 Thank you. Singapore, we need to talk. I really need to talk about a very important issue. No one agrees on what to do. No one knows where to put their commas. So I mean, I love SQL. I hope you love SQL. And it's one of the best languages invented in 1978. No one has found anything better to deal with data. But there's a problem here because, well, we all started learning SQL with something like this. Pretty easy. You get one column of data. You can add another column of data. You can add a third column of data. But this doesn't work because I left an extra comma. Any modern language will allow me to put an extra comma there. But SQL doesn't. So now people are fighting. Should we put our commas at the start? Should we put our commas at the end? Who likes their commas at the end? Commas at the end? Commas at the beginning? You monsters. That's ugly. Well, I'm one of you. I like my commas at the beginning. I think they are more productive. But it's not easy to convince people. So trailing commas versus leading commas was better. I happened to work at Google, where we like making decisions based on data. And we have this Google Cloud, where we share our best tools with everyone else. And one of these products is BigQuery. That allows me to analyze a lot of data, share it within the company with more people, et cetera. And it allows us to collaborate between us. With the same data, we can look at things. And for example, in BigQuery, I have all this data from GitHub. And I can count all of the events in GitHub in my table with a query like this. Turns out we had 360 million events this year. Cool. Then I can go deeper and count the number of users, how many distinct users. And I can join it with other table to see what country they're coming from. And I can find out that the country with the most users is the United States, followed by China. But that makes, that's a little unfair as a ranking because, of course, it's the biggest countries. So let me bring in another table and see it per capita. With another join, I can find that the countries with the largest concentration of programmers is Iceland, Sweden, Norway, cold countries. But I really, the cold country is the one with the most programmers. If I bring all the weather from the whole world and join this with these tables, I can find out that, yes, there is a correlation. These are the cold countries that have the highest concentration of programmers. The hotter countries have a lower concentration of programmers. But that, on the top right, that's Singapore. You have the highest concentration of Github programmers within the hot countries. So that's pretty cool, but this doesn't solve my question. I need to know what's better, come as at the beginning or at the end. And I also have Github code. I can see gigabytes of SQL code and find out what do we have more, more lines with a SQL at the beginning or at the end. And the result is this. There are way more commas at the end than at the beginning. So I lost. Or not. Because the question is, what projects are more successful? The ones with the most stars. You have to define success. Stars, the number of contributors, the activity. And it turns out, in all of these dimensions, I run this query and I found out that these numbers, and it turns out that projects that allow people to put leading commas are double as successful as the other projects. You can check this out. So I win. Well, in Google, we have a lot of other products that collaborate within each itself. I'm going to be presenting this at Strata. I have my colleagues here that will be presenting there too. And my final message here is that maybe we will not stop all fights. Some people started asking me about the statistical distribution, why did you average, causation, correlation, all of that stuff. And some people you will never convince, no matter how much data you show them. We call them industry analysts. But at least we should be able to embrace data within our fights. And that's my five minutes. Thank you. Thank you so much very best. So, OK, let me.