hurts in a way that probably nothing else ever hurts in that same way.
Rob Collie (00:15:14): All the different skits that were pitched on Saturday Night Live ultimately didn't make the final broadcast. Those people have got to develop some thick skins over time.
Eric Vigesaa (00:15:23): That's right.
Rob Collie (00:15:24): Yeah. So, there was another difference though in the fact that the product of mine that had been cut, completely canceled, the whole product. Not a feature, entire product.
Eric Vigesaa (00:15:33): That's right.
Rob Collie (00:15:33): A new start menu shortcut in the Office suite that never happened.
Eric Vigesaa (00:15:40): Wow.
Rob Collie (00:15:40): We were two years into it, but the huge distinction here between your feelings and mine was that for me, it was a tremendous relief when they canceled it. This thing was awful.
Eric Vigesaa (00:15:53): Had you admitted that to yourself, or was it more just like an underlying feeling that then was confirmed when it was actually cut?
Rob Collie (00:16:01): I knew that the product we were working on at the time just didn't work, but I had so little to go on. I didn't have a lot of experience, a lot of track record. So, maybe this is always how things work and we'll just get through it. It is always darkest before the dawn, so I didn't have the wisdom to be sitting on the sidelines rooting for a cancellation, thinking that was the right thing. But then when the powers that did cancel it, oh, it all just crystallized all at once, but it hurt many, many, many other people.
Rob Collie (00:16:36): Most people on that team were very disappointed that this had happened. I didn't come to work the day after the cancellation and everyone read that as, "Oh, the poor guy's just devastated." But I knew I wouldn't be able to come to work that day and hide my delight. I'm like, "It's just better to lay low."
Eric Vigesaa (00:16:55): That's funny. I feel like, tell me if you agree with this, there is a bit of program manager at Microsoft culture that people feel like they need to be the cheerleader, the evangelist, the believer. Sometimes that can cover up or squelch healthy doses of realism. I don't know if that was a factor in this case, but I'm well aware that some people believe that in order to be a PM, you've got to be rah-rah-rah, the true believer leading the charge.
Rob Collie (00:17:29): I get the sense that that has declined in intensity.
Eric Vigesaa (00:17:32): I think you're right.
Rob Collie (00:17:34): You and I both left the program, product management line of work at different points in time, but what we have in common is that when we left, it basically looked the same. We're basically talking about the same culture. You and I both had the same time capsule image of what that's like. Now, of course, you've stayed really, really close to Microsoft Corporate Centers, so you get a little bit more clarity into the changes than I do for sure. The job that you and I did, basically the same. So, you needed to be all in. You needed to be a believer.
Rob Collie (00:18:02): We don't need skeptics on our team. They just drag us down. Now you can label it a healthy skepticism. You can label it a wet blanket pessimism, and it's hard to tell the difference between the two, isn't it? The difference between the two is that one of them is correct. They might be exactly the same words. Now you got to figure out which reality it is.
Eric Vigesaa (00:18:22): That's right.
Rob Collie (00:18:23): There was another thing that one of my compatriots or my colleagues on that release that got canceled called Office Designer, and it was basically a competitor to SharePoint. SharePoint didn't exist yet. SharePoint became named during that same release. SharePoint made it to the release date. Office Designer did not. Exchange was going to be the server, the collaboration server center of the universe, not the Microsoft Office web server, which became known as SharePoint. In fact, Office Designer was in the running to get the SharePoint name. The SharePoint name was known to be such an awesome name that they'd really nailed it on the naming front. Now, they were like, "Okay, which one of you kids gets the good candy?"
Eric Vigesaa (00:19:03): I did not know this. This is great.
Rob Collie (00:19:06): So, one of the products turn out to be first thing to a billion in Microsoft's history famously, and one of them turned not to be nothing. Anyway, so one of my compatriots on that said, "One of the things I've noticed, Rob, when we're writing all these specifications for features is that the program managers who do a good job and write a really detailed specification about what's required to build this, they're at a disadvantage because their specs get costed by the dev team as being a lot more expensive than the ones who just sort wing it and say, 'Yeah, it's going to be easy.' They don't think it through."
Rob Collie (00:19:39): All the fractal discovery of all those million things that you have to do to make that product work that you would not bother to think of ahead of time, all of the specifications, all of the features eventually hydrate to their full form. But if you want to survive that early round of cuts in the old process, you just mail it in.
Eric Vigesaa (00:19:58): That is hilarious. So, I have my own way of internalizing what I think is that same concept, which is you better get your engineers excited about what's to be built here and want to build it. Because if they don't want to build it, there's a very easy way for them to do that. It's just cost it so pessimistically that it'll never get built. The inverse is also true. If you get them so excited that they want to build it, then they will be optimistic in their costing and it will get built. So, that was one of my very early takeaways is that getting your engineers on board was essential and it actually made things way more fun too versus the misery of always headbutting with the engineers over various elements of the design.
Rob Collie (00:20:48): Absolutely. You and I also, even though we have these, I hesitate to call them cynical because there's nothing cynical really about us. The viewpoint that we just described, if someone heard us describe this is how it used to work, they would think that we were really just trashing the whole affair. I'm like, "No, we're just really describing it how it really was. We're appreciating it in its true form." So it's not cynical. Cynical means you've got to be down about it, but we're actually smiling and giggling and this is the best thing possible.
Eric Vigesaa (00:21:19): Once you realize that that's the way it is, you actually can conduct yourself accordingly. You can work within that system. Not to say it's the best system, but if that's the environment that you're operating in, you can operate accordingly.
Rob Collie (00:21:31): Exactly. So, even though you and I can agree that that's how the feature prioritization process worked back in that snapshot of time, coolness matters. Nerd coolness mattered a lot. Don't overly specify it. Don't get super precise because it'll scare people. Get them on board so that they take their thumb off the scale when they tell you how long it's going to take them to do.
Eric Vigesaa (00:21:55): Exactly.
Rob Collie (00:21:57): In fact, not just take the thumb off the scale, start lifting up on the scale to make it even lighter, but compared to most people that we worked with, you and I, I think, going into all of this would've been most surprised to find that that's how it worked. We would've taken it at face value that this was an efficient process. We would've naively believed the official version.
Eric Vigesaa (00:22:20): Absolutely. Absolutely true. Yeah.
Rob Collie (00:22:22): More so than the average person. But by the time we were done, we were talking about the process in a way that the average person wouldn't. We'd move past and through their point of view, right? One day, we were average for maybe an hour. We were on the timeline and then we were off into the other.
Eric Vigesaa (00:22:42): Yeah. You think that was a naivete or it was just like, "No, we went in with a certain set of assumptions. We realized very quickly those were not valid. We adjusted and we figured out a way to move forward." It's a bit of a practical take on it, I think.
Rob Collie (00:22:58): I have a long term and still shelved passion project, but I want to get around to someday. It all started when I discovered that the domain name 3PO was available. I was just looking for new names for Power Pivot Pro, 3PS, P3. So, I was trying out different spellings of put the three first, put the three later. You got to dodge the domain trolls. I mean, they're everywhere, right? So now I am a domain troll because I discovered that if you spell out in letters 3PO, but put the dot before the io, that was available for $9 a year. It wasn't even a premium domain. So, our company P3 Adaptive, we own the 3PO domain.
Eric Vigesaa (00:23:45): Nice.
Rob Collie (00:23:45): The subtitle for 3PO will be translating the human experience. I think it's that journey from more naive than average, going in more naive than others. It's like this toes in the hair type of thing. It's very difficult in the early going. Everyone else seems to know all these unknown rules and no one ever tells you. It's such a surprise and it feels like a betrayal. Are you kidding? That's how you've all been acting all this time and no one ever bothered to tell me. That's my whole life up to a certain age. I still experience this from time to time, even in my middle age. I still occasionally run into instances like this. Oh, my God. Really?
Rob Collie (00:24:30): Very rarely anymore. Very rarely. When you have to go and do the hard work of reverse engineering something that everyone else is already navigating intuitively, instinctively, whenever those instincts were installed, the factory was off that day and didn't plug the chip in for me. I just said, "No, let that one go." Like Laverne and Shirley, they put the glove on the beer and it didn't get a cap or something. so doing the hard work of reverse engineering all these other things has this weird long-term advantage and that you end up actually understanding it sometimes better than the people who had it instinctively installed.
Eric Vigesaa (00:25:11): Wow. Okay.
Rob Collie (00:25:13): So that's the feel I get. Let me tell you how it really works. We could probably explain to others how it really worked back then better than any of those other people could have, because we had to do it the hard way.
Eric Vigesaa (00:25:24): Right. I had never thought about it in the way you're laying it out. That totally makes sense. It's really resonating with me. But I do have just one question related to the 3PO vision, and that is will you be translating into the language of moisture vaporators? Because I think that is a really a must have for your v1.
Rob Collie (00:25:45): Well, I do speak Bocce.
Eric Vigesaa (00:25:49): Excellent, excellent.
Rob Collie (00:25:52): 3PO can speak like 17 million different forms of communication, but all the only ones we know about are the binary language of moisture vaporators, Bocce-
Eric Vigesaa (00:26:01): And Bocce. That's right.
Rob Collie (00:26:02): ... and whatever they call English in Star Wars.
Eric Vigesaa (00:26:07): Anybody who has to think through something more explicitly, either because they got it wrong the first time or they just didn't intuit it, therefore they had to run it through a different part of their brain. I do think they have a different take on that. That reminds me a bit of my son who's on the autism spectrum. It felt like a prophecy at the time when he was evaluated and diagnosed that many of the things which most kids pick up intuitively, he will have to learn explicitly. The inverse would also be true. In other words, subjects which most kids need to learn explicitly. We may find that my son would learn intuitively. That ended up also being true. For example, reading. We never taught him to read, just simply read to him and he figured out how to read intuitively.
Eric Vigesaa (00:27:02): On the other hand, basic social interactions, there's no class for that. Read procedurals on how to conduct yourself in certain interpersonal situations. My son has been much more explicit like that. It doesn't mean he can't get it. Absolutely, it doesn't mean he can't get it, but needs to be taught in a different way. Anyway, just reminded me that we're all different and the way we process things are different, but that has lasting consequences, some of which could be very positive.
Rob Collie (00:27:29): Yes, getting through some of those challenges, you can end up in a better place on the other side than if you hadn't been challenged in that way.
Eric Vigesaa (00:27:38): Absolutely.
Rob Collie (00:27:38): I wish someone could go back and tell 24-year-old me, A, that this is what was happening to me, and B, just stick with it. The crisis of identity that I was going through around the time that you and I first met, it was brutal. Yeah, I can look back on all the stuff now and laugh, but I would not want to re-experience it. So, we had this release, this pie in the sky release that was in the very beginning was called Office.net, but they quickly pulled that back. It ended up being called just Office 2003, because in the end, there just wasn't a lot of net in Office 2003. Basically, the application shipped unchanged from their previous form.
Rob Collie (00:28:25): You and I and Chad and others had participated in exemplifying a lot of aspects of Excel, which allowed marketing to go out and talk about something as if we'd done something. But in the end, all of the functionality that we built, the stuff that's still in there anyway has been very deliberately hidden. It made really zero impact on the world. So, let me just call it a zero.
Eric Vigesaa (00:28:50): Very formative for us, but as far as output and benefit to the world, negligible.
Rob Collie (00:28:56): Incredibly formative. Yes, I agree. I appreciate them giving us a whole release of training.
Eric Vigesaa (00:29:05): Expensive, but hey, we were worth it.
Rob Collie (00:29:07): And then we got put on a real project, Excel 2007. Still to this day, the conditional formatting functionality in Excel is largely unchanged since your work. I can't detect any way in which it's been improved or modified.
Eric Vigesaa (00:29:24): Yeah, it has a few minor things, but yeah, very minor. Yeah.
Rob Collie (00:29:28): I still to this day show people one of my favorite buttons of all time, which is the on object UI that appears in a pivot table after you format a cell. I think you'll appreciate this. When I used to teach classes on Power Pivot, at some point in time during the class, I'd be sitting there with a pivot table and this pivot table would benefit from having data bars in it, the data bar conditional formatting in it. I would deliberately run through the clicks to format it.
Eric Vigesaa (00:29:58): Just to show off?
Rob Collie (00:29:59): Super fast, super fast. I go super-fast. I pretend like that was nothing. Okay, so I'm just going to go do this, right? I would just keep a straight face the whole time and just deliberately pretend that everyone in the room knows exactly about this. I would wait for the hand to go up. They would say, "Hey, that was really awesome. Can you show that again?" It was a way to almost prime the pump of people being willing to ask questions. I knew this would be so mind blowing for them to see it, that they would ask the question in the early going. There weren't many things that were like that where total impact of being shown these three clicks again would get them to change their lives.
Eric Vigesaa (00:30:39): Well, that easy to understand, easy to demo visual appeal I think is why it became one of the things in the release that marketing and the press and whoever else really latched onto. Not that there wasn't a lot of other great stuff in that release, but anything that's visual and easy to grok is easy to market probably, easy to write about.
Rob Collie (00:31:01): These are lessons that we learned from Dave. For the reverse engineers of humanity like you and I, these are really, really, really powerful principles that changed our worldview in a positive direction. For example, Dave's principle of, "Hey, all things being equal, which they rarely are, but all things being equal, the feature that's more visual and more conducive to screenshots. Could we screenshot it and put it on the back of the box?" We still talked about a box back then.
Eric Vigesaa (00:31:30): Yes, absolutely. We did. Yeah. That's going to be on the box, man. That's that good. That's that hot.
Rob Collie (00:31:36): Yup. Cardboard is not involved anymore. Shrink craft is not involved anymore. The more visual features should win in the contest for resources in that world. I also remember the early versions of the upgraded conditional formatting spec that you were working on was what we think of now as the managed rules dialogue. That was the spec. I was on board, I liked it. We went to Dave and he said, "Hey, everything that you would ever do in conditional formatting is now equally difficult. It's this one central interface for doing everything."
Eric Vigesaa (00:32:15): I see.
Rob Collie (00:32:16): We're sitting there in the room saying, "The tone you're using, Dave, sounds like you're criticizing it, but you keep saying things that sound like it's good." Why wouldn't I want an elegant place to do everything? And he said, "But what about all the simple things? That should really only just be one click, right?" Lasso a range of cells and slap a data bar on that. I remember you and I going, "You mean you don't like this beautiful managed rules dialogue?" No, no, no. I like it. It's an and not an or. That one little vignette, just so much repeatable wisdom.
Eric Vigesaa (00:32:50): Yup. He was right on, of course, in that.
Rob Collie (00:32:52): Oh, yeah, yeah.
Eric Vigesaa (00:32:55): Maybe most products are like this, I don't know, but Excel for sure. You got legions of basic users, but then you've also got rabid sizable groups of very Power users and you need to cater to all of the above. That's where Dave was going with that feedback.
Rob Collie (00:33:13): There's a principle in programming that inevitably infected the culture of design, of end user design. I've been noodling on this nonstop ever since that era. Under the hood, you only want one way to apply a conditional format to a range of cells. You only want one function call that does that. You want to elegantly design the software, the machinery under the hood. You would definitely not want a completely separate code path that applies the quick and easy stuff. So, fine. But that elegance of only one way, everything's the same, which you do want in the machinery. We were working in an organization that the programmers outnumbered us. There were more of them than us. They made up a heavier portion of the culture than us at the lunch table even, right?
Rob Collie (00:34:07): And plus, we'd come from that corner of the woods anyway. We were already at that point in our lives, like ivory tower academics of a sort. We would very hungrily sign on for this philosophy, and so too should the user interface be elegant and only have one way to do things. Well, that's just dumb. Under the hood and what works for the user are not the same. You've got to break that math. It's a different math.
Eric Vigesaa (00:34:36): Think of how many basic users would never have used any of that stuff if we hadn't provided that one click and apply.
Rob Collie (00:34:45): Yeah. Even in the pivot table case, pretty clearly, it wasn't discoverable enough because I learned very reliably that I could show this to a room of 40 people, every single one of them would light up and want me to show it to them again. I could tell on their faces that there weren't like three or four people in the room who were knew the secret. It was like everybody. That was the simple version, not the complicated version. So, look at all those lessons in product design all encapsulated into this fine little example.
Eric Vigesaa (00:35:15): You're totally right. You and I haven't spoken about this. I could tell it's still so fresh in your mind as it is in mind, but it's fun to hear what stuck with you over time.
Rob Collie (00:35:26): I think it's the cleanest example of the form is that feature and how it turned out. I wanted to mention to you that first of all, this is one of the best things I think anyone could ever watch. The other day, I watched a 90-minute interview between a football coach and three former football players, all of them from the NFL. I think it is the single most inspiring and amazing example of what a true leader, someone who is the upper 10th of a percent of a percent of a percent of leaders, maybe even in human history, is this coach, Mike Tomlin of the Pittsburgh Steelers.
Eric Vigesaa (00:36:06): Oh, I knew who that is.
Rob Collie (00:36:08): This 90-minute interview, think about it. Oh, my gosh, I would never watch a 90-minute interview. That's true. I would never watch a 90-minute interview, but I watched every bit of this. It's like the equivalent of a page turner.
Eric Vigesaa (00:36:17): Wow. No retreat.
Rob Collie (00:36:19): He says so many amazing things. First of all, he said some things like how he "loves" it when other coaches will say things about their players like, "Yeah, this guy's just not getting it. This player, he's not teachable." Tom was like, "What is our job if we're not here to teach? If we're not here to teach people who are hard to teach, what's our whole purpose? Remind me of that." He said, "So what you're doing when you're out there in public throwing someone under the bus, what you're really admitting is that you're teaching is failing and you're making it about the player because you're seeking comfort." He over and over and over again says throughout this interview, "I do not seek comfort. I've trained myself to not seek it. I run away from comfort."
Rob Collie (00:37:04): So, then the players ask him, "Okay, so you're patient. You take a lot of responsibility. It's not about the player wasn't good enough. You don't give up so easy, but you still have to give up sometimes, right, Coach, because you've got a roster of 45 or 53 and all these people in the training camp. So, how do you decide when to pull the plug? The other coaches do it too early." This segment of the interview is so good, you just have to watch it. But he says, "Look, there's no cookie cutter rule for when it's over, but when you're out of time, you're out of time. Not because of you, but because of things going on around you. The situation is what it is."
Eric Vigesaa (00:37:38): I see.
Rob Collie (00:37:39): If they decide that they're really weak at wide receiver and they need to keep seven of them just to hope that three of them turn out to be decent, that's going to subtract a spot. They're going to keep a player, for example, who is not as good as the one they're going to let go. They believe that objectively the one they're letting go is not as good, but because of just the utility of where they're going to be used and how they're going to be used. I think that something like that, things going on around you, I watched that play out a little bit with you on that Excel release. You revectored your career and you went "back".
Rob Collie (00:38:12): You had come to us from Microsoft internal IT, and you went back to Microsoft internal IT. by all accounts, you have thrived. So, tell me what's your job been like? We would still have a beer together from time to time, but largely professionally, when you went back to internal IT, that's like where I lose track of the story.
Eric Vigesaa (00:38:34): Got it.
Rob Collie (00:38:35): Are you comfortable with my characterization with the Mike Tomlin story?
Eric Vigesaa (00:38:38): Totally.
Rob Collie (00:38:39): I figured you would.
Eric Vigesaa (00:38:39): Absolutely.
Rob Collie (00:38:39): I wanted to be 100% respectful. Okay.
Eric Vigesaa (00:38:42): Yup. Absolutely. Going back to Microsoft IT, I could see a perspective of, hey, you tried to make it out there in the big leagues. It didn't work out, went back to the minors or whatever. But I won't put labels on it. I'll just say I think life in a lot of ways is the intersection of opportunity and your own agency. You can have opportunities or negative opportunities I suppose are a thing as well, but ultimately, it's not just that in isolation. It's what you choose to do about those opportunities, positive or negative. Turns out at this particular time in this particular part of Microsoft IT, we were at the very, very beginning of piloting something pretty. Just it turns out was pretty fricking cool. That is dipping a toe into online services.
Eric Vigesaa (00:39:35): I just want to point out, this is before the SharePoint engineering team was doing anything about online services. This was simply Microsoft IT saying, "You know what? We think we've gotten pretty good at operating SharePoint and Exchange and Office Communicator, but let's stick with SharePoint. We think we gotten pretty good at operating SharePoint for Microsoft. We'd like to try doing it for customers." So, this is taking our on-premises bits, the same bits that you'd run on prem, but in a Microsoft data center where we are operating those bits. That was going on. We had one customer at the time, so that's how early it was. It had passed a Bill Gates review to even get off the ground at all, but we had one single customer.
Eric Vigesaa (00:40:23): I have to say probably if I had to pick one phase of my career, which was the most exciting, felt fortunate and also felt like I couldn't get this experience, I can't imagine being anywhere else right now, it was figuring shit out in the early days, trying to stitch together an online service, wearing a ton of hats, feeling like a small business inside of a big, big corporation. It was so much fun, but also, wearing hats, it gave me the ability to try out a lot of different skill sets, try out a lot of different roles, and develop a point of view on, "Where's the need, but also where's my aptitude? Where do my strengths really seem to come to the surface?"
Eric Vigesaa (00:41:13): It may have taken me years to learn those things in another environment, but in that environment, it felt like an incubator for trying a lot of different types of roles and skills and seeing what stuck with me, seeing where I really wanted to take it further. So, incredibly exciting time. A lot of things happen between there and now, but I'll take a pause and just say, wow, you exit the product manager, program manager role, yet somehow you have this wonderful opportunity to really get in at the beginning of a huge transformation. It felt like incredibly lucky, but also I went all in on it and it's just served me very well to this day frankly.
Rob Collie (00:41:54): When you were saying an opportunity to figure out where your strengths lie or aptitude, completely unbidden. I had this fully formed Gary Larson Far Side cartoon jumped into my head of you sketched like Gary Larson style sitting at a desk and the caption is something like, "After a period of struggle and self-discovery, Eric finally came around finding his hidden strength." The picture of you is you, except you're sitting there. Your right bicep is just ginormous, right? I don't know where my strength is.
Eric Vigesaa (00:42:28): He finally noticed that it was his massive bicep.
Rob Collie (00:42:33): Quarter of your body weight is in the right bicep and you're like, "I'm still trying to figure out where my strengths lie." So, I want to react to a number of things there. The minor leagues thing, what's funny I just now realized is that I think I actually beat you. I went to the minors before you did. Before that release was over, I went to work on the Fantasy Football org and everyone believed I was committing career suicide at the time. I was probably just too naive to understand that they were probably right. Looking back, I'm like, "Oh, yeah, okay, I see what they're saying."
Eric Vigesaa (00:43:04): I had totally forgotten about that. Yeah.
Rob Collie (00:43:08): Another thing is that what year approximately were you involved or years with running the on-prem SharePoint bits in a cloud light capacity for a customer?
Eric Vigesaa (00:43:21): It was for sure the first half of the 2000s. How far into the 2000s it went? I don't recall at the moment, but for sure, early to mid-2000s.
Rob Collie (00:43:32): Like 2008?
Eric Vigesaa (00:43:33): Probably even then. Our earliest customers, I'm remembering this now. Were on three different versions of SharePoint on-prem bits. It was a long time and we have a later phase of when we finally did have a cloud first offering that was ready, feature rich enough, and ready to meet the needs of these big enterprise customers because that's what they were. That was a multi-year project to transition them to that offering, but I just can't tell you how exciting it was and just how much I treasure that time.
Rob Collie (00:44:04): I believe it.
Eric Vigesaa (00:44:05): It's hard to imagine just my life in general, certainly my working life without that experience. That was pretty freaking cool.
Rob Collie (00:44:12): What was the official motivation for MS IT doing this? Was it let's go make some revenue? Was it let's go be Guinea pig researchers for the SharePoint team to see what it's like to bring essentially real customers in house in a way so that it was more transparent for the engineering team to learn from? Was it a combination of factors? You've seen Kill Bill.
Eric Vigesaa (00:44:36): Oh, yeah.
Rob Collie (00:44:36): Where Michael Madsen asked Daryl Hannah, now that Uma Thurman's character is dead, which she isn't, are you feeling relief or disappointment? And he says to her, "Don't weasel out and say you're feeling both because we both know you're feeling one more than the other." So, what was the official on the record motivation for Microsoft IT to do such an outlandishly different thing at the time?
Eric Vigesaa (00:45:01): Yeah. There's a story that gets repeated a lot that I'll tell. I believe though, first and foremost, it was IT wanting to not just be a cost center, but to also make some money. IT had a bit of a chip on his shoulder, I think, maybe back in the day about just being a cost center and maybe we didn't have our best and brightest and most creative minds there. I wouldn't necessarily agree with that, but I think that was a motivation. There is a story that gets told and that is there was, perhaps you remember, I believe it was called the SQL Blaster virus.
Eric Vigesaa (00:45:36): It was a particularly bad virus in terms of damage done in the industry. It took down a lot of our customers, like a lot of our big customers, SQL of course being the backend for SharePoint among many other things. In contrast to our customers, Microsoft's own implementation of SharePoint survived Blaster very, very well. Why is that? You might ask. Well, it's simply because we were very diligent about keeping our machines up to date, patched, et cetera, because the latest patch levels were actually protected against Blaster.
Eric Vigesaa (00:46:12): It was only when you were not on the latest versions that you were vulnerable and somewhere that got remarked on by a Microsoft executive and that sparked a thought in one or more people to say, "Hey, we're actually pretty good at running this and we think we're actually better than our customers. We can offer our customers a better experience than they're getting from running SharePoint themselves. Oh, by the way, why are they spending energy operating SharePoint instead of just being a consumer of it?" So anyway, that story, I don't know how influential that was, but I think that makes a lot of sense to me.
Rob Collie (00:46:44): It does.
Eric Vigesaa (00:46:45): That it gave an opportunity to say, "Hey, it takes a certain amount of diligence and knowledge of best practices to really do this well. Our customers should not to worry about that and left on their own devices. They're actually not doing that at the highest level and they're paying the price for it."
Rob Collie (00:47:00): That makes a lot of sense to me. I do secretly hope that somewhere the idea started off being pitched with, "Wouldn't it be cool if..."
Eric Vigesaa (00:47:12): Maybe it was. It wasn't until the Blaster thing came along that there was enough credence.
Rob Collie (00:47:18): Now is the time, now is the time to strike, my children.
Eric Vigesaa (00:47:19): That's great.
Rob Collie (00:47:24): Oh, that's great. A lot of what you were saying as well about that experience, that green field, fresh territory like uncharted domain was both super important to you career wise, but also intoxicating in the moment.
Eric Vigesaa (00:47:40): Yes, absolutely.
Rob Collie (00:47:42): I get it. It'd be a bit of a stretch as an analogy goes. That's what I do. One of the things we've been talking about on this show recently is how I probably wouldn't get into DAX, the formula language of Power BI. I wouldn't take to it if I joined the story today and looked around and saw what everyone was up to with DAX. Now, in the real timeline though, one that we actually lived, I still probably have written more articles on DAX than anyone, because I was doing two a week for years.
Eric Vigesaa (00:48:17): Oh, wow.
Rob Collie (00:48:18): I wrote a book that now that the market's become fractured across many, many, many books, I wrote a book that still might be the all-time best seller on DAX. At least it would be in that conversation. I haven't checked the stats in a long time. So, I definitely got into DAX, right? I mean there's no two ways about it. But if I had come to a table that was already set, I probably wouldn't have. There's two components to it. I think one is the excitement of exploring that new territory provides an energy. It provides a motivation. It provides a drive. That's the fun thing to talk about is that component. I like that stuff. The one that doesn't speak is highly of me.
Rob Collie (00:48:57): When there's already a tremendous amount of prior art and the community coalesced around something already, there's an opportunity to be judged. There's an opportunity to judge your own work, your own knowledge against others. When you're out and there's nothing to compare yourself to, there's a benefit there. I think I've learned this about myself. Today's DAX that is written on blogs and shared in YouTube videos, there's a strong self-regulating pressure in the community for that DAX to be optimal. Not most understandable, not most brute force, let's get the job done. For me, DAX is like an alley fight, down and dirty. It's like, "Let's get this done. Who cares how many corners we have to cut as long as it's always going to work? If it works, it works." That's the test.
Rob Collie (00:49:50): If it works and stays working, great. So, I've been joking about we need to have a YouTube channel or a blog or something that's dedicated to bad DAX, because we've developed an unconscious gate keeping mentality. I think making it harder for others to see themselves as part of that community in the future. Can I get to be like them? The answer might be no because there's a handful of luminaries out there. So, does that resonate with you as well about that greenfield experience? Do you see both components?
Eric Vigesaa (00:50:24): Yeah, let me put it in a different framing just because it is how I think about it. You tell me if I'm touching on some of the same core things. This actually gets into something that's so deep in me now, I think, in my DNA, particularly work DNA. That is it is completely different mental approach and just different mindset and I would argue different type of person in each of these approaches. One of them is you're jumping into a problem space out of necessity because you're close to the customers or you're close to the problem anyway that needs to get solved. You're in there, you're rolling up your sleeves, you're iterating. You've got constraints, budgets and time pressures, customers waiting, and you're figuring stuff out as you go.
Eric Vigesaa (00:51:17): For some people, there's an exciting energy that feeds off of that. For others, maybe it's chaotic and they want to take a step back behind the curtain to think in peace and quiet maybe. But for those of us who love that, there's an energy and a sense of urgency and you're always the problem. The customer with the problem is front and center because they're very close to you. You're trying to make it work for them. I think the other one is a bit more white lab coat. Let's give somebody the time and the budget, a big brain, and let's have him go design something awesome. Some people thrive in that environment. For me, going back to your Gary Larson big arm, the other arm was probably this wasted little thing.
Eric Vigesaa (00:52:01): My big arm is the necessity, feeding off the energy of making something work, figuring out for that customer in the moment, and then trying to learn those lessons in a more structured way. Man, that arm though, where I've got all the time in the world and I'm the guy in the white lab coat, that is an underdeveloped muscle for me. I thrive, I feed off of that energy and that sense of urgency from our customers and from our problems. I'm a frontline guy in that sense.
Eric Vigesaa (00:52:31): I don't know if that's the right metaphor, but to me, that's massive. You're of this ilk or that ilk and I just feel like there are two different mindsets. We need people of both sorts. But I'm clearly from this formative experience where throw me in there with urgent business problems to be solved and we got to figure shit out. So, I don't know. Does that resonate?
Rob Collie (00:52:56): It does.
Eric Vigesaa (00:52:56): Am I making a slightly different point?
Rob Collie (00:52:58): It is a slightly different point, but it's still really important access to look at. So, I also want to say, I don't think this is an either or thing. You showed up just like me. You had your white lab coat. You would've been a fine long term white lab coater. Let's not oversimplify the narrative. I think you could and would have continued to do very well in the white lab coat world of building the software. By the way, their philosophies on how to build software have become much closer, not as close as actually solving the problems, right? But it got much, much closer. So, a lot of the bullshit we were indoctrinated into, wouldn't it be cool if and all nerds fear me, I have best nerd idea. All of that stuff has probably not been completely eradicated, but it's definitely been pushed to the margins.
Rob Collie (00:53:50): There is definitely a difference and I've experienced it as well. Call it just like distance from the actual impact, distance from application. I find that it's emotionally far more satisfying to be closer to the impact.
Eric Vigesaa (00:54:06): 100%.
Rob Collie (00:54:07): I suppose if you got super, super, super high in the organization at Microsoft and you could say, "I'm literally in charge of entire applications or entire products," that's got to scratch the narcissism. That's got to be good for the ego. That might be pretty satisfying. But I think we're much more wired as humans not so much for achievement as for being useful as part of the hive. So, being close to impact, me out here in the consulting world, I found a joy in using the products that I had helped build that building the products would never have ever scratched. I mean not have even come close.
Eric Vigesaa (00:54:52): I would imagine a much clearer sense of their product strengths and weaknesses than you ever would've had.
Rob Collie (00:54:58): That's right. I like to joke that only now am I qualified to do the sorts of jobs that Microsoft used to pay me to do.
Eric Vigesaa (00:55:05): That's the supreme irony, right? Yeah.
Rob Collie (00:55:08): But at the same time, I probably wouldn't enjoy those jobs as much as I used to. Maybe not. I don't know. It's a complete unknown, let's call it an unknown, right? You know what though, honestly, now that our company's grown and I'm more of an executive than a doer, I'm back to distance from the impact again.
Eric Vigesaa (00:55:27): Maybe you're more able to handle it because I think you carry all that experience with you that's now part of your being, whereas that wasn't true in your earlier days.
Rob Collie (00:55:37): Yeah, I mean I guess it makes sense then that we do things like this. We have the podcast. I have a super-secret skunkworks project that we're going to finally take out a stealth mode here hopefully pretty soon. It's meant to connect with people. Even if I can't scale my ability to converse with everybody, I do have a need to be directly impacting in a positive way other people's thinking.
Rob Collie (00:56:01): So, I need to find ways to scratch that itch and find them in a way that that also supports the company, which isn't really that hard to do to align those things. So, 3PO for instance, a series of blog posts about figuring out how the world really works and explaining it back to others, sharing my learning experiences. I don't think that really supports the company very much. Boy would it scratch my itch. I have to earn the right to someday go do that.
Eric Vigesaa (00:56:28): By the way, short lived breakfast cereal, 3POs. Just wanted to let you know that.
Rob Collie (00:56:32): Oh, really?
Eric Vigesaa (00:56:33): Yeah.
Rob Collie (00:56:34): Is this a dad joke? Are you setting me up for something?
Eric Vigesaa (00:56:36): No, it's real. I just thought of it.
Rob Collie (00:56:39): Do you know why Yoda was afraid of the number seven?
Eric Vigesaa (00:56:42): Why?
Rob Collie (00:56:43): Because six, seven, eight.
Eric Vigesaa (00:56:48): Nice. Nice. Okay, that's good. I'm telling my son that one.
Rob Collie (00:56:52): All right. Because normally, the joke is why was person X afraid of seven because seven, eight, nine.
Eric Vigesaa (00:57:00): Seven, eight, nine, right?
Rob Collie (00:57:01): They're expecting that joke. Actually, the Yoda joke is even funnier if they've heard the original.
Eric Vigesaa (00:57:06): You get the Yoda ease.
Rob Collie (00:57:09): The best part is when they go, "Is it just because seven, eight, nine?" No, because six, seven, eight. 3POs were a cereal.
Eric Vigesaa (00:57:19): Indeed, A little bit of a rip off. They just took an existing cereal and then rebranded it. It wasn't very creative.
Rob Collie (00:57:27): Did they look like 3PO?
Eric Vigesaa (00:57:28): Not really, but it was almost two Cheerios stuck together into an ate thing.
Rob Collie (00:57:34): I see. Yeah. Okay. Yeah.
Eric Vigesaa (00:57:36): So maybe vaguely droid-like, but a poor effort.
Rob Collie (00:57:39): So you were involved in running on-prem SharePoint bits in essentially like a private cloud capacity way.
Eric Vigesaa (00:57:47): Yes.
Rob Collie (00:57:47): Guess what I was doing in 2010?
Eric Vigesaa (00:57:50): Oh, my goodness.
Rob Collie (00:57:50): I was running in a co-located data center Power Pivot for SharePoint with customers using this from hundreds of companies.
Eric Vigesaa (00:58:02): I know from a previous podcast that you found that incredibly easy, fulfilling.
Rob Collie (00:58:08): Yeah. Just shoot me.
Eric Vigesaa (00:58:10): Yeah, it was not one of your best work memories.
Rob Collie (00:58:14): That was the worst. So, I take it that MS IT never got into the game of running that particular type of config for anybody.
Eric Vigesaa (00:58:24): We never hosted Power Pivot for SharePoint. If I could genericize the point and this will sound maybe very quaint now, but at the time, it was a big set of problems we faced is what subset of features and controls should we expose, should we allow our customers to fiddle with and which should we not? Of course, we would've liked to have given our customers the ability to fiddle with everything, but we couldn't operationalize that or we couldn't continue to maintain our availability SLA or we couldn't figure out to make that work in a hosted model.
Eric Vigesaa (00:59:02): We had essentially a subset of the features in SharePoint and a subset of the admin controls that we allowed our customers to change, and then we had operational process around that. So, essentially, our feature roadmap was how do we unlock more and more of the capabilities in the native product? And so for years, that was one of our biggest challenges is how to have that subset be as big as it could possibly be.
Rob Collie (00:59:33): So the Power Pivot integration, this is obviously tongue in cheek. You mean the feature that in order for it to work properly, you had to go and basically set about a dozen other SharePoint settings in the exact opposite of their normal recommended operating condition.
Eric Vigesaa (00:59:50): Oh, geez.
Rob Collie (00:59:52): You all didn't get around to that one, huh?
Eric Vigesaa (00:59:53): No, no. I think we must have determined that was not friendly in a hosted offering for us.
Rob Collie (01:00:01): First step, crank the file size limit up to its max of two gigs. Second step, go into Excel services and crank up its maximum. Our customers were in a 100% read only mode. They had no capability to create new sites, no capability to create new users, no new lists, nothing. All they could do was consume the reports. They could click slicers. They could interact with the Excel services front end. They certainly couldn't upload new workbooks or anything like that. All of the writing happened on our side. It was 100% read only, and we still couldn't keep that thing up and alive.
Eric Vigesaa (01:00:38): That is brutal, man. There's something back in the day called right back where you could enter in an Excel. You could modify a value and have that written back to the data source.
Rob Collie (01:00:48): Oh, this is a legend. We were always singing songs of this ghost.
Eric Vigesaa (01:00:54): It was a mythical thing.
Rob Collie (01:00:57): As far as I can tell, the platform is still reliant on third parties for this or at least heavy customization through Power apps and things like that. It's roll your own right back. It's certainly proven to be a difficult problem. The famous Microsoft product that did right back in planning and budgeting, Performance Point, got abruptly mothballed, one of those changes at Microsoft that caused more harm in the community. It would rank highly on the hall of fame of damaging careers outside of Microsoft, the end of Performance Point.
Eric Vigesaa (01:01:35): Because we pulled the plug so abruptly and changed course you mean?
Rob Collie (01:01:39): Well, I mean there were entire companies and consulting companies and software companies that had built themselves up around being part of this ecosystem. When Microsoft killed it, these companies had no reason to exist anymore.
Eric Vigesaa (01:01:51): Got it.
Rob Collie (01:01:51): Entire companies vaporized.
Eric Vigesaa (01:01:53): Wow.
Rob Collie (01:01:54): It'd be the equivalent of Microsoft pulling the plug on Power BI for my company and that's not going to happen. We've bet on a pretty reliable horse.
Eric Vigesaa (01:02:03): Yeah. Way beyond critical mass, that's for sure.
Rob Collie (01:02:06): Yeah. I think it's going to stick.
Eric Vigesaa (01:02:08): Yeah. I think you're safe there. You're safe there.
Rob Collie (01:02:12): Yeah. You really can't go more than a year in the community without running into someone that bears scars of the Performance Point cancellation.
Eric Vigesaa (01:02:21): I know that it was not our team, but a team that we knew about. Microsoft did try out... I don't know how many customers they had. It might have been very, very few. ... a hosted Performance Point offering that also didn't last very long.
Rob Collie (01:02:35): Yeah, it was weird, right? There's that arrow where everything at Microsoft, if it was going to have a web presence of any sort, intranet or otherwise, it was like a commandment. You had to integrate it into SharePoint. The SharePoint API and extensibility model of then is very different from the one today. The one today is well thought out, and the one from before, it evolved the Winchester mansion in California.
Rob Collie (01:02:59): And then these other teams wouldn't be necessarily SharePoint experts. However good the programming and development model was for SharePoint at the time, these teams would never quite live up to it anyway, but this incredibly loose leaky integration. They weren't jokes. They were real. We just eventually just capitulated and had scripts running that were just checking the health of services, and if one of them's dead, we just restarted it. Don't even try to diagnose.
Eric Vigesaa (01:03:27): Oh, geez.
Rob Collie (01:03:27): Eventually, there was no chance we were going to ever diagnose it and robustly correct it.
Eric Vigesaa (01:03:32): How long were you in that business? Did you have that offering up?
Rob Collie (01:03:37): So I was there for three years at this company that was, in hindsight, really just my first client for the consulting company. This was the incubation client. I incubated their business and they incubated mine. We didn't know that at the time. That wasn't our business deal. That's what it turned out to be. They went on and continued after we had our creative differences like The Eagles and Behind the Music or something.
Eric Vigesaa (01:04:05): Behind the Music. Yeah, that's a good reference.
Rob Collie (01:04:07): They continued. I think at some point, they inevitably switched to the Power BI service. I wasn't there when that happened. They kept things running well enough in the meantime on Power Pivot for SharePoint that they made it. I think they got bought and then someone bought that company and now someone's bought that company. So, they've like Russian dolled three times since then. So, I guess things worked out for them.
Eric Vigesaa (01:04:34): It was an important period for you, it sounds like. Even though you went on to do different things, that was a key time.
Rob Collie (01:04:40): I was convinced in early 2010 that Power Pivot even, we didn't have Power BI, we didn't have Power Query, but I was convinced in 2010, early, early 2010, that Power Pivot was going to dominate the BI world. I'd seen enough, and so I just figured that the world would get around to very efficiently reaching this new equilibrium. Maybe it'll take two and a half years. So, for three years, there was only enough demand in the market really for me to work for this one company and demand started rising. By 2013, early 2013, there was sufficient demand in the market that I had the ability to branch out, turn what had been a blog into a business.
Rob Collie (01:05:25): So, yeah, I mean I would've needed to do something for those intervening three years that I don't know what it would've been. If it had been anything other than what I was doing, I would've lost the thread. I had a reason to get up and work on DAX and on data modeling and on SharePoint every day for a paycheck. To build a company that I thought was someday might even be so successful, that would be the exit and make enough money at that company. I ran out of patience.
Eric Vigesaa (01:05:53): Not being overly modest. Do you feel like you influence the evolution of the development of where Microsoft was going in the BI space during those years?
Rob Collie (01:06:02): Probably not very much. I would say let's start with zero and then try to find a couple places where maybe things were a little bit different because I was doing what I was doing. I don't know, because the thing is they would never really tell me and not because they were stingy. They would never tell me when something that they'd learned from talking to me or watching me made a difference.
Eric Vigesaa (01:06:23): I see. Yeah.
Rob Collie (01:06:25): It's not part of their workflow.
Eric Vigesaa (01:06:27): You told them a lot of stuff. They were listening, but what actually they did with that or didn't do with that is a pick.
Rob Collie (01:06:32): I certainly noticed the places where they didn't listen. I got plenty examples of that. I remember Jeffrey Wang asking me, just during a training that I gave near Microsoft to some Microsoft people, he asked me, "What about you said this idea? If we gave you this concept of variables that you could declare at the beginning of the formula and they got evaluated once right up front and then they resolved to that value from that point forward, would you use that?"
Rob Collie (01:06:56): I'm like, "Yeah, I would totally use that. Oh, I totally get what you're going. I think that's awesome." They probably would've done variables anyway, right? I mean, if anything, maybe I help nudge them a few weeks faster towards the same conclusion they would've reached anyway. I think it's one of those things that's like it's inevitable. It's going to run its course.
Eric Vigesaa (01:07:15): And then another question. How do you feel about the gap that you fill between the software and customers? In other words, are you thriving in part because there is a big gap there and Microsoft could have and should have gone further and made it easier? It would've cramp your style, but because they didn't quite get all the way there, it's a more lucrative space for you to have been able to operate.
Rob Collie (01:07:42): Yeah, it's a really fascinating question because I worked on the product team, I would've never been able to come around to anything resembling a confident feel here, confident maybe. Correct and confident, probably not.
Eric Vigesaa (01:07:54): Yeah. Naively confident.
Rob Collie (01:07:57): Confidence, no problem. Here's the way I look at it now. It would take a lot to change my mind on this. A lot of new developments, tremendous breakthroughs would be required, earth shaking breakthroughs. Making data work for you in any given specific business environment is always going to be difficult. It's never going to be easy. Stop waiting for easy, stop hoping for easy. Now, as a software vendor, there's this tragedy of the commons type of thing where everyone has to go out and lie that their software makes it easy. The lesson of Tableau is that they sold the lie better than anyone else. They committed to the lie and it is a lie. It is dishonest.
Rob Collie (01:08:54): They committed to the lie that now data will be easy and we'll make it look easy by having the price be super high and a team of dedicated technical sales people in the back frantically working to solve your business problem so that we can come in and give you a demo with your data that looks like click, click, click, click, never mind the potentially hundreds of person hours that you didn't see. Okay. It is a lie. It will always be a lie. Every time Microsoft tries to introduce something that makes things easier, it just really increases the surface area for people buying in on the platform and discovering that it's not easy. Now, it's easy.
Eric Vigesaa (01:09:32): Wow.
Rob Collie (01:09:32): It's easier than previous tech. The citizen developer thing is real. I took to DAX. I recently had Dany Hoter on. He tried to teach me MDX like 10 times. I failed. It's funny. He tried to teach me 10 times and I failed 11 times. I don't know how I failed more than 100% of the attempts.
Eric Vigesaa (01:09:56): You overachiever, Rob.
Rob Collie (01:10:00): So it has gotten easier, but I mean the cold realities of all the nuance and all the difference between every single business environment, it's the same thing with SharePoint, right? Before SharePoint went cloud and every company had to run their own instance, the moment someone got good at SharePoint working for a particular corporation, like the moment that Skynet went aware, right? The moment they got good at it, poof, they're gone. Because now they're worth way more to a consulting org than they are to anyone client, right?
Eric Vigesaa (01:10:35): Totally.
Rob Collie (01:10:36): That was primarily because running the infrastructure was so damn difficult. Oh, yeah, you need the right number of spindles and the blah blah, blah. It was racking and network topology problem. Then when Microsoft made all that cloud stuff, it took that problem away. I mean, a lot of these people had to go. They went and found new jobs, but we're the ones writing scripts and formulas and pipelines and integrations and things like that.
Rob Collie (01:11:04): Sure, there might be a breakthrough out there coming for us that writes good formulas for you, that understand your intent, that read the pressure of your fingers on the keyboard and the timber of your voice as you're instructing it and say things like, "Oh, I know Jimmy says he wants this, but we know Jimmy. What he really means is this." I think by the time a company like P3 is no longer needed, nothing is needed, because that's when general AI has solved everything. Look at how hard it's been to get the last 10% of self-driving car problems worked out. I think they're discovering what they need is general AI for that last 10.
Eric Vigesaa (01:11:49): I think you're right. Yup. Yup.
Rob Collie (01:11:51): They're going to solve the self-driving car problem long before they AI me out of a job. I even have a blog post where I call it one of the three big lies in data is that this new thing's going to make it easy. No, never, never, never. Introduce quick measures. Fine. It's not a threat. What is going to end up being, in my cynical opinion, is another really, really slick demo someone can give to get someone to buy the product. Now, once they're in-
Eric Vigesaa (01:12:18): When you want real value, you've got real questions to be asked and answered.
Rob Collie (01:12:21): That's right.
Eric Vigesaa (01:12:21): Interesting. Yeah. I'm hearing your tone. That's not just around the corner. There are real complexities and nuances here that you don't see tech filling that gap anytime soon.
Rob Collie (01:12:32): It's in the space of things that humans uniquely do well relative to machines. I think it's as close to the top of that pyramid as you can get, very improbable that there's some breakthrough in technology that drowns us and doesn't drown everything else. Microsoft is committed as a business to the idea of you needing less and less consulting. It's this theory of compliments in economics. Great example is if the only use of peanut butter in the entire universe is to make peanut butter and jelly sandwiches and the demand for peanut butter and jelly sandwiches is what drives the demand for peanut butter. Jelly is also only used for PBJs, and the amount of money it takes to populate a PBJ, the ingredients, it's like a dollar of jelly and a dollar of peanut butter.
Rob Collie (01:13:23): Let's ignore the bread, the average consumer on the street. The supply and demand curve works out so that they will buy seven PBJs a month, and then there's a breakthrough in jelly production that cuts the price of jelly to let's say, 25 cents per sandwich from its original dollar. What's going to happen is that I originally had $14 a month to spend on PBJs. It's money that I was willing to allocate for the PBJs, right? But now each PBJ that used to cost $2 is going to cost a $1.25. So, my hunger for PBJs is going to equalize at higher than seven PBJs a week now. So, the demand for peanut butter goes up. Nothing to do with peanut butter. Only jelly changed and the demand for peanut butter goes up, right?
Eric Vigesaa (01:14:20): Got it, got it. Because the product being consumed is not peanut butter, but this composite thing called PBJ.
Rob Collie (01:14:29): Yeah. So, Microsoft software is just the jelly or just the peanut butter. It's the peanut butter in this example, right? So if you make consulting cheaper or less necessary, this means more demand for your product.
Eric Vigesaa (01:14:43): You think it's a healthy goal or maybe unachievable, but Microsoft should be trying to come up with more complete solutions that require less complexity from consulting, right? That's a noble goal.
Rob Collie (01:14:54): Totally. Yeah. That is what the market "wants". Think about it. The transition from MDX to DAX is a perfect example of this. There were probably like 2,000 people in the world who claimed to know MDX and maybe 10% of them actually did. The world's hunger for dimensional data models with an OAP query front end was far greater than what was being represented in the actual transactions. DAX is a breakthrough that drove the price of consulting dramatically lower, breakthrough in jelly production.
Rob Collie (01:15:35): Even just the invention of DAX itself is 100% motivated by this and look how much good it has created. If you really wanted a good MDX solution built and you knew what you were doing, you were going to have to wait for one of the top 200 to become available and then win the auction for their time.
Eric Vigesaa (01:15:56): It's that scarce of a resource.
Rob Collie (01:15:59): Yeah. Now, of course, what you're really going to do is hire someone that doesn't know what they're doing.
Eric Vigesaa (01:16:04): That's the more likely outcome. Chaos and tragedy ensue, right?
Rob Collie (01:16:10): Yeah. Yeah. Bring it on. That mentality created the space. The universe in which P3 can exist was created by that spirit, so just keep after it. The people with the data gene are always going to be at the interface between the software and the problems. That's us.
Eric Vigesaa (01:16:28): By the way, I think my son is a budding data disciple. When he was younger, he used to maintain a ranked list of his favorite people. If I would have to make a tough parenting call and sometimes deny him something he wanted, he would let me know that following through with that decision would change my ranking. Was I prepared for that? Was I prepared to go down from number three to number six if I denied him something he was looking for? Maintaining those rankings was very important to him.
Rob Collie (01:17:01): I love this. Were there columns and attributes, or was it just like, "Look, I've got a single column list of people ranked"?
Eric Vigesaa (01:17:11): Well, this is where I feel like perhaps you could've guided him in ways that would've been eye-opening for him. Different ways to visualize the data trending. I should have had my own conditional formatting stuff in there like trend lines, like spark lines. Eric's the climber these days. Look, he's gone from number 10 to number 2 just in the last three weeks. Watch out, number one.
Rob Collie (01:17:34): You can do things like plot it by hour of the day. Look, dad, you always tank at bedtime, right?
Eric Vigesaa (01:17:43): That's right. I feel like this could be a Calvin and Hobbes. He was always critiquing his dad's performance. Remember? He didn't use enough data though. We should have more data in there.
Rob Collie (01:17:52): So I still got to know though. Did he just keep this in his head or there was a scoreboard on the wall?
Eric Vigesaa (01:17:58): It was written down on paper. Later he had it on a whiteboard. He keeps other tabular data sets in Excel, but he wasn't using Excel back in those days when he was ranking family members.
Rob Collie (01:18:11): I mean, just think about it and I'm sure he was frustrated by how clumsy the updates are. If he wants to move you down three spaces, he's got to change a lot of people. Everybody in between has to be erased or crossed out.
Eric Vigesaa (01:18:24): This is a big problem. So, he's a guy after our own hearts. Somebody who's close to the problem and I have high hopes for him, Rob. We have a Peloton in our house, and he spends at least as much time reviewing his graphs and assessing his performance as he does actually on the bike.
Rob Collie (01:18:46): As he said the words, I can't believe they don't let me export this.
Eric Vigesaa (01:18:51): Yes. Yes, he has. He has, because there's this overlapping of different data sets. There's the output that you have. We have a heart rate, so there's a heart rate data. There of course is resistance and cadence. There's all these things, and he's not able to manipulate these things in ways that he wants to.
Rob Collie (01:19:13): This is basically an IoT level of data field. He's got sensored data, right?
Eric Vigesaa (01:19:19): Yes.
Rob Collie (01:19:20): He needs to be storing stuff in Gusto, Dany Hoter's new-
Eric Vigesaa (01:19:25): That's right.
Rob Collie (01:19:27): You're firing up the KQL.
Eric Vigesaa (01:19:31): That's right. I'm telling you, I need to get this guy. There's such a clear transition from these childhood problems to career and data. It's so clear to me now.
Rob Collie (01:19:41): So, Orange Theory Fitness is similar, right? It generates a lot of data, this gym that my wife and I go to. They've had these water rowers forever, and then one day, the water rowers now have iPads clipped to them. So, to see how far you're going and what wattage you're producing and all that. You're no longer looking at the little, tiny LCD readout that used to be there. It still is now. It just shows nothing. Of course, I go and I look. I'm like, "Oh, look at this. This whole time, this water rower had this USB interface that you could plug into." They didn't replace the rowers, just a peripheral.
Eric Vigesaa (01:20:20): I see. They just plugged an iPad into that interface that was always there.
Rob Collie (01:20:23): You start thinking, right? It was like, "Okay, look. I go in there with a USB pass through device, right?" I'll plug my little USB dongle in, put it in between the rower and the iPad, so that it will log for me down to the second every last thing that's going on.
Eric Vigesaa (01:20:43): Get all the raw data.
Rob Collie (01:20:44): Yeah, because the iPad has just given you the summary. As soon as it goes by, it's gone. Now, at the end of the day, they tell me how many calories I supposedly burned and what my heart rate was on average and how many minutes I spent in each heart rate zone. But if I wanted to know how many calories I burned in minute 31 or what my heart rate was at 31 minutes and 12 seconds, that's lost. It's just lost. Like tears in rain to quote Roy Batty from Blade Runner. It's just gone. What a shame. I'm generating all this data and they're just throwing it away.
Eric Vigesaa (01:21:20): Yeah. Yeah. There's a career there for innovative minds to unlock these data sources.
Rob Collie (01:21:29): You know what the addressable market is for obsessing over the data out of Orange Theory? It's like me and you and your son.
Eric Vigesaa (01:21:38): Okay. Well, maybe that was the addressable market, whether we knew it or not, back in that first release of Excel.
Rob Collie (01:21:43): Well, Azure data market went and spent a lot of money to find out the same thing.
Eric Vigesaa (01:21:47): That's true.
Rob Collie (01:21:49): Again, we did it. We did it on the cheap.
Eric Vigesaa (01:21:52): Fail fast, man. Fail cheap.
Rob Collie (01:21:54): Exactly. Well, Eric, it's been a long time. I've really enjoyed this.
Eric Vigesaa (01:22:00): Guys, this is so much Fun.
Rob Collie (01:22:01): Thanks for listening to the Raw Data by P3 Adaptive Podcast. Let the experts at P3 Adaptive help your business. Just go to p3adaptive.com. Have a data day.
Sign up to receive email updates
Enter your name and email address below and I'll send you periodic updates about the podcast.