Episode 179: SQL Trail Recap & Buying Power BI

Episode 179: SQL Trail Recap & Buying Power BI

Episode 179: SQL Trail Recap & Buying Power BI 560 420 Carlos L Chacon

SQL Trail continues to be an experiment in the making and year three proved no different. Eugene joined us this year and we discuss some of the changes, how the vision has changed for the conference and what the future will hold for our event. While there is plenty to be excited about with our event, we couldn’t help ourselves and get into a discussion about Power BI and purchasing options.

Episode Quotes

“I will say that I was pleasantly surprised, which sounds like a back-handed compliment, but it isn’t, I swear.”

“We’ll put it together, and of course those who are interested that want to come and see, we’d still love to have you, but we just recognize that it seems to cater more for the local folks.”

“I think nobody in their right mind is going to be consuming Power BI reports through Power BI Desktop. And that’s not a criticism against people, but just the fact that it’s not really designed for that.”

“You’re going to get Power BI Pro and pay $10 per user per month, and that gives you the vast majority of features and functionality with Power BI. That’s going to be the most common solution in a majority of cases.

Listen to Learn

00:38     Intro to the team & topic
02:28     Management Studio has a Schema Change Report
04:38     Why our SQL Trail conference was started
07:32     Building “conversation triumphs content” into the culture
11:04     Social media idea and why the venue is great
13:12     Scheduling is a challenge, especially in conference season
14:27     Thoughts on some of the favorite sessions
17:07     A little bit of information on buying Power BI
19:05     Licensing Power BI can be a little confusing
20:27     Do you need to be using Office 365?
21:48     All the frustrating options and you still pay more for report creation
24:49     Closing Thoughts

*Untranscribed Introduction *

 Carlos:            Compañeros, welcome to another edition of the SQL Data Partners Podcast. My name is Carlos L Chacon and I am joined today by Eugene Meidinger.

Eugene:           Hi, everybody.

Carlos:             And we’ll excuse Kevin for today’s episode. He’s over in some, I don’t know, country doing some SQL Server-y things.

Eugene:           That doesn’t narrow it down at all for him, honestly.

Carlos:             No, that’s right. He’s become a world traveler. So, he just finished up in New York City, is now actually over in Great Britain, and then he’ll head to Australia.

Eugene:           Oh, my goodness.

Carlos:             Yeah, he’s a busy guy. And what is it, while the cat’s away, the mice will play. So we have an episode today, we’re going to talk about a couple of different things. We batted around a couple of ideas, and then we’ll actually see what we get to today. So today’s episode’s kind of a mixed bag. Let’s see, I should have some compañero shout-outs and I’m not ready to do that, unfortunately, so we’re going to have to skip those today. But we’ve had some interesting time and I appreciate your patience, compañero. Eugene was just chastising me. I mean reminding me that we’d taken this break from the podcast and trying to get back up to form has been a little challenging. We’ve had some stops and go’s here, so we appreciate you hanging in there with us. So Eugene and I just got together. We’ve actually just come back from– well, he’s just come back. I live here in Richmond, so maybe it was a little easier for me, from SQL Trail. So we just had that, and I think we’re going to talk a little bit about that. Does that sound reasonable? And I think we might get into some Power BI. So those are all some things that are on our minds.

Eugene:           Milk and orange juice. Two great tastes that taste great together.

Carlos:             Well, they manage to make sorbet work, right? Isn’t that kind of what the–

Eugene:           Okay, I stand corrected. Fair enough.

Carlos:             You just have to know how to blend it just right.

Eugene:           Sure, sure.

Carlos:             Okay, so again, compañeros, knuckle-dragging Neanderthal, and I don’t know how I missed it, necessarily, but one of the things that comes up from time to time, I’m not sure if you’ve ever experienced this was, “hey, who dropped this object?” and ultimately, you’re looking for schema changes. Now, that information is in the default trace, but I have found it sometimes difficult to get that information out. Well, I was just pointed, the other day, to in Management Studio, and I’ve been trying to figure out what version it was introduced in, but in the standard reports, they actually have a schema change report.

Eugene:           Oh, that’s interesting.

Carlos:             And the nice thing about that report is if your trace file loops over, and you have a couple of copies laying around, it will take care of transversing those copies for you. So it won’t go back all the way in time, but inevitably, which actually happened to me, right, I got a request, “hey, can you tell me who dropped this stored procedure?” And my initial thought was, “no, because I don’t have auditing turned on.”

Eugene:           Yeah. Trying to get retroactive information is a pain. I did auditing work for one customer, and we got a bunch of stuff set up and it was brutal, because I was doing Extended Events work in SQL Server 2008 R2, so we had no GUI or anything.  But they asked me, like, “hey, is there any way to get information from before we did all of this?” And yeah, it’s like, “well, there’s some trace if it wasn’t blown out by a bunch of events and stuff, and that’s probably it.

Carlos:             Right, and so it is interesting that generally when we think about system trace, it’s just that last file, because it’s the one that these things run or it’s the one that’s currently running, but you can have other copies there. But then I don’t see too many blogs pointing us to those older files, and again, so I think just like your log collection, it can vary depending on your settings as to how many it’s going to keep around, but I found that to be very cool and just a handy little tool. Obviously, it’s a report, so then you can give it to anybody and you can export to Excel and I can get the answer. And particularly in this case, the default trace didn’t have the detail I was looking for. It had already gone to another file. And so yeah, that was something that I learned that I thought that was handy and wanted to talk about it. And speaking of learning, we wanted to do a quick recap of SQL Trail. And I said in the beginning, when we actually opened, that we were still trying to figure out what SQL Trail was, and I think I have a better understanding of what it is, now that we’ve had it for three years. At one point, in the beginning, the thought was “hey, we have this podcast, we have thousands of people that listen to it–” well–

Eugene:           Thousand. We have thousand of people.

Carlos:             We have thousands of listens. But yeah, we have a thousand. Yeah, we haven’t quite broken that– well, now I’m curious, I’m going to take a look. I don’t think we’ve broken the thousands. There may actually be one episode that’s broken the 2000 mark. But the thought was, “okay, so I like getting together with people, there’s the hallway track, you know, all of these things.” The thought was, “hey, can we bring the conversation to people, you know, kind of in a real-time event?” So, it turns out that whether that’s marketing, whether that’s what have you, and I digress here just for a moment. So 1700 is the most listens we’ve had of a single episode. Episode 144, as a matter of fact. The thought was, I’m not sure if that’s through marketing or what it is, but there’s also travel, I mean, I recognize that people wouldn’t come from Europe or Africa or things like that. That’s not necessarily the idea, but could we get people from the East Coast? And now, PASS is actually going to move next year because of the Convention Center, and there’s SQLSaturdays, so could we do something that’s longer than SQLSaturday? It’s not Summit, we’re not looking for 5000 people. They’ve had relays in the past, but again, could we have a different kind of event for those, basically on the East Coast that aren’t going to go to Summit but that could still come and get together? And it appears the answer to that question is no. I think people enjoy listening to the program, but it’s another matter entirely to demonstrate the value of then, “hey, let me, you know, pony up for a hotel and whatnot to come to an event.” So then, the question then, which we then ended up asking ourselves this year is “why are we doing this?” So one of the changes that we made, actually, is we tried to have a two and a half day event, weren’t getting the sign-ups, and then with about a month remaining, we actually just cancelled the first day and a half and just did a one-day event. I think it went well. I mean I guess we can get into that. But one of the things that I really enjoyed was getting our consultants together. We had some brown-bag sessions, some networking, if you will, because all the folks that work with me don’t always work together on the all the same projects, so it was a time for us to get together. And then, can we then have or present some training for ultimately a Richmond or a very Virginia-centric crowd? Folks that probably won’t have to– they can drive in a day and then drive home, so a one-day event seems much more palatable to folks. And I know you were kind of on the receiving end of some of these changes, Eugene, but in your first SQL Trail experience, what were some of your thoughts?

Eugene:           Sure, I will say that I was pleasantly surprised, which sounds like a back-handed compliment, but it isn’t, I swear. But I think part of that was like, I think some of my first impressions before SQL Trail were kind of that mismatch that you talked about, where it’s like, okay, it’s this small, close-knit thing, but you’re also trying to get people from like different states, and it’s like, “okay, where’s the sweet spot with this?” What I ended up liking a lot, being there, was that at least this year, a lot of it was, “okay, here’s other consultants with SQL Data Partners and then you apparently have an impressive amount of sway with Microsoft, because you had like 2 or 3 Microsoft speakers there.

Carlos:             We actually had 5.

Eugene:           Five? Yeah, I think there was three that I spotted, because there was Tim–

Carlos:             Devin, Brian and–

Eugene:           There was a bunch. I think kind of the sweet spot in what I really liked seeing was there was kind of two audiences that you were able to serve, that worked really well in the sense that you’ve got the random kind of local people that are like, “okay, I need to do some kind of training and this is $100. This is cheaper than a pre-con, most of the time.” But then you also had maybe a second group of people that overlapped with people that you’ve probably either done work for or had some sort of business interaction with. And I think that was awesome, because then, those people could see the expertise on display, and it was a chance to kind of kick the tires implicitly. Like an example was, I got to meet Greg Baldini in person, and I got to watch him present. And I’d met him like once online, whenever I was running in some DAX performance issues. And the first time I met him online, I was kind of intimidated, because we were running this issue and he sends this like long email of all these different things it could be, and I’m like, “oh, he’s going to think I’m an idiot, and I don’t know anything about DAX.” But then you meet him in person, and he just reminds me of like my cousin’s husband. Just very, little nerdy, but down-to-earth kind of guy, and it was great seeing him present and just seeing, “okay, he’s clearly dealt with a lot of these problems before. He’s probably made so many date tables he could do it in his sleep.” And so I think that to me had some real nice value add, where it’s like, “okay, for the people that just want some training, they can come and get some training. But for the people that have had some kind of business interaction with SQL Data Partners, they can kind of get an idea of, “okay, here’s the various skill sets we have on hand, on staff. We can do data warehousing, we can do data modeling, we can do reporting, all of this kind of stuff, and so there’s like an implicit escalation path, and you can get some of that for free if you stick around after the presentation and be like, “here’s my problem.” And you can get some of that for not free if later you’re like, “all right, I watched this guy and he seemed like he knew what he was talking about.” So, I liked that a lot, and I don’t think you would have gotten that nearly as much in past years, when you were trying to do more of that un-conference sort of thing. I think the networking piece is important, but I think there’s other ways you can kind of weave that in, still.

Carlos:             Right. Well, I think building that into the culture is kind of important, and letting people know that it’s okay to kind of ask the question and getting away from the– not that content is unimportant, that’s not the case, but the conversation triumphs the content. And that, if there’s a specific problem or experience or “hey, how would you do this?” that those conversations are ultimately the focal point, and if we leave a couple of slides left unmentioned, that’s okay.

Eugene:           Yeah, something I think would be great for next year, it’s like, if you had like a social media wall in the sense that there were a number of people that wanted to connect with me after I presented and all of that kind of stuff, and I think you would probably even get some of that between the listeners or learners or however you want to describe it, with each other, and so like an opt-in kind of thing of like, “hey, do you want to be able to connect with people? Give us your Twitter, give us your LinkedIn, and then just have a big poster. At the top, “okay, here’s all the speakers” and then a little line and then, “okay, here’s attendees” so people could go up and just be like (swishing sounds), and connect with people that way. I think that would be something that would be really cool for next year.

Carlos:             Yeah, we fell down big time in the social media department and there are lots of boring reasons for that.

Eugene:           Sure, sure.

Carlos:             One of the things we actually did last year, to that point, was the LinkedIn app on your phone actually has the ability to open it up and say, “find people near me.”

Eugene:           Yeah, I’ve done that the once.

Carlos:             And then you can just connect, assuming they have their picture, it’s very easy to make those connections, because people are also accepting it, it’s right there, and that would have been a great idea. We probably had time to do that at lunch and didn’t do that, and now as you bring that up, I’m kicking myself that that didn’t make the list. The other piece for me is that the one day, I mean, there was a lot of preparation, but I have been doing a lot of these one-day events for a while, and man, one day flew by.

Eugene:           Yeah, yeah, absolutely.

Carlos:             In compared to the two and a half days that we’d done in the past.

Eugene:           That makes a lot of sense.

Carlos:             But again, good and bad.

Eugene:           Yeah, no, no, I liked it, and if I’m being honest based on some prior impressions, I was kind of on the fence, because I’d never done it before, and like, “what is this like,” and I definitely came out of that excited for next year.

Carlos:             Right, right. Yeah, I mean, I also think our venue works well.

Eugene:           Oh, it’s gorgeous. No, it’s great.

Carlos:             Yeah, the Virginia Museum of History and Culture. I mean it’s not the Microsoft office, and it’s not–

Eugene:           Well, but it still feels professional. I think whenever you first told me you’re like renting this space at the museum, it reminded me of like whenever you’re trying to find a place for your high school graduation party and you’re like, “all right, what fire halls can we rent” kind of thing. But no, I mean, walking in there, you see all this art, you see the architecture, and you’re like, “oh, this is serious business.” I liked that a lot.

Carlos:             Right, that’s funny. And you’ll forgive me, I had a senior moment there, so Tricia, Tim, Brian, Frank and Devin were our Microsoft folks. Give a little shout-out to them. Yeah, so that was SQL Trail. I think the purpose of ‘why’ has changed a little bit for me. I’m still happy to have it, and like you said, I think making it a little bit more client-centric in getting together the consultants and kind of talking about what other– I mean, that idea still resonates or still is at the center of what we’re trying to do, is to get people together and talk. The audience might have changed slightly, but yeah, we’re looking forward to making it again. Now that we are just going to do one day, the timing might change a little bit. You know, we may not have it in October.

Eugene:           Yeah, if you could not do it the day before SQLSaturday Pittsburgh, that would be great.

Carlos:             Yeah, that’s right. Yeah, Matt, I chose first. I had my date first.

Eugene:           Right, right. Well, you heard I was going to be in Orlando, and you’re then like, “oh, we gotta move it.”

Carlos:             No, that’s right, that’s right. So yeah, October’s a very busy conference time, so I think it does lend, potentially, to some maybe even a different time.

Eugene:           It can be a challenge.

Carlos:             It can be a challenge. Scheduling is tough. But yeah, we’ll put it together, and of course those who are interested that want to come and see, we’d still love to have you, but we just recognize that it seems to cater more for the local folks.

Eugene:           That makes sense.

Carlos:             So, I guess having said that, Greg had lots of good feedback on his sessions. Any takeaways?

Eugene:           Yeah, I mean the key is you need to include pictures of otters in your presentation. That’s what I learned. He had his “OH NO Otter” for whenever you run into a problem. I joke, but I really liked his strategy of, “okay, here’s a common data problem”, like, “oh we’ve got a quarter column, like Q1, Q2, Q3, how do we show the previous quarter?” And I think it’s a strong– you know, before the show you and I were talking about like instructional design and I’ve been reading about that to be a better Pluralsight author, and I think it’s a strong design pattern to be able to be like, “here’s a problem, how do you deal with this?” And then you shut up for a second, and you just let it sink in and engage people that way. And he did a really great job of like, “our data looks like this, we need to do X. How do we do it?” And then everyone’s like, “eh.” And then you give that solution and people are invested, so he did that repeatedly, and I found that to be very engaging. I enjoyed that quite a bit.

Carlos:             Very cool. Yeah, I’m always interested in where people are and the things that they’re talking about. We did talk a lot more about Power BI this year than in year’s past. So one, when we thought we were going to have many more people, and actually we had a customer that said they were going to come that wanted to talk about artificial intelligence, and so we had Frank La Vigne there, but there weren’t too many people in his session, but the people that were there really, really liked that session. Talking about neural networks and I learned some things there that I had never known before.

Eugene:           Yeah, I asked both of them, “hey, tell me why I should attend your session,” and Frank’s definitely sounded very interesting, but from talking to him, it sounded like he was kind of going over “here’s the fundamentals of what a neural network is,” and so I picked Greg’s cause I need to get better at DAX logic.

Carlos:             Right, no, I think that was a good choice. I mean, Frank obviously did a good job. He and Kevin could have geeked out for a long time.

Eugene:           Well, it was funny, I was talking to Frank and I’m like, “you know you’re showing that, all right, anybody could build a neural network once you understand that it’s just like nodes and weights and a little bit of algebra.” And I was saying, “well, you know what you should do? You should build it out all in Excel formulas and literally show people that you could just do this by hand in Excel and it’d be ugly, but I think people would have that galaxy brain moment and be like, “oh my goodness, I guess this isn’t like some robot Skynet, it’s just like Excel formulas all hidden away.”

