The right title for this blog post should be “Guidelines for SQLBI content publishing”, but it would be too boring. And, at the time of writing, the more common questions we receive about articles that are nowhere to be found on the SQLBI website are related to DirectLake and Visual Calculations.
When talking about DirectLake and Visual Calculations, we must keep in mind that these features were released two months ago and less than two weeks ago, respectively. This seems like a significant amount of time from the perspective of weekly announcements of new features, but it’s definitely a very small amount of time to produce the quality of content we like to publish. Thus, I would like to explain how we make the content you can read and watch on the SQLBI website.
In general, we do not publish an “article” about a feature that was just announced or released. Of course, there are exceptions, but we tend to rely on blog posts (like this one), which do not have the same constraints we apply to “articles” (more on this later). Our newsletter has a “Latest News” section where we link to announcements and content (from Microsoft or other authors) related to features and analyses that we consider interesting for our readers and subscribers. On the X/Twitter and LinkedIn social media platforms, I occasionally comment on the “latest cool new features/gadgets”, most of the time without taking it too seriously. But when it comes to SQLBI articