Welcome to Shop Talk, a regular Q&A-style broadcast hosted by the Triangle Area SQL Server Users Group (TriPASS).
You can watch each episode on the TriPASS Twitch channel, and get information on this show and everything else TriPASS has to offer on our Meetup.
After each episode airs, you will see the show notes as well as an embedded link below.
Our first topic of the night came from Mala, who couldn’t make it this evening. The people behind ggplot2 (implementing the Grammar of Graphics for R) have come out with a SQL variant called ggsql. It’s still pretty early on in development, but it does look promising.
Our next topic comes from Andreas Wolter and deals with a permissions bug in SQL Server. To me, it looks like certain permissions get cached (or maybe not cleaned up correctly) and can lead to people having a bit more access to specific external models than you’d think.
The third topic was a blog post from Randy Knight on Group Managed Service Accounts (gMSAs). These have been around for quite a while, but they’re definitely worth pointing out to DBAs working on-premises or outside of Azure.
The final topic also came from Mala and covers SQL Server’s MCP offering. Jerry Nixon describes using the Data API Builder (DAB) to serve as a database intermediary, so that you don’t offer up direct database access to a language model.
Our first topic of the night was an update on Raleigh Day of Data, coming up on May 23rd of 2026. You can still sign up for the event at the Day of Data website. Attendance is free and lunch is a nominal fee.
Our major topic this episode covered a Rebecca Lewis blog post on multiple SQL Server vulnerabilities. We reviewed the (lack of) information available and I pointed out a few additional details.
I also spent a bit of time talking about a pair of posts from Anthony Nocentino on the SQL Server Kubernetes operator, as well as planned failover. Since the time of recording, Anthony also put together a post on unplanned failover.
Our first topic of the night was an announcement that Raleigh Day of Data’s schedule is live. You can sign up for the event at the Day of Data website. Attendance is free and lunch is a nominal fee.
Our major topic this episode was a discussion of two interesting SQL Server tools. The first is Erik Darling’s performance monitoring tool. The second is David Seis and Straight Path Solutions’ SQL Server health check using dbatools.
Jeffrey Snover has a great rant on how Microsoft has had multiple, overlapping but incompatible strategies for implementing graphical interfaces on Windows.
Along the way, I got sidetracked a bit on Trusted Platform Modules and my annoyance at Windows 11 for not working without a module.
Our first topic of the night was an announcement that we were sending out speaker notices for Raleigh Day of Data. The schedule isn’t out quite yet but it’ll come soon. You can sign up for it at the Day of Data website. Attendance is free and lunch is a nominal fee.
Our major topic this episode was a blog post about the challenges of software development when the industry is putting a heavy focus on generative AI. Along the way, I ranted a bit about how Google’s natural search fell off a cliff. I also spent some time searching for Firehound, a neat service that appears to be down now and chatting about xv.
Our first topic covered several upcoming conferences. We have SQLcon in Atlanta, with multiple SQL Saturdays and Days of Data following it. The southeastern US will have several major events between now and the end of May, with Raleigh Day of Data 2026 taking place on May 23rd in Durham. The call for speakers is still open for another week or so, but get your sessions in!
Our major topic this episode was Redgate’s 2026 State of the Database Landscape. We went through a PDF of the report, though admittedly, I got side-tracked multiple times on graphics I hate and a detailed discussion of how brown, white, and black are not actually colors.
As far as the actual report goes, I did mention some spots that seemed surprising to me, and do recommend picking up a copy of the report, even if using circle sizes to describe relative percentages is a terrible idea.
We spent a brief amount of time talking about current issues in SQL Server 2025, including one around slow behavior when bringing online a large number of databases, regardless of whether you have availability groups set up. We also took a bit of a diversion around this to discuss trace flags, trace flag scopes, full-text search, and how much I dislike James Joyce.
Our final topic was a blog post from Greg Low. Mala was shocked that SQL Server allows you to update an underlying table using an UPDATE statement on an inline table-valued function. I noted that you can do the same thing against views when there is no ambiguity in the view definition, and we speculated as to why you might do this. My guess is that it was to entice some Oracle developers and DBAs over to SQL Server, as the Oracle perspective for a very long time was to create a set of views separate from the base tables and only grant user access to those views rather than the underlying tables. The why behind this behavior is speculation, but I will say that regardless, I don’t recommend doing this because it’s pretty confusing behavior.
The first major topic of the night was a question that Mala asked about diversifying from SQL Server onto other platforms. I talked a bit about the overall landscape, the types of data platform technologies available, and also what does and doesn’t make sense. I think, in a lot of cases, the better answer is to dance with the one what brung ya, but there are specific circumstances in which other technologies can outweigh the cost of additional administration.
Mala also covered an interesting development from the world of ScriptDOM. A new commit to that project means that the tool now keeps comments in place. This is great when extracting database objects from a database to go into source control or rewriting database code automatically.
Our last major topic of the night covered an article from Addy Osmani on some of the more recent developments in software development, particularly when using language models to write code.
The first major topic of the night was about a survey Brent Ozar runs annually regarding data professional salaries. The 2026 results are in and I did a bit of poking around with a pivot table. We didn’t do any real in-depth analysis simply because this was my first look at the data. But we did talk a bit about inflation rules of thumb, sample sizes, and a few interesting things in the dataset. I definitely appreciate Brent continuing to put this together and making it available for people to use.
The next topic covered Azure pricing and exchange rates. Thomas Rushton retold a story (in blog format) regarding some exchange rate volatility in 2022 affecting British Azure customers because everything is ultimately priced in USD and then spot exchange rates determine the conversion factors to other currencies. For currencies that are normally fairly stable, that’s not a huge deal. But in this case, there was a 15% or so difference, simply because a spike happened during the time of month that Azure bills.
Our last topic of the night comes from a John Deardurff blog post. John recently wrote about several tips for teaching people about technical topics and we gave our takes on the tips. In general, I fully agree with John. There are a couple of tips that I don’t necessarily observe, but we did talk about when and why it can make sense to do so.
The first major topic of the night was about a pair of Days of Data. The first one is that Raleigh Day of Data 2026 is live. You can register (either free or a nominal fee for lunch). You can also submit for our call for speakers for the event. Our event is going to take place on Saturday, May 23rd in Durham.
In addition, Day of Data Richmond 2026 is also live. This event will take place on April 11th in Richmond, Virginia. The registration page for this is also open and the call for speakers will be open until January 18th.
Our other major topic was around vector search capabilities in SQL Server 2025. Mala has been investigating this and shared her thoughts so far.
The first major topic of the night was an article from Steve Jones on the evolution of SQL Saturday. Steve has officially announced the new name, Day of Data. We talked a bit about this as well as SQL Saturday Raleigh 2026 (or Raleigh Day of Data 2026?). That’s coming up soon.
In case you missed it, the TriPASS call for speakers for 2026 is currently up. We’re accepting submissions for our three groups: advanced DBA (virtual, 300-400+ level topics), the main meeting (general purpose, with a preference for speakers who can meet us in person), and data science/business intelligence/artificial intelligence (virtual).
Mike hit me up before the show and asked for my thoughts on PGlite. I took a look at the website, speculated a bit on good uses cases, and compared it to a pair of other libraries I’m much more familiar with in SQLite and DuckDB.
Mike also wanted to talk about DAX Lib, jumping off of a recent Curated SQL post. This leads to DAX Lib, a website hosted by the sqlbi team of Marco Russo, Alberto Ferrari, and Kurt Buhler. The idea is to provide people with a repository for user-defined DAX functions that they can use and re-use in Power BI reports. I think it’s a neat idea and it does fit in nicely with sqlbi’s (paid) Tabular Editor 3 product.
A member of chat asked us to end with a useful SQL tidbit for newbies, and that reminded me of a Louis Davidson blog post on using FORMATMESSAGE. This was built on a bit of yes-and development between Louis and me and I was able to show the audience the final product and how the FORMATMESSAGE() function can make this sort of string formatting considerably easier. At least, so long as you aren’t using date types (or don’t mind converting them to strings first).
Our major topic was the release of SQL Server 2025, and this took up pretty much the entire show. We looked at many of the new features in the product, discussed expected adoption curves, and I punched down on SQL Server Integration Services for a while. But in fairness, if they didn’t want me to complain about it, they’d actually have provided meaningful updates to the product this decade.