Carlos:             Yeah, so interesting stuff. And it once again reminds me that there are a lot of smart people out there, and I am not one of them. So, it was nice to re-remember my place in the universe.

Eugene:           Hey, someone’s gotta be the visionary, right?

Carlos:             There we go. So, okay, well, I think we were going to talk a little bit about Power BI. Should we do that?

Eugene:           We can if you think it’s not going to be– I don’t know what you’re going to title the episode, but yeah, we can talk a little bit about it.

Carlos:             Yes, “stuffage” is the word that I always use when it’s just like a whole bunch of stuff so that may get that idea. One of the things that I think we wanted to tackle was a quick review on when you think about buying Power BI, and we will maybe save more pieces for another episode or if people have questions, we can kind of tackle those. But that idea of, hey, $10 a month, right, and actually I put together a presentation on just this topic, and people are like, “$10 a month, I thought it was free!” So we start with free. You can download the Developer, Power BI Developer and start for free, so I suppose that’s true.

Eugene:           Well, yeah, Power BI Desktop is completely free. It’s kind of a like a razors and blades model. I think nobody in their right mind is going to be consuming Power BI reports through Power BI Desktop. And that’s not a criticism against people, but just the fact that it’s not really designed for that. You can do it, but it’s just from a user interface, perspective, not designed for it. If you’re going to be trying to consume file-based reporting locally, and you want to use Power Query and DAX, you’re better off just doing it in Excel, because Excel’s a lot more of a powerful and a lot more common and attractive, from a user interface standpoint, when it comes to consuming reports. And Power BI Free, I mean, yes, it’s free, but short of doing some reporting on your daughter’s lemonade stand sales, it’s not for business, because there’s no way to share the reporting at all, aside from publish to web, which makes your data publicly available, so, it’s a non-starter for any significant business use. But yeah, it’s free, which is great for playing around with and learning. I mean, especially compare that to say, some of the more esoteric BI solutions, like let’s say Domo. I heard about it and I wanted to just poke around with it. I’m pretty sure it was Domo, it might have been Sisense. But one of these, you go to the website and you’re like, “hey, I just want like a developer version. Can I just get a trial?” And they’re like, “fill out this form to get a call from Sales.” I’m like, “no. I’m not doing that.” So I do definitely appreciate that there’s a free version with Power BI. You know, you don’t have to go through a quoting process with Sales just to be able to play around with it. But yeah, so just like the deployment methods, licensing on Power BI is confusing, because they decided at the very beginning to do agile development, and kind of user-voice driven development, you know, the highest voted features get implemented, and that’s great for agility and delivery speed, but it makes the messaging less coherent, because every six months it seems like they’re adding either a new licensing method or a new deployment method. But for like 95% of users, what you’re going to do is you’re going to get Power BI Pro and pay $10 per user per month, and that gives you the vast majority of features and functionality with Power BI. That’s going to be the most common solution in a majority of cases.

Carlos:             Yeah, that’s fair. I mean there are some limitations, that we can talk about, but that can get you going. You can share your reports and you get to take advantage of all of the other pieces. Now, there are some other things that you’ll need, ala the gateway to get set up.

Eugene:           Yeah, unless your data’s all living in the cloud, yeah, you’ll need the data gateway to be able to do refreshes and that sort of thing.

Carlos:             Right, and if you’re not using Office 365, and I would imagine that most candidates are probably either thinking about Office 365 or have done it. It’s almost kind of funny, mail migration was one of the first cloud adoptions, and there has been a lot of Office 365. So I guess I’ll ask, if someone’s not using Office 365, do you think they’re just going to need it, to use Power BI or do you think it’s possible to use it without having that Office 365 license?

Eugene:           That’s not an issue that I’ve run into. But yeah, I would think you would need it. You would either need it for just the raw log-ins, you can have an Office 365 tenant and have users that the only licensing that’s assigned to them is Power BI. I’ve done on my private tenant of like, “okay, I’m going to create test_user and just assign them a Power BI license.”

Carlos:             But then that’s another account they have to keep up with.

