Raw Data By P3 Adaptive
Stay Away from Guadalcanal, w/ Olivier Travers
MS Data Platform International Man Of MysteryListen Now:
For the second part of our 2022 New Year double feature, we sat down with Olivier Travers. Olivier has had quite the data journey that has taken him from France to Germany, Portugal, Italy, and finally, to Chile where he serves clients all over the world as a Business Intelligence advisor and consultant…an expert in Power BI, he’s focused on data visualization, SaaS integration, and software architecture! He also knows quite a bit about MMA and Brazilian JuJitsu, and a spirited discussion on who would win in a fight ensues!
Check out Olivier’s website HERE!
References in this episode:
Rob Collie (00:00:00): Hello friends. If you don't yet know why we're releasing two episodes in a single week, just go back and briefly listen to the intro at least, to the Jeff [McNeily 00:00:09] episode. Our guest for this episode is Olivier Travers. Now, Olivier is kind of like a Microsoft data platform international man of mystery. And of course that's the subtitle to Austin Powers. But I think Olivia is a little bit closer to the James Bond end of the spectrum that inspired that. He's just your average French citizen who moved his family to Portugal, and now has spent the last decade and a half in Chile. Pretty standard stuff. And that international flavor pulled us in some unexpected directions. We pretty quickly got into the topic of international trade and the role that different currencies from different nations play on that stage. There are some positively fascinating aspects of that domain that, believe it or not, citizens of basically every country on earth are familiar with, except for American citizens, because the dollar holds a very specific role in international trade. And it just never comes up in the USA.
Rob Collie (00:01:06): Continuing that James bond-like theme, we talk a little bit about his jujitsu hobby and exactly how long it would take to make me give up. Spoiler alert, not long. In general, he's just a very calm and deliberate person. So as you're listening to this, try to imagine him wearing like a white tux sitting in Monte Carlo, playing baccarat.
Rob Collie (00:01:25): On the technology front, we spend a pretty reasonable amount of time talking about Power Query, pros and cons. We talk a little bit about the idea of using Power Query as like a SQL macro recorder. We again lean on Tom to get him into Power Query. I feel like we got him a little closer this time, but still not quite over the edge. Just top to bottom a fascinating fellow, and I hope you agree. So let's get into it.
Announcer (00:01:52): Ladies and gentlemen, may I have your attention please?
Announcer (00:01:56): 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.
Rob Collie (00:02:20): Welcome to the show. Olivier Travers ... that's a exactly how it's pronounced in the native tongue, yeah? What does it sound like when you say it?
Olivier Travers (00:02:28): [foreign language 00:02:28] Well, that's only with other French speakers, so Travers is good enough.
Rob Collie (00:02:34): I think you should deploy that more frequently than just with French speakers, because the French speakers won't be as impressed with it. It sounds so good.
Rob Collie (00:02:42): Now, it does put a lot of pressure on the other person you're introducing yourself to, right? It puts a lot of pressure on them. But depending upon the situation, maybe that's appropriate. Damn. One more time.
Olivier Travers (00:02:53): [foreign language 00:02:53].
Rob Collie (00:02:53): Oh, so good. End of episode. Hang it up. Let's go.
Olivier Travers (00:03:00): It's hard to top it.
Rob Collie (00:03:02): All right. So you're just the average Frenchman living in Chile?
Olivier Travers (00:03:08): Yeah. I don't know what's average about that, I guess.
Rob Collie (00:03:12): It's such a cliche these days. We're really tired ... oh, another one?
Olivier Travers (00:03:17): There are so many of these, right?
Rob Collie (00:03:19): Is there anything about that origin story? How long have you been there?
Olivier Travers (00:03:22): We've been there for 14 years after spending four years in something in Portugal. So we left France 18 years ago now.
Rob Collie (00:03:32): Wow. World traveler. Do you have Chilean citizenship? Are you still on some sort of long visa or ... ?
Olivier Travers (00:03:40): It's close enough. I have permanent residency. So for most intents and purposes, it's the same. I can vote in any election, but I don't think you can be president or buy land at the border. There are a few things that are off limits when you're not a citizen, but by and large, you it's the same. Legally, in most instances, I would be represented and have the same rights and obligations as a citizen.
Rob Collie (00:04:07): Can't buy land at the border? It's very specific.
Olivier Travers (00:04:10): Well, I don't think they're the only country that does that, is they don't want some citizens from the country at the other side of the border to buy and seed that land to Argentina.
Rob Collie (00:04:22): I see. So it's like a Crimean Peninsula situation.
Olivier Travers (00:04:26): There's not a lot of trust in the area, generally speaking.
Thomas LaRock (00:04:30): You're in Chile, but you're from France. And as far as I know those two countries don't border each other. So I would think they would let you buy land anywhere, but ...
Olivier Travers (00:04:44): Probably they wanted to have something generic. You don't really ... a law made of exceptions is not really good law, so probably they wanted to have a generic case. And also you could say, well, you can be anything, but we don't trust your loyalty since you're not a citizen. So I'm guessing that's the Russian element.
Thomas LaRock (00:05:01): This is weird to me because I don't know that we do this in the US.
Rob Collie (00:05:05): No. We're not really concerned about being invaded in this way. The world's largest military by far tends to not be quite so worried about selling an acre of land.
Thomas LaRock (00:05:16): It's the Canadians. You can have North Dakota, all of it. What do we care? Buy all the land you want.
Olivier Travers (00:05:24): You can always invade Canada any time anyway, so ...
Rob Collie (00:05:28): And Mexican cartels don't really need to own the land on the other side of the border to tunnel under or whatever they're doing, right? And it's just part of NAFTA.
Olivier Travers (00:05:36): The drug trade, you mean?
Rob Collie (00:05:38): Yeah. I just got finished watching Narcos Mexico season two or three.
Olivier Travers (00:05:42): Don't spoil it. It's on my backlog.
Rob Collie (00:05:45): It's a really good show, but it's semi-documentary. NAFTA plays a big role in the most recent Narcos Mexico season. I'm not spoiling anything, but who needs elaborate smuggling when you have an open border for commerce. The only reason you need to dig a tunnel under the border is to try to break someone out of prison or something.
Olivier Travers (00:06:07): There are some edge use cases.
Rob Collie (00:06:09): Yeah. Still some use cases. But again, whether you happen to hold an American ... USA passport, doesn't stop you.
Rob Collie (00:06:15): Okay. So world travelers, you said we, so you're not alone. You have a family.
Olivier Travers (00:06:21): I'm there with my French wife and our two kids. Our daughter was born in France three years before we left. Our son was born in Portugal, then we moved to Chile when he was still a toddler, so they grew up mostly in Chile, which makes them, I don't know what exactly. I don't think they know either. French, but not quite.
Rob Collie (00:06:39): So was it an itch to see the world?
Olivier Travers (00:06:41): Yeah, I'm sure it looks that way. There was a bit of a method to the madness. So the itch, there was an impulse. Let's put it one way. There was a strong urge to leave France. I was spreadsheeting public deficit over 40 years and I was ... you can and put whomever in power and they're still running a deficit year after year for 40 years. So just no way this is sustainable. It's not a good environment for entrepreneurship and to raise kids. So I'm out of here. I love French culture and the food and the landscapes and the architecture. It's great to go on vacation. It might be a good place to retire. For me, it was not the like to raise a family and have an active life.
Olivier Travers (00:07:21): So we wanted to get out of France and originally went to Portugal, where it fit my motto, which was more send, less taxes. So check two boxes in one move. But then most of my business was in the US where I was eight hours ahead of the US for one, and number two, at that time, the Euro was really getting expensive relative to the dollar. So even in a relatively local area such as Portugal, I could only stretch my income so far in terms of actual purchasing power. Also, there was a kind of real estate bubble and we wanted to buy a house. So planets were not aligned for us to stay long term in Portugal, even though we like the place.
Olivier Travers (00:08:02): So it was okay, where do we move that? We get closer to the US. And first we consider moving to the US itself. We looked at Austin, Texas fairly closely. But if you're not a student, if you don't have a big company sponsoring, et cetera, US immigration is not that easy. In my case, I looked at E5 visa where I had to put some money on the line where I was not sure I would be able to stay long term regardless of the investment, so it was. This is a lot of risk they're asking me to take, can I get the benefit of being closer to the US and having a better exchange rate without moving to the US? So we looked at the entire Americas from Canada down to Chile. And Canada was too cold. Mexico already had a drug war.
Rob Collie (00:08:47): Oh, did you hear that? He said that. Already had a drug war. We need to go someplace where we can start a drug war. We don't want to compete.