Eugene:           Yeah, it’s a pain. I’m not aware of any like, “okay, here’s how to use accounts from somewhere else like using Azure B2C or whatever it would be to do that. The other option is, you can use Power BI Embedded and manage security yourself, but it just feels like overkill at that point.

Carlos:             Sure, yeah, you have to have that application.

Eugene:           Yeah, I mean you’re just doing a lot of custom dev just to be able to get access to Power BI and then at that point, are you better off just doing Tableau or Qlik or something?

Carlos:             Sure. Yeah, so you use Embedded when you don’t know who the users are or you don’t want to manage those log-ins, i.e. your SaaS is the ideal scenario, there. Internally, it doesn’t quite generally jive, so it’s a slightly different use case. And then again so one of the other things we have on our presentation is the slider. So everything is great about the cloud is that, “oh, you just slide, and you need more? Slide.” The challenge with Power BI is that you go from $10 a month to $3500 a month. That’s the slider, right?

Eugene:           Yeah, there’s a big jump there. This is what I don’t like about Power BI licensing is it starts off with very clear messaging and very simple pricing. It’s $10 per user per month, whatever.

Carlos:             Period.

Eugene:           Yeah, exactly. And then it’s like, “unless you want to do embedding or you have more than, say, 250 users, and then we start getting into like three different tiers and 10 different skus.”

Carlos:             And you’re like, “oh yeah, this is a Microsoft product.”

Eugene:           It’s like, “aw, we’ve wandered into like Azure VM territory.” Because you have the A tier, which is for application development, or– I can never remember if it stands for Azure or application, but that’s for, “okay, your SaaS, you’re embedding it into your software,” and that has eight or nine different levels, like A0 to A8. And then if you’re looking more for internal usage, there’s the EM skus, which stand for embedding, but not to be confused with Power BI Embedded, and that’s for Office 365 embedding. So, basically, you want to be able to start leaning on free users, and you’re fine with just putting this in Teams or SharePoint online or that sort of thing, and I think that starts somewhere around $750 a month and goes up to two or three thousand? I forget. And then there’s Premium, which is just everything, and that starts at $5000 per month and goes all the way up to like $23,000 or something like that. But, you basically get the full PowerBI.com experience, and the way I think of it is instead of buying people these bus tickets for $10 per user per month, you just buy a whole bus and you cram as many people on as you can. So again, the licensing is frustrating. You still need to pay extra for the bus driver, so even if you’re putting in this big burlap sack of $5000 of money and handing it over to Microsoft, you go to them and you say, “hey, this covers me being able to create reports, right?” And they say, “no, you have to pay for a Pro license.” And I don’t understand why. Like, “well, if I’m giving you $5000 per month, you would think you would do the courtesy of giving me like the free mint or the fortune cookie.” I’m just saying.

Carlos:             Yeah, yeah, that’s a gimme.

Eugene:           “You know what? Yeah, we’re not going to charge you for report creation.” But no, they do. Like if you’re doing Power BI Report Server, you’re doing Power BI Premium, any of these things where you’re just throwing oodles of money at Microsoft, you still have to license on top of that, the actual report creators, not just the report consumers. So, yeah, it’s a mess.

Carlos:             SQL Trail licensing is much, much more straightforward.

Eugene:           It’s true. You’ll have to deal with it for more than a day, that’s for sure.

Carlos:             Yeah, that’s right. Okay, well, unfortunately, our time is now short and we need to wrap this up, but thank you Eugene. Oh look, there goes my even little– I don’t know what that was.

Eugene:           Some reminder that says to get off.

Carlos:             My phone was telling me, “come on, you gots to go. You’ve got people to see.” Not that we don’t want to still be with you, compañeros, but we’ll have to tune in another time, another episode. So remind us, Eugene, how can people get in contact with you if they had a question about today’s episode?

Eugene:           Well, in honor of Kevin, I’d say, “you can hunt me down in the greater Pittsburgh area,” or you can find me on Twitter at sqlgene or my blog is sqlgene.com.

Carlos:             There we go, yeah, we need a “Where in the world is Kevin Feasel” time.

Eugene:           I’d buy that game.

Carlos:             Yeah, that’s right. Much easier to find me on LinkedIn at Carlos L Chacon. That is going to do it for today’s episode. Thanks again for tuning in, compañeros, and we’ll see you on the SQL Trail.

Leave a Reply

Back to top