Olivier Travers (00:08:55): Why compete if you're going to smuggle? Be a leader. You saw right through me.
Rob Collie (00:09:00): Yeah. I just watched three seasons of why moving to Mexico might not have been the move at the time. Yep.
Olivier Travers (00:09:05): And I've done a couple data projects with Mexicans and those were the most lovely, easy to work with, serious people I've I've worked with. And I understand from talking with them that the security situation is really in certain states. It's not the whole country. And Mexico is a large, fairly populated federal country, so it's not all over the place. But you have places where it's pretty much a war zone, not an improvement from Europe.
Olivier Travers (00:09:30): And you could say to some extent that's the case ... with variations, but for lot of Central and South America. We looked at Brazil. There was affinity with that from Portugal, but Brazil has European bureaucracy with third world efficiency. So not a great mix. Even though we love Brazilian culture like sports, my wife practices, [inaudible 00:09:50] I practice Brazilian jujitsu. We have Brazilian friends. They themselves, when they can't move to the US or other places, a lot of the time. The security situation in Brazil, not middle class suburb America. So we ended up with Uruguay and Chile on the short list. And Chile at the time had better internet access than Uruguay, so that sealed the deal for us.
Rob Collie (00:10:14): This is the most deliberate journey of all time. I love it.
Olivier Travers (00:10:17): For something random, right? So yeah, there was a spreadsheet, there was banking, advisory notes and cannabis reports and whatnot. And then talking to people, trying to get a bit of the real deal as well. But then you end up really discovering things when you live there, it's always different from what they tell you. But yeah, we're still there. So far, it's worked out pretty well for us.
Thomas LaRock (00:10:38): So I have a quick question on all of that. Wonderful journey, by the way, and I love the thought process. Now, you were working on ... you said you were tracking deficit and GDP, and you were looking at the 40 year history, and I'm guessing you built a bit of a prediction as to where things would be. So after 18 years, how's that trend line? Has it continued in the direction that you looked at 18 years ago?
Olivier Travers (00:11:03): I'm in the political dynamic where you can have left wing, right wing, and the mix, the combination of both between executive and parliament. It stays the same. It's the same course that adjusts the margin to slightly have less deficit. But there's no structural changes that can really change the course for good.
Olivier Travers (00:11:24): I think the numbers are slightly worse than I thought. I thought some countries would've gotten out of the Euro, not just the European union, like what happened with Brexit. Actually in my forecast, Italy, would've gotten out about eight, 10 years ago and that's not happen yet. And they pull out all the stops with Greece, with the whole whatever it takes policy where we have magic money now.
Olivier Travers (00:11:46): But the fed does do the same in the US. And the Bank of Japan was a pioneer of creating fake money by their trillions of dollars. So now everybody has funny magic money. For me, the way I look at it is like a road runner who doesn't know he's up in the air and should fall until he looks to the ground and then reality catches up. So how long can we run up in the air? I don't know. Funny money works until I guess it doesn't. Right now, we are having inflation like we haven't had in 40 years, so I'm guessing maybe not forever. But for sure the macro forecast was fairly good, fairly accurate, but the political ... because I would say that what Central Banks do right now is more policy than pure economics was no, I didn't see that. I didn't think they would just unleash free money. It's not even truckloads. I mean, there's no vehicle to hold all the streams of dollars at this point.
Rob Collie (00:12:35): Okay. So I had no idea we're going to be talking about this, but I love it. So I have to continue along Tom's line of question. I have to jump in here and oh so gently ask you, do you perceive any irony in escaping from the printing fiat regime of the average Western government, such as France, to the relative safety of South America. You know what's going on over in Argentina, right? And are you familiar with the history of Chile?
Olivier Travers (00:13:08): Pick your poison. Right now we have elections coming in on Sunday, presidential elections, and I'm waiting for the result to come out with trepidation because I'm okay with one result. Sophie is my wife. Where do we go next in the next two, three years? And I've done a bit of business in Argentina, a beautiful country, great food, great people, but you don't want to live there. You don't want to live where there's 40, 50, 100% inflation. It's insane. Chile on the macro perspective was much more stable. It had a primary surplus until a few years ago, but now the demographics are catching up. They've been a bit less responsible. The difficulty for countries like that is they need to invest more, but they don't necessarily have the human capital to know how to invest responsibly and in a way that's going to have an actual multiplicative effect.
Olivier Travers (00:13:54): So where they look at the first world with envy and think they can get there, but how do you bootstrap that when you don't really have that culture yet? Even just personal discipline ... I mean, you, you're not going to become German overnight. So you can say, "I want to copy and pay something from Denmark or Sweden," but the discipline of the day-to-day interaction and sticking to deadlines and saying what you will, say and doing that then, it's a different culture.
Olivier Travers (00:14:21): There is an irony to that. Japan has a messed up economics, but somehow they're sort of spurted for two decades of being flat, but not completely [inaudible 00:14:32] either, who's really performing well these days it's hard to pin down a country that is going to have some sort of imbalance somewhere in their numbers, metrics, policies.
Rob Collie (00:14:41): Two stories that seem very relevant here. The first one, I don't know if it's true. It's anecdotal. It's a legend perhaps. There's a story about this guy in mid-1930s Europe who saw it all coming, saw the conflagration, saw that it was all going to burn again, maybe worse, and knew. And so did his homework, didn't have the internet, did his homework and said, "We've got to get out of here. We got to get out of the way of this." So packed up and moved his family to the other side of the world in a place that would never have had any of these problems, an island in the Pacific called Guadalcanal.
Olivier Travers (00:15:18): Yeah, it's ... or you could ... I'll go to Hawaii. I'll just surf. For five years, I'll just sit another one out. What can happen? Nothing can happen, right? So no, it's really hard.
Rob Collie (00:15:32): Yeah. Didn't see in his research the staring match going on between the United States and Japan over resources and embargoes and things like that.
Olivier Travers (00:15:40): I don't know that it's going to be a better outcome. I will retain some sense of personal agency and free will that I think I was deprived of in my home country. And some other people will say, "Well, France has a great safety net and whatnot." And if that's what they like, then great. I'm more of a libertarian. I'll make my own life and get the benefit and also the drawback of my own decisions and mistakes.
Rob Collie (00:16:04): So the other story is my wife and I went to Israel on business nine years ago and all the prices in Israel on all the menus and everything have the letters N.I.S afterwards, N period, I period, S after it, where the dollar sign would go, where the number of euros, not in front of the number, but after the number.
Rob Collie (00:16:25): And eventually I had to go look this up. It's the new Israeli shekel. Not the old shekel, the new shekel, which was issued in the 80s, the 1980s. So this is ... in many ways we think Israel has much, much, much more in common with the west than most other regions of the world.
Rob Collie (00:16:42): And in the 1980s ... and actually my friends at Microsoft who grew up there told me that there was a moment in time when their parents went down to the bank and paid off their mortgage with a handful of change out of their pocket because of this runaway inflation, that was eventually ... they just redenominated the shekel. The new shekel was worth 10,000 or a hundred thousand of the old ones. And so people started getting paid in the new Israeli shekel. So now they were getting paid essentially like a hundred thousand times as much based on what their previous debts were denominated in, which was the old shekel. And they go down and they pay off their house.
Rob Collie (00:17:20): And Israel is in a tough spot, geopolitically. The whole borders thing ... we can talk about the borders and who's allowed to own land on the border. That gets very contentious pretty quickly. And they had experienced runaway inflation in the 1980s, while we were all alive and paying attention, and they didn't fall off the planet.
Rob Collie (00:17:39): So I'm with you. I've spent long periods of my life, very, very, very concerned about these sorts of things, and at the same time, slowly coming around to also tempering my concerns. It's amazing the things that human society seems to be able to survive, even when it seems like there's just no way that it could, right? A six order of magnitude or five order of magnitude revaluing of your currency to essentially catch up against runaway inflation, you would expect utter collapse. And if you had enemies, they would overrun you at that moment.
Olivier Travers (00:18:13): It's easy to overstate ... economists have forecasted a nine of the past three crises. So the old joke where you can't live and be consumed by worry all the time when you there, and there's balancing elements. That said, last time, you also had high inflation in the West. Not to that extent that the US felt compelled to issue the new dollar, but [inaudible 00:18:35] they could, when a house used to cost five or 10 grand, and now it's $500,000. You have a new dollar by a factor of a thousand right there between the 50s and now.
Rob Collie (00:18:44): I agree. I think the difference here is in the United States is that we don't have to redenominate. We just print it to the new level.
Olivier Travers (00:18:50): It's the privilege of being the reserve currency of the world. And China is trying to balance that and is not catching fire. The Euro is a fraction of the dollar in terms of trade.
Rob Collie (00:18:59): So China actually is catching fire at the moment in the way that you don't typically hope for.
Rob Collie (00:19:04): Yeah. So reserve currency, this is a phrase that I almost never hear come out of an American's mouth.
Olivier Travers (00:19:09): Yeah. Because it's like you're fish swimming in water. You don't know that since all around you they have that benefit.
Rob Collie (00:19:14): Americans don't know what a reserve currency is.
Olivier Travers (00:19:17): The other countries need to have dollars.
Rob Collie (00:19:19): You've got to be a real weirdo. To be an American ... Whenever I'm on an international flight and I'm sitting next to someone who's not an American, I lean over and say, "Hey, do you know what a reserve currency is?" Nine out of 10 times, they do.
Olivier Travers (00:19:30): Well, if you go to Argentina, that's really ... everybody needs to be somewhat of an expert on macro economy, because for them, inflation is like you're making daily spending decisions based on these things. Anywhere in south America, really, everybody will know the exchange rate with the dollar. Even in Chile, the Chile is not dollarized.
Olivier Travers (00:19:48): The dollar is used a way to express sums of money, because Spanish doesn't really have a word for billions, so you have to say a thousand millions and then you have thousands of thousands of millions of pesos. That's a bit ridiculous. So large sums of money, they tend to speak in dollars, but not because that's actually handled in dollars, just as a shortcut. And you know the exchange rate with the dollar, because if you're going to buy with just [inaudible 00:20:10] or what. But you have countries in South America which are dollarized otherwise informally. Like in Argentina, the dollar is really what holds value against the peso, which is a complete joke. So people have to have that awareness in their daily life, even if they're not in the international trade business themselves. It's a different outlook.
Thomas LaRock (00:20:31): I love how this podcast cast has become a discussion on international and monetary policy. And I'm sure the listeners are just going to enjoy this pivot if you will for the next hour.
Thomas LaRock (00:20:43): I wanted to mention this documentary I saw years ago called the Ascent of Money. I don't know if either of you have seen it or heard about it.
Olivier Travers (00:20:51): Yeah. The one in Ferguson. Actually watched that with my son a few months ago.
Thomas LaRock (00:20:54): So it's brilliant, and I would recommend anybody listening that hasn't seen it should probably watch it as well. I think it's worth the time.
Thomas LaRock (00:21:01): But the other thing, more importantly, is for a Frenchman to be criticizing how numbers are done in a different language. You, sir are from the land of four 20s because you don't have a word for 80.
Olivier Travers (00:21:17): That's true. Wait until you hear about 90. That's four 20 10 plus 1. 91. [foreign language 00:21:25] Four 20 11.
Thomas LaRock (00:21:27): I just had to mention that. Wow. Yeah. He's criticizing Spanish.
Olivier Travers (00:21:32): That's the French origins for you.
Rob Collie (00:21:34): We just finished recording a podcast with Lars Schreiber, who's German. We were joking about the way Germans make words is they just take a sentence and take all the spaces out and call it a word, right? It turns out that the French looked over the border and went, "Oh no, uh-uh (negative). Oh, you're not going to run unopposed. We're going to do this with our numbers." Why don't we just count to 91? As we count out loud, that becomes the word for 91. It's 1, 2, 3, 4, 5, 6, 7, 8, 9, 10. I wonder if this has some weird side effects. In France, no one ever wants to price any at 80. They're just like, "No, just make it 79."
Olivier Travers (00:22:10): When you say it in French, you don't think about the root what you're saying. It's just a word. It's just a sound. You're not breaking down, oh, I'm actually saying four multiplied by 20. You're just saying [foreign language 00:22:22] and that's it.
Rob Collie (00:22:22): Tom, did this documentary explain to you what a reserve currency is?
Thomas LaRock (00:22:26): I don't remember that, but I fascinated with how money moves and flows around the world. And when you start to understand even why some countries have wealth for the past few centuries and that wealth could have been elsewhere had a couple of bankers in London, decided to fund France instead of England, it's really amazing when you just see and track the wealth over hundreds of years, and now understanding how and why China is basically such an economic power. And I think the important lesson is knowing that none of this is permanent. The money flows in, and then the money flows out.
Rob Collie (00:23:08): Well, let me finish springing my slow motion trap. Luke, Tom, as American citizens, do you know what a reserve currency is?
Thomas LaRock (00:23:18): Oh, that's what the federal reserve has.
Rob Collie (00:23:22): The world has had a reserve currency for a very, very long time. It wasn't always what it is today.
Thomas LaRock (00:23:28): Are you going to say it's being backed by gold?
Rob Collie (00:23:30): Similar. Basically the reserve currency ... I'm going to explain it, and then Olivier can explain it for real. When it comes to reserve currencies, I'm like a slightly improved American. Basically, it's the currency in which international trade can be safely denominated at all times.
Thomas LaRock (00:23:47): So Bitcoin?
Rob Collie (00:23:48): That's actually the danger. And I've got a cryptocurrency expert themed lined up for soon as football season's over. We're going to go after it. We'll come back to that later. But if you're in France ... let's say you're pre-Euro France, and you want to buy something on the international market. You're a company and you need to get supplies from Brazil. You go to Brazil and say, "Here, I'll pay you so many Francs for your steel or whatever." And they go, "What am I going to do with Francs? Can't you pay me in my local currency?" And you go, "Oh, come on. I don't have any of that." But it turns out if you pay in dollars, everyone's happy.
Olivier Travers (00:24:33): It's just the depth of the currency market so you have liquidity. You have almost no spread. Otherwise, if you want to trade in your own currency and there's ... you're going to be stuck. You're going to ask for a huge spread if there's liquidity all. So at some point, Russia, for instance, was non-convertible. They were trading. I'll sell you weapons for oil or timber or whatnot because the currency itself doesn't have liquidity, ergo, it's hard to access its value, because it could be worth double or half of it next month. You're stuck with it.
Rob Collie (00:25:03): If you accept a large quantity of Francs, it's going to take you a little while to resell those. Now you have to go back to the international market and sell those Francs and trade them in for your own currency to use it again at home, right? First of all, it's a hassle to go sell those Francs, right? It's not a lot of fun. By the time you're done offloading them, Olivier's right. It might be worth much less. There's a huge risk there. It's really, really crazy. So because the dollar is the world's reserve currency, one of the side effects of all of this trade being denominated in dollars is that everyone holds dollars.
Olivier Travers (00:25:38): Our currency, your problem.
Rob Collie (00:25:40): Every country on the world is holding onto large supplies of dollars so that they can go buy things on the international market, but also because they just sold a bunch of things on the international market. Exactly. So when the United States dilutes its currency, the world pays the depreciation. Of course, when the dollar loses its value, American savings accounts also lose value, but the rest of the world loses it too. And so it's much, much, much safer, it's much easier, this tremendous privilege that the United States has that no single American citizen seems to be aware of. So all these things that are trying to attack and displace the dollar as the reserve currency, this is the single greatest threat to current American way of life and current American national security, all of that, is if we'd lost that privilege, we'd actually have to play fair.
Olivier Travers (00:26:33): You'd have to pay for the real price for things. You'd also be exposed to fluctuations that right now everyone else is exposed to. So if the dollar goes up and down, it's everybody else's currency exchange risk. American companies see that when they have to [inaudible 00:26:47] earnings from say the Eurozone and then, well, we grew by 20% in Euro, but Euro gained 10%, so in the end it was just plus 10% in dollars and whatnot, you'll see that in SCC filings ... American companies have such a huge local market, it's not as material to say for the average German company, which is 70% expert-driven, so where they need to control that so much more, because it affects their balance sheet and their real operating income every year. An American company could be, well, it's just extra money. We leave it abroad for tax reasons and whatnot, but we can be happy with 80% of our revenue in the US, such a huge consumer market, huge B2B market. What you need other countries for is gravy.
Rob Collie (00:27:31): Now where things get a little hazy, where you end up in the realm of conspiracy theories that might be true, but also might not be, I've read things like the real reason that Saddam Hussein got taken out the second time around? This is the conspiracy theory. Okay? The real reason is, is that he tried to sell oil in a currency other than dollars. So the story goes, whenever you try to sell oil for something other than dollars, if you don't have nukes, you don't exist anymore.
Olivier Travers (00:28:04): Well, without those extremes, you see how ... for instance, how China is pressuring any country that says we're going to recognize Taiwan or have dealings with Taiwan, is going to be under huge pressure. A mix of incentives, like they will bribe countries, Jamaica or whatnot, Caribbean islands that have sometimes recognized Taiwan as a sort of defacto country. "Oh, well, if you'll get a $2 billion investment if you stop doing that." I think they we're doing that with a small European country this week, play more of the stick as opposed to the carrot policy of, well, bad things are going to start to happen to you if you don't stop your shenanigans, because of China. So I don't know how all, all this relates to data, but it is very data-driven.
Rob Collie (00:28:46): This is the curious mindset that keeps picking at a story to find out what the next level down is. When I got ahold of all of this stuff, probably, I don't know, 15, 20 years ago ... we're going to change gears in a second. But it actually kind of started for me when I was sitting around at Microsoft going, "We have a 40 billion dollar cash hoard ... " this is in the 90s or whatever, right ... as a company. And I asked myself, "Why aren't we splitting that up? There's 15,000 of us here. I don't know. Why don't we get together and just divvy it up," right? And I knew that we weren't going to, but I didn't have the precise reasons why. As I started to pick at that string, slowly over the course of years, it kept leading me into all kinds of new places and everything, and really truly understanding how the financial system works and what it means to be a publicly traded company and all that kind of stuff. And then eventually it led me into all of this stuff.
Rob Collie (00:29:40): And I kind of came around to the conclusion that, like you said, the economists, the famous economists either don't know anything or are deliberately ignorant to keep the music going, and everything is so damn complicated that you end up trying to avoid, try to play it, try to surf it, and you end up in Guadalcanal at the opening of World World II.
Rob Collie (00:30:00): And so I went this long, long, long, global journey in my mind anyway, there and back again, Bilbo Baggins, right? And all I did was just come back to where I am today and go, "Okay, so what are we going to do today with the people that I know directly? How are we going to build a good business?" And all of that. But I had to go on that long, protracted journey.
Rob Collie (00:30:24): And if you met me at various points along that journey, you would've discovered me in varying degrees of radicalism in a way. "Because you don't really know how things work. Let me tell you how things really work." And I think I know how things work now better than I ever did on that previous points on the journey. But I'm kind of more moderate now about it because everyone's on the take, including us.
Olivier Travers (00:30:47): Right. I think there's a bunch of red pills. A single red pill doesn't unlock the whole thing. So you think you understand it and then wait, you're telling me LIBOR rates that underpin any mortgage in the world, or almost, you're telling me that those currency exchange rate that we're talking are rigged? And that's not a conspiracy theory. There's been criminal charges on the biggest markets in the world.
Rob Collie (00:31:12): Brutal fines have been levied that were equivalent to 0.001% of the profit gained by rigging it.
Olivier Travers (00:31:19): Yeah. So it's hard to get the whole encompassing, oh, I know, I understand, top to bottom, and from the macro ... from the universe down to what happens within a cell perspective at each zoom level, which I guess it goes to what we do with BI, which is how can we drill down from show me the main cash flows of a company down to how much money do we make in North Dakota before it was sold to Canada last week on the red widgets?
Rob Collie (00:31:47): It's the North American Crimean Peninsula.
Olivier Travers (00:31:50): Weren't there were plans somewhere in the US history, semi-seriously some president considered invading Canada, or ... ?
Thomas LaRock (00:31:57): I'm not sure we considered it. But yes, plans for the invasion of Canada did exist.
Rob Collie (00:32:04): I think I saw this in the South Park movie, right? Isn't that when this happened?
Thomas LaRock (00:32:09): I don't believe that there was any seriousness about it. I'm sorry. Let's say after the War of 1812. I look at it and to me it's almost like it was busy work. There was somebody in the cabinet at one point and they're like, "Hey, you know what would be great? Give him something to do that we'll keep him busy. Tell him we're looking to invade Canada. Have him drop plans." And he spent two months on it and he came back with everything like, "Oh, that's great. That's great. Now what can we give him to do?" Just to keep him out of the way for a while. That's how I've always thought of it. This wasn't a real project. This was something to just keep somebody else busy.
Olivier Travers (00:32:43): If you read the Federalist Papers, the state of mind of the founders of the US and of 18th century competition with Spain and the potential war with Spain was a big deal. It was right there in the main list of being able to thwart this huge competitor and a potential enemy 250 years later to Spain. Don't you go there once a week to race bulls or have tapas? How's that a foe?
Rob Collie (00:33:08): It turns out that the Americans ended up being the perfect engine for warfare over the last few hundred years, right? Tremendous amount of resources, not easy to invade geographically, and a bit of a chip on it shoulder. We were attempting to make that right turn into BI. But we talk about BI a lot on this show and we're going to come back to it. But I saw it in your Twitter bio. Before you even mentioned it, we were going to go there. We're going to turn this into an episode of the Joe Rogan experience.
Olivier Travers (00:33:34): Three hours?
Rob Collie (00:33:35): We're going to talk about ...
Olivier Travers (00:33:36): Mushrooms?
Rob Collie (00:33:37): Brazilian jujitsu.
Olivier Travers (00:33:38): Oh yeah, don't get me started.
Rob Collie (00:33:40): Let's get you started, because we've never talked about Brazilian jujitsu on this show. When did you first get into that?
Olivier Travers (00:33:45): Well, so I did judo as a kid and teenager a little bit in France. It's like wrestling in Midwest, in the US. It's what you do when you're a kid to bulk me up a little bit, give me confidence, and I did. And then I did other sports, but no wrestling of any sort. And I really needed to get back into shape. And I started getting a bit judo a few years ago. But in the end, when you're middle-aged, something that's really focused on takedowns, and high amplitude at that, is a bit hard on the body, so it's just the recovery.
Olivier Travers (00:34:13): Brazilian jujitsu does takedown, but is mostly focused on [inaudible 00:34:17] fighting. For people not familiar with it, if you look at MMA UFC fights, when they're on the ground, most of what you see is a combination of wrestling and Brazilian jujitsu. So a dominating position and finding some way to choke out or apply a joint lock on you, shoulder, arm lock, leg lock of any sort. But because it's ... you have less explosion than in judo, because you are on the ground for the most part, it's a bit less harsh on the body. You can train more often.
Olivier Travers (00:34:46): And yeah, I fell into that rabbit hole about seven years ago, which for me is the fun way to do something that's very physical and a very serious role in one swoop. So when I was young, I could go and work out and bench press. I've lost any interest. If it's not fun, I'm not going to do it. I will go and have a run every two months. That's just not a way to stay in shape. I'm limited by just my recovery time, but I'll do it as often as I can. So I train three, four times a week. And it's never ending, the techniques and the counters. You think you've got it figured out and then, well, but if you move your arm in that way and your head in that way, then you can get out of that position and reverse it. So it is really fun. It's not for everyone, but I find it extremely fun.
Rob Collie (00:35:33): I've never done it, but I have heard it described over and over again as almost akin to chess in a way, right? There's a physical component to it, obviously. Everything has to be physically expressed. But it is a very cerebral thing. I'm assuming that Chile, you guys use kilograms and not pounds.
Olivier Travers (00:35:49): Yeah. It's metric system.
Rob Collie (00:35:51): How many kilograms do you weigh?
Olivier Travers (00:35:53): I'm about 80. So I think that's 170, 175?
Rob Collie (00:35:56): 178, 175. Okay. So I'm 220. I'm middle-aged, like you. I've never done any organized wrestling or jujitsu. How many seconds does it take you to make me give up?
Olivier Travers (00:36:13): Do I want to work for it or do I want to throw yourself into the submission?
Rob Collie (00:36:17): You're competing in a worldwide competition to submit Rob Collie as fast as possible.
Olivier Travers (00:36:22): Are we starting standing or on the ground?
Rob Collie (00:36:24): Tell me both ways.
Olivier Travers (00:36:26): Short of striking, but you don't want me to take you down and you're going to exert your weight and strength.
Rob Collie (00:36:31): Okay. I'm fighting for my life. You're competing in an Olympic sport where every 10th of a second matters.
Olivier Travers (00:36:37): 40 pounds is nothing to sneeze at. I don't know your overall condition, if you do other sports or whatnot. Do you have any cardio otherwise?
Rob Collie (00:36:44): I don't know if you have it down there, Orange Theory Fitness. I do that, let's say, three to four times a week. It's high intensity interval training.
Olivier Travers (00:36:52): Yeah. So you're in good shape. So it might take me a minute, two minutes just for me to protect myself and not put me in a bad position. If it's on the ground and we're chilling, you want to practice Brazilian jujitsu, you want to go into that game, then 20 seconds, 30 seconds. Because your energy ... you'll try to, for instance, [inaudible 00:37:12] you'll try to throw my legs, et cetera. You'll go in a direction. You'll go way too much in that direction, and I'll take your back ... untrained people do sometimes silly ... they don't do jujitsu. So sometimes it puts you back in more of a real fight mode. If you know jujitsu and then you're 220 ... one of my spar partners is 220 and muscular, 220, 28 years old. That's work. Yeah. He makes me work.
Rob Collie (00:37:36): I don't want to play against him.
Olivier Travers (00:37:40): You don't want to oversell it. Technique doesn't overcompensate everything. Attributes, it's strength and speed and everything through technique. But if you're a complete newbie and short of strikes, which completely changed the game. So if it's MMA, it's different.
Rob Collie (00:37:55): I don't have any training in that either, so it wouldn't matter.
Olivier Travers (00:38:00): I did that a bit, and then I realized I didn't like to get punched in the face. Kneeing other people or elbowing them or punching them is great. But being on the receiving end, I was ... nah, I'm just not going to do that.
Rob Collie (00:38:13): That's actually surprisingly common, even at the highest levels of these combat sports. There are some people who very clearly are not into being hit.
Olivier Travers (00:38:21): And especially if you're hit in the head. You're exposed to other risks like CRT, like concussions, et cetera, and it could affect my work. I mean, I got hurt several times last year. I got a displaced rib a little bit. I couldn't sleep well for a week. I got my eye scratched with someone's nail a couple times, and that's really painful. I got tweaked fingers. You get hurt.
Olivier Travers (00:38:43): But then many of those issues, you can get playing soccer or going on a bike ride with your kids. So it's not like it's the only sport where you can get hurt. You can control to some extent, and also selecting your aspiring partners, gym culture, et cetera. You can control to some extent the level of risk that you're comfortable with. I warned you. Don't get me started.
Thomas LaRock (00:39:04): I wanted to. I think the secret here is because he's going to be expecting you to do ... or you're going to be doing ... you have to keep him off balance by not doing those things. So I would suggest you just do nothing.
Rob Collie (00:39:17): You know what I'm going to do?
Thomas LaRock (00:39:18): Just stand there.
Rob Collie (00:39:19): I'm going to walk into the ring or the octagon with him and I'm going to immediately go into the crane kick position from Karate Kid. Like as far as I can tell, that is an unbeatable technique.
Thomas LaRock (00:39:30): You have to do what's unexpected. You have to do things that he's never seen before, and by doing those things, you'll last longer. You ain't going to win, but you could last longer. And now, this is the important thing, is there's an over-under here and there'll be some wagering. And what I need from you is I need to know if this is a sure thing. So I want you to try all of those weird things and let me handle the money side.
Olivier Travers (00:39:54): The doing nothing is funny actually, because beginners will try that after a couple of weeks of trying too hard, and then they realize, "Everywhere I go is a trap. Oh, there's an arm right there. Oh, he's choking me out. How does that guy have my back when I thought that was winning two seconds ago?" So at some point there's like, "I'm going to try and sit there and see what happens. And then while I'm going to push you, I'll go get mount and you'll get choked in 10 seconds as opposed to 20." So you can scratch that off the list. Doing nothing is not an option.
Thomas LaRock (00:40:26): See, he's just saying that because he knows that's really the answer. He's throwing you off.
Rob Collie (00:40:31): The next three months, I'm going to practice rolling into a perfect ball like an Armadillo and just rolling around. There's nothing you can grab a hold of.
Olivier Travers (00:40:38): My son does that. It was, "Oh, that's my technique. I invented it. I called it The Ball." I'm like, "That's not going to work for long, son."
Rob Collie (00:40:48): Ah, smart kid, smart kid.
Olivier Travers (00:40:52): He was very proud of himself. It's funny, my son is a French guy raised in Chile, and somehow default speaks in English. It's like some kids have been raised by wolves, he's been raised by the internet seemingly. So he's like, "Oh, I call it the ball ... " with his. He has a very good American accent. The ball is not a thing. That's not going to work.
Rob Collie (00:41:11): I just have a real ... it's like a fascination with things that are expressed physically, but that involve technique and end up being just gigantically superior. If I were really rich, a big part of my life would be spent going around just getting absolutely smoked by people in their various fields, right?
Rob Collie (00:41:36): I remember watching Marshall Faulk the football player and thinking, "Okay, how narrow does the hallway have to be for me to be able to actually get two hand touch on Marshall Faulk if he's trying to get past me? There's a point at which the hallway just wide enough that I'm never going to be able to make contact. His job is to get past me. We're in a hallway. He's still going to get past me. I don't have to tackle him. I just have to touch him." Stuff like that, I'm always fascinated by it. If you make it to Indianapolis, you make it to the States anytime soon, Olivier, or even not, let's film this. It'll be two social media stars, Jake Paul fighting ...
Olivier Travers (00:42:13): Tyron Woodley. Yeah.
Rob Collie (00:42:15): Right. We'll just use this. Olivier versus Rob. I'll get 15 minutes of training and then you'll just smoke me.
Olivier Travers (00:42:24): Usually it takes about a year, I would say 12 to 18 months of training, to not have too much trouble with a brand new beginner, except with their linebacker, taking outside professional athletes with a serious body mass. But outside of that, the average newcomer, even if they're lifting, a regular Joe just doesn't stand a chance ... again, outside of striking. Striking has always that one punch, bigger chance you can knock out almost anyone, right? You connect well on the jaw or temple, you could knock out almost anyone. But you'll break your hand in the process.
Rob Collie (00:42:57): I'm as likely to hurt myself throwing punches. So I'm going to ... no striking. Tom, did you have something there?
Thomas LaRock (00:43:05): I'm amused that you think you have a chance.
Rob Collie (00:43:06): Look, I'm betting on me.
Thomas LaRock (00:43:08): The thing is right now, Rob, what's happening is your mind process thinking about this is you're thinking as if you're still 20 years old, but you have the wisdom of something smarter. You don't think you're a middle-aged man in Indianapolis.
Rob Collie (00:43:23): No, that's true. That's totally true. That's totally true. I can't move at the same speed. However, I would pick me today over the 20 year old me in a fight. No doubt about it. I would tear up 20 year old me. I am so much thicker, stronger, bigger. I was still filling out when I was 20.
Thomas LaRock (00:43:39): Now, that's fair. I haven't seen you in the ring. I haven't seen you compete.
Rob Collie (00:43:44): Well, neither have I.
Rob Collie (00:43:48): All right. You're in Portugal before you move to Chile. Most of your clients are in the USA at that point. You need a time zone change above all else, and maintaining the sunshine. I understand these are important KPIs. And that was what year, roughly?
Olivier Travers (00:44:04): 2007.
Rob Collie (00:44:05): Okay. 2007. So let that sink in, dear audience. There was no Power BI. There was no Power Pivot. So what were you doing? I glanced through your blog articles and I was seeing all kinds of things that they're just general technical stuff. How'd you define yourself as a professional? What was your primary marketable skill?
Olivier Travers (00:44:25): Right. So at the time, I had a couple ventures where I was involved in an operational capacity, more than ... I went back and forth with consulting. So I had for a while a gig where I was managing a series of B2B blogs, trade publications, but on the web in newsletters. So I focused on that for eight years or so, something like that.
Olivier Travers (00:44:46): So in that case, I was more the general manager and that was more of a content business. I applied data to running the business. I had my dashboard for sales and finance. And also as a publication ... actually I contacted you when you were about ... what was it, Pivot Stream? 10 years ago. Because where I was trying to get a retail trade publication out the ground, and they were already retail publications, magazines, and I wanted to something that was data-driven. And I saw Power Pivot, but I was ... I need something that is hosted that I can embed in my own website.
Olivier Travers (00:45:18): And then I was trying to have a same store sale tracker going on. I wanted to have something that was across retailers with Walmart and Target saying, "Well, here's the store to store comparison where you don't take growth in terms of numbers of stores into the equation." You're looking at the existing base and you see whether business is growing pretty much organically.
Olivier Travers (00:45:38): So it's a metric that some investment bankers will try to compute. And you see some numbers sometimes, but there was no real dashboard of same store sales. I was trying to build that out, for instance. So that's one area where I was trying when ... now we are talking 2010, 2011, where I wanted to have pivot tables on the web for subscription services or stuff like that. And I don't think the technology was quite there yet at that time. So we ended up having one off spreadsheets, more of an ad hoc, a bit podunk, to be honest, but a handcrafted approach to that.
Rob Collie (00:46:11): Artisanal.
Olivier Travers (00:46:13): Very artisanal. And that was part of the problem. I never made that scale where it became a sustainable long term business. So when ad sales creator, post-2007, 2008, things became faster. We tried to increase content sales subscription and sort of made do for a while. But in the end, that was not enough. So I went back to consulting, leveraging my operational skills, and I would say my breadth of experience, as I think now it's apparent from the various countries, various businesses, industries, languages.
Olivier Travers (00:46:44): I'm a specialist in generalities. I'm not the best at anything really, except maybe being good at a wide variety of things. So what I'm good at, I would say, is from that background is diving in something where I'll get to understand maybe the key drivers for your business in your industry, in your country, whatever that is, fairly quickly. So the business and the acumen, I would describe that, generally speaking, drives and then the technology follows. So yeah, Power BI is a good tool. We'll use Power BI, but we can use other tools as well. I don't to think of myself as a one tool guy. I think that's not good job security. And also then you become the hammer that is only seeing the nails everywhere. I don't think that's good service for clients.
Rob Collie (00:47:24): Go back to the jujitsu metaphor. You need all the tools so that you can show those beginners that there's no escape, no matter what they do.
Olivier Travers (00:47:32): In this case, it's not as antagonistic with our clients. You try to make that more of a corporation, a partnership, but yeah.
Rob Collie (00:47:39): The opponent is the problem. The problem tries to wiggle out this way. "No, no. Uh-uh (negative). Submission. Arm bar". So do you remember what it was, or how you even discovered Power Pivot as early as you did? Because you're right. We were in touch back in my Pivot Stream days, which ended in 2013. So that puts you in the first 0.000001% awareness worldwide of the existence of Dax, the existence of the verte pack engine, [inaudible 00:48:10] whatever was behind Power Pivot. You'll never randomly bump into someone in your life that discovered it before you. You might be at a conference or something. In the random world, you will always be number one to have discovered it.
Olivier Travers (00:48:24): So a couple ways to look at it. First off, I used Excel from when I was in business school. I had a side gig with a local computing PC assembly company. And the CEO of that company hired me on the side to create a monthly price striker, a dashboard of all the competitions. So I was spending my Saturdays reading the monthly PC shopper magazines and typing in the data and charting specs and this hard drive and that monitor and that much ram. Then here's the graph of everywhere, and you're here. My first dashboard was 30 years ago. That was Excel. And I also did a bit of sales for that guy. At the time, I sold 17 or 18 286 PCs to my business school. So I went to the IT of my business school like, "I want to sell you computers." They're like, "Okay, what's your quote?"
Olivier Travers (00:49:12): Then I worked at Microsoft in the mid-90s and that was obviously a tool where there was a lot of use. There was a data warehouse for sales numbers called MS Sales, very creatively, as you would expect from Microsoft. And I would go and create those awesome pivot tables. And I would macros refreshing them from that data warehouse in I think '97. And I would show that to my peers and colleagues and they went, "How do you do that? Where is that even coming from?" I was, "Let me show you."
Olivier Travers (00:49:39): I was in charge of larger country sellers, the companies selling to the select at the time, licensing mass deals. And I was able to go and meet the CEOs and CFOs of those resellers, partners of Microsoft saying, "Well, here's exactly how much you're selling," in Francs at the time, and this product line against that type of license in that region. So we had already a fairly good breakdown. So it was all pivot tables. So I wanted ... for the publication business I was telling you about, I wanted to have pivots, but for the web. And before that, maybe there was something for SharePoint I don't quite remember.
Rob Collie (00:50:13): Yeah. Excel Services.
Olivier Travers (00:50:15): Yeah. There was a startup ... I don't remember the name, that that's something that ... like the first pivot technology I saw, maybe a year or two, and they were purchased by whomever and they sort of killed it. I was on the lookout from the 90S for a pivot on the web. And then it took forever to materialize into something that is now credible without blowing in your face. Took 10, 15 years to really happen, I think.
Olivier Travers (00:50:40): Power BI now is mature. I looked at Power BI desktop designer. It was not [inaudible 00:50:45] was not called Power BI Desktop, but in 2015 ... I was, "This is complete garbage." I was playing with Tableau at the time. I blogged about it, saying, "Microsoft has the work cut out for them," because it was buggy. The visuals were fairly crappy. But you could see the intent was better than the whole Silverlight Power View fiasco, which was ... I could have told you that was a technical dead end. What committee came up with that architecture?
Rob Collie (00:51:08): Oh, I know.
Olivier Travers (00:51:11): You know where the [crosstalk 00:51:12] is? I buried it.
Rob Collie (00:51:12): Yeah, but the people that were involved in that, they also have very, very selectively short memories. Not my department, mm-mm (negative).
Thomas LaRock (00:51:20): Are you saying somebody kicked Mr. Silverlight?
Rob Collie (00:51:23): I saw a tweet today ... Olivier, we'll give you a little background here. This is a National Football League coach, urban Meyer. The funny thing about NFL coaches' contracts is that they're almost always guaranteed. Five years, 7 million dollars a year, and if I fire you, you get the rest of the money. It's just an incredibly perverse incentive.
Olivier Travers (00:51:46): Well, I was about to say, what are my incentives?
Rob Collie (00:51:46): Yeah. It just turns out that you rely on the demographic being so competitive that they don't want to go out on their shield. That's not how they want to go. And I think Urban decided very early in his tenure with the Jaguars this year that, nah, I'm just going to keep being awful, like just really awful until they have to fire me. And I saw a tweet today that said ... it didn't even mention who it was about. You didn't have to. Everyone knew. It said, "We just witnessed a masterclass in stealing tens of millions of dollars in plain sight."
Olivier Travers (00:52:19): Yeah. Silverlight was really that architectural ... let's compete with the product that's going to be phase out by the entire market in two years. Yes. I'm saying that a bit with hindsight regarding the fate of Flash. But really Web 2.0 was already well established. Dynamic HTML was already a thing. What are you doing? What is this Runtime? The .net ... I mean, it was like architecture from the mid-2000s, 2005, that they tried to pursue eight years later that it made no sense. Why not ActiveX while we were at it, right?
Olivier Travers (00:53:29): Yeah, I remember seeing the first cross-browser drag and drop in some financial web app 15, 16 years ago. And I was, "Huh." I liked to see it because I was doing some web coding and I had developers working with me and that was ... and now you're thinking drag and drop and all those affordances that you have on the desktop. We expect them by and large to have them in the web browser, right? But go back 10, 15 years ago, that was edge demos, not something that you would expect to really work, the whole cross-browser thing. As long as Internet Explorer was around, that was a big deal. Now everything is Chromium, that's, to a large extent, not an issue anymore.
Rob Collie (00:54:05): We just had another wordagami, affordances. I remember when I was first introduced to that word five years in my Microsoft career. And it was new to all of us. Oh, a new way to say capability or type of UI. It was like an all purpose word that sounded cool. I think it had about an 18 month lifespan at Microsoft. We burned through it pretty quickly. But it is a good word.
Olivier Travers (00:54:25): I think it's a bit more specific than functionality because it's about the way that it responds. So for me, it's almost something that's tactile, click and I see that it feels like the button can be pressed. If you look at the affordances in a game, any game where you interact with 3D objects, like things magnet and click and there's noise and there's boom, and you see that things connect or not.
Olivier Travers (00:54:47): Why do we have business applications that feel like they're mainframe screens from the 70s? Why don't you apply what's done in the gaming world where it's much more like tactile almost? It's virtual, but you feel like it's boom, it's clicking. You feel like a physical emulation of a physical mechanism on screen and reinforced with haptic reactions on your phone, made data entry possible. And if you don't have the haptic feedback, it doesn't quite work to type on a smartphone. Right? That's what I think the breakthrough with the iPhone.
Olivier Travers (00:55:19): Now, apply that to those dead UIs in business software. Have you looked at what's been going on in gaming for at least 20 years? I feel everything in terms of what we accept as adequate performance is laughable in the business world. You have a whole generation of managers that go 120 FPS or 240, just because oh, 60 is not quite ... we're coming from a world where TV is worth 24 Hertz, 24 images per second. Now you're pushing 120, 240, and you're locking that frequency. And then we're there waiting for five seconds for a screen to refresh. And you have a blank screen in between. What is this? I want to send every enterprise software developer working for a gaming studio and learn what's real acceptable frontend performance and come back and lock us out of the ... like I feel we have something that's shinier and better than it used to, but only marginally. There's a leap when you go into a game, and boom.
Olivier Travers (00:56:15): I have three screens. The main one is a 49 ultra-wide. And I have two 27 on top, so I can feel like a trader, even though I don't trade anything. But play a more ... say a driving game on a screen like that, it's amazing. And then you go back to Teams and you're like, okay, how do you get that level of streaming? I would say bottlenecks. Don't get me started on downloading data from SharePoint. Say you're, oh, five megabits per second. Wow. Welcome to, I guess, 2002 broadband.
Rob Collie (00:56:45): Oh, well SharePoint, it's running a lot of complex garbage SQL queries to ... let's not go there. So we should send like the CarMax that made Doom ... we should send them to be technical fellows on the intelligence platform at Microsoft, bring some of that quality and response time and production value discipline.
Olivier Travers (00:57:03): Yeah. I think it's the expectations. We've just grown accustomed to, well, if you wait two seconds and nothing happens, that's good. That's fast. If you play a competitive P2P, player against player game online ... I think you played World of Warcraft. I used to be sort of semi-hardcore casual 15 years ago.
Rob Collie (00:57:23): Yeah. Me too. I haven't played World of Warcraft in about, I don't know, five hours. I was on earlier. Just doing some auction house work, folks, I wasn't ...
Olivier Travers (00:57:31): Yeah. I've done that. I've done the arbitrage.
Rob Collie (00:57:35): I know you have.
Olivier Travers (00:57:36): You know I have, right? And so ... and I have that transparency.
Rob Collie (00:57:43): Have you done any cross-server arbitrage?
Olivier Travers (00:57:45): I've done cross-faction where you swam for 20 minutes to get to the neutral auction bay in Booty Bay or whatever, and then you could buy, but you had to make sure that people were not scamming your trades because you had to go on the open auction. I made 67 gold in two months. That was the game.
Rob Collie (00:58:02): Just making the gold, right? In fact, I'm trying to give the gold that I'm making away to my friends and they're like, "I don't want to have to pay it back." I'm like, "You don't understand, this is the game."
Olivier Travers (00:58:11): I'm hitting the limit. I'm hitting the account limit. So anyway in those games, a hundred milliseconds, 200 milliseconds ... 200 milliseconds is about human eye to finger reaction time, more or less. It increases a bit with age. But you know how that's the difference? We're talking a fifth of a second. You're going to perceive it. But then we accept we're saying three seconds is fast. That's an order of magnitude slower than it should be for people not to start having their ... your brain is already starting to wonder and open other tabs. Power Query, I love Power Query as a tool, but it's impossible to be in zone with Power Query. You cannot be productive in the sense of I'm in a task and that's all I do and boom, boom, boom, boom, boom. For me, I can't just be there waiting for the next step to refresh because I changed an earlier step and all the evaluation to go on for 30 seconds ... it just doesn't work for me.
Rob Collie (00:59:01): I feel your pain. You should see our Power BI report that measures are advertising effectiveness, connects to an Azure data warehouse. And the way that it's structured is the way that Google, for instance gives us data. It's just tremendous overkill. So it ends up with you kick off a Power Query, anything that triggers the refresh preview, go get a cup of coffee.
Olivier Travers (00:59:22): Yeah, exactly. So you have to start computing, is it worth it for me to create a view that is just a thousand rows that I'm going to develop all the ETL from? Can I build that to the client? Is it fair? Is it worth it? And you're like, "No, I won't do it," unless that's really marque project that you're going to keep making and adding to. But oftentimes it's not worth it and you're just there frustrated that it's not faster.
Rob Collie (00:59:45): And of course, with those sub views, those sample data sets that you would use to develop against, part of the problem is that you still need to validate that your answers are correct. And you can't validate artificial numbers in any way.
Olivier Travers (00:59:57): Yeah. Because it kills your profiling. So then if you are trying to find exceptions, you can't do that with just a thousand random views either. So yeah, that's an issue.
Rob Collie (01:00:06): Nope. Nope, nope, Nope. Tom's going to fix all this for us as he learns Power Query over the holidays.
Thomas LaRock (01:00:12): Oh, is that what I'm doing?
Rob Collie (01:00:13): Oh yeah. You're going to splice that together with SQL, and you're going to develop a school of operating with Power Query and SQL that will be called The LaRock School. You know how Matt Allington named my way of laying out tables in the Schema view of Power BI the Collie method? If I can have something named after me like that for just arranging things in two rows, I think you can actually contribute something of far greater value to the world and we will make sure that the name sticks.
Thomas LaRock (01:00:44): Yeah. I'm good.
Rob Collie (01:00:47): Such disappointment.
Thomas LaRock (01:00:48): Well, it's Raw Data by P3, not by LaRock. So ...
Rob Collie (01:00:53): Listen, this is a point in time, Tom. This is the pre-LaRockian method, the pre-LaRock School era.
Thomas LaRock (01:01:00): True, because it could be just Power Query all the time.
Rob Collie (01:01:03): Well, Power Query and SQL, the two of them together.
Thomas LaRock (01:01:06): Together.
Olivier Travers (01:01:07): Oh, I see a query folding winter of love for you.
Rob Collie (01:01:12): I agree. Until Tom is throwing query folding around like the word affordance ...
Thomas LaRock (01:01:17): There's so much out there yet for me to learn.
Rob Collie (01:01:22): Ich bin ein query folder.
Thomas LaRock (01:01:24): There's so much and it can be hard for me to have focus on tackling any one thing.
Rob Collie (01:01:33): I totally get it. That's why I'm here for you.
Thomas LaRock (01:01:35): So is Power Query a thing for me right now? You know what? I don't know
Rob Collie (01:01:41): Top of the stack.
Thomas LaRock (01:01:41): I don't know.
Rob Collie (01:01:41): You owe it to yourself, to the world.
Olivier Travers (01:01:43): Here's a third thing that I did with query folding that I'm told, "Oh no, you should not do that," but I think the rationale for not doing that was a bit being a purist, is you can look at the SQL generated by our query underneath. You get the query. So I thought, well, do the ETL there and it's a defacto SQL generator, and then lift that query and create a view with that. Do you think that's really bad because it's poor SQL that it writes? Or do you think that's brilliant?
Thomas LaRock (01:02:10): I won't say poor because I'm not sure it's that good.
Rob Collie (01:02:12): Here's your in. You're a DBA by trade. You like to complain about things. You like to criticize, right? That's your people.
Thomas LaRock (01:02:18): Even if I go into Power Query and I, like Olivier said, we can look at the SQL that gets generated, and I'm certain it's not optimal. But I'm also certain I have no idea if it would be even possible to have a machine generate optimal code. And on top of that, what I'll say is going forward, the engines are going to be smarter and smarter and they're going to be able to take that code at some point, make a recommendation.
Thomas LaRock (01:02:47): For example, I happen to work for a company who can already do this for you, right? We can look at what you're doing and we can make suggestions as to how to optimize certain queries. So that already exists. I imagine that'll get baked into the engines at some point and things get better for everyone.
Thomas LaRock (01:03:04): But if I could do it in a way that would be helpful, but I don't know enough about how it actually gets generated. So I wouldn't know enough to say, "Hey Microsoft, instead of doing it this way, you should probably do it this way." But yeah, if I could help, if my involvement with Power Query and SQL could actually help move things in a forward and positive direction, I'd love to be able to do that. I just don't know that I can justify the time on that versus all the other shiny things I want to be doing.
Olivier Travers (01:03:32): Also, you could say it's a suboptimal optimization where the computing has been let's just throw more hardware at the problem, generally speaking, the last however many decades where it's hard engineering scarce power to go and do those marginal optimizations because you have to go deeply into the code as opposed to just buy Power BI Premium and you go from say for PPU, for 10 bucks to 20 bucks. Unless you have a massive number of users, it's trivial the amount of extra money to unlock, applying much more computing resources to the same issue. And then apply your human talent to things that you have no readily press a button and apply more hardware power to solve the issue, I guess.
Olivier Travers (01:04:13): So I think in many cases, there's the, well it's okay-ish and good enough when we amp up a bit the resources against it, as opposed to, well, now it's a bit of an 80-20. To push for the 20% in extra performance, we have to double the amount of engineering talent. And that talent might be deployed into brand new problems, for instance, as opposed to chasing marginal utility.
Thomas LaRock (01:04:38): Here's the thing though. If they gave you information about the statement or if you could get information from the statement the first time it was run, let's say out of the Query Store ... so here's the thing. You go into Power Query, you generate something, we go and we look at the SQL and I tell you it really isn't the most optimal thing in the world. But the generation of that code doesn't know if it's being run once by one user one time or a million times by a million users. It has no idea about scalability. It's like, "Here, here's the answer that you were looking for." And now a human is going to have to figure out if we need to make that thing scalable.
Thomas LaRock (01:05:13): So if you were just given information at the bare bones level that said, "By the way, to let you know, this was 10 gabillion, logical reads, logical IO, things like that. It's fine if you're just going to run this once, but if two people need to run it, you should probably look into making this code better, because chances are that's not optimal." And if you just had that information, now you can talk about the human part and say, "Hey human, go focus on the ones that we can see are being run multiple times and consuming the most resources, because maybe instead of throwing hardware at that, maybe we should spend your time and salary on fixing those queries first." That at least would be a step in the right direction.
Olivier Travers (01:05:52): That feeds back into how much human talent are we wasting. In the case of Power BI, that would be an issue for [inaudible 01:05:58] query or for the author waiting, getting their coffee while Power Query is refreshing the preview. But otherwise, if you're doing imports on the scheduled ... it doesn't matter. Just schedule it where by the time your users need to have the fresh data, it's in there in memory and that's it.
Olivier Travers (01:06:12): So I think that was more ... the impedance was do your authoring once and maybe you wait a little bit, but the UI sort of makes up for having to write manual queries. So [inaudible 01:06:22] net is still productive ETL even though you feel frustrated. And then it's important, and now you can schedule however many times. As long as your query doesn't run out, by the time you need to refresh again, you're good. I guess that's how they saw sort of making those trade offs.
Rob Collie (01:06:37): I'm just sitting here feeling very, very satisfied with myself, almost like this puppet master, as I listen to Tom suddenly start to enthusiastically spout off about some of the implications here, some of the directions it could go. I mean, this definitely isn't the sound of Tom getting engaged with the Power Query SQL server interface, now is it? Nope. Definitely not. It's not what that sounds like. Now, by taking the victory lap here, I'm probably sabotaging. If I was truly a puppet master, I would sit back and say nothing, just let Tom get started. But he's in, folks. I can tell.
Thomas LaRock (01:07:12): Yeah, I'm going to download it right now. Power BI for Mac and I'll have it up and running here in no time.
Olivier Travers (01:07:18): I'm losing track of the reverse, reverse, reverse psychology there, so I'm not sure who's getting whom there.
Rob Collie (01:07:23): And we're losing respect for Tom and his Mac. These things happen. But you started to ask that question about query folding a few minutes ago. Just look up Power Query query folding. And by the time you're done, Tom, by the time you're done reading that ... in fact you should just stop right now. You shouldn't read that because you're going to get hooked.
Thomas LaRock (01:07:41): Do this. Don't do that. I feel like George Harrison. Just tell me what you want me to do and I'll do it or not do it.
Rob Collie (01:07:47): So you started to ask that question, Olivier, about query folding, and I'm like, "Oh god, here it comes. I am not going to have any opinion at all on this. I'm not even going to understand the question. I'm going to look really silly." But then you got to the end of the question and I'm like, "Oh, I do get this." So using Power Query essentially like as a macro recorder.
Olivier Travers (01:08:05): And that's what it is.
Rob Collie (01:08:06): To then write some SQL for you. Because it's one thing if it's a macro recorder that writes M for you. But now you've found macro recorder prime. It's even producing SQL statements that you can use. Okay. And then you can take those SQL statements, however clumsy they might be because they're machine-generated, like they always are, and move that SQL statement to the server, closer to the source, and have it run that and therefore simplifying your Power Query script. You connect to the view that you create with that exported SQL essentially. And hopefully it's a sip of coffee from now on when you make a change to your Power Query and not going and brewing one, drinking it, thinking about another one, and then going back to see if it's done.
Olivier Travers (01:08:51): So when I did that, I thought that was brilliant. And I felt a little dirty, which makes it feel even better. But then for some reason, my data flow would not refresh. I would not get the latest data, and I was ... nobody's ... that I see is talking about this online, that kind of an edge case where I'm going to waste more cycles trying to troubleshoot this than moving on with my life. So I thought I should bring that up.
Rob Collie (01:09:13): Tom's got you covered. He's coming for it. Until you said that part where the data flow didn't work anymore, I was prepared to give this the Rob Collie Two Thumbs Up Seal of Approval, right? Number one, it's clever. Number two, it works, right? This is where at least one of the thumbs gets retracted. Not quite a two thumbs up. But the third one is, is that it's dirty enough that going to upset people, who are the kind of people that when they get upset, that confirms you're on the right track.
Olivier Travers (01:09:43): The projects I do, I either do departmental projects within large companies or enterprise projects within small or medium companies, right? I don't do huge enterprise projects with enterprise companies. So there's the value of being expedient. And usually I don't deal with IT. I deal with a CEO or a business owner or a CFO. I'm going to lose them fairly fast. If I go into the nitty gritty of my [inaudible 01:10:09]. I try to explain them that it's not a black box to establish trust, but at some point that's established and I can skip on some of the technical details and just focus on, yes, we have this challenge. We're going to solve it this way or that way. And we'll get the clean data that we can chart it. So it's not the most elegant. It's not something that's going to win architectural design awards, but it works and lets me stay within budget constraints where I'm still going to focus enough time on the end user result and value and adoption and everything else, as opposed to I spend 80% of the time cleaning up the damn data.
Rob Collie (01:10:43): And even the SQL that's generated in that situation, at least that's potential starting point for someone who's really good at SQL to go look at this and go, "Okay, I can write this functionally equivalent in a better way, but in a way it's the clearest possible specification," because you see the code and you can run it and you can test for the equivalent of result.
Olivier Travers (01:11:03): If the result is right. I did the transformations I wanted. So it is the right query in terms of what is spitting out. Are the inners clean and performing? Well, it is or it is not an issue. Maybe your SQL server or VM is slipping 95% of the time anyway, so we're wasting resources in some way by not using them, you could argue. Those are trade-offs. Yeah. It's generated art. It's logo by computer. Maybe that's good enough for 80% of or whatever. Or maybe it's good enough for prototyping, and then by the time you know that it's actually the query you need for the data that your CFO wants to see, then have it written properly, and you'll probably drop a Power Query by then and do it with the ADF and have a proper data warehouse and whatnot, and sell them on CNAPS and upsell on mid six figure consulting.
Rob Collie (01:11:51): We were sent to put an end to that as sort of our credo. So where are your clients largely geographically today? Are they USA, given time zone? Have you put down a lot of local business route in Chile?
Olivier Travers (01:12:06): Yeah. The local stuff I do once in a while. I used to push for more to get it when it was easy to get face to face with people just because I was craving getting out of the office. It's not been possible for the last couple of years, so I've not pursued that nearly as aggressively. Also, pricing is a bit more of an issue relative to first world countries. So the US is large part. I have a project right now in Canada. I've worked as far as Australia, where I have a slice of two hours to get them on the phone early in the morning. So you have to work a bit to make it work. But if you look at the curve, then it's mostly the US, and then a bit of this, a bit of that.
Rob Collie (01:12:42): Are there user groups for Power BI and things like that in Chile? What's the local talent level?
Olivier Travers (01:12:47): There's a user group. Again, back then, when it was easy to connect physically, I went and spoke a couple times, demoed maps and also how to connect to APIs. Had a lot of fun doing that. And then COVID happened. So I already do enough remote. I don't want to just be stuck to remote on the user group, so I have not been that active. And you see that the same thing in Brazil, they're getting there, but they're still a bit more of a state traditional IT approach to things, generally speaking.
Olivier Travers (01:13:13): Also just the business culture is risk-averse and very hierarchical, so it's harder for people who are not in a clear chain of command to go and do something. So it's a bit more of a top-down, here's the exact requirement, you're trying to do my job for me. That's just not going to work out. Let me worry about that. What is it that you want to know to drive your business? Don't tell me that I need a gauge or ... you can give me preferences. I'm not going to be an absolutist again, but give me guidance. Don't give me directives. Otherwise hire a junior staff. Don't hire a guy who's been working for almost 30 decades.
Rob Collie (01:13:45): I love your avoidance of the absolute. Because as we know, only a Sith deals in absolutes.
Olivier Travers (01:13:51): I never speak in absolutes.
Rob Collie (01:13:54): Well, Olivier, this has been great. I'm glad we got to catch up like this. Thank you so much for making the time to be here.
Olivier Travers (01:13:59): Thanks for having me. This was really fun.
Announcer (01:14:01): 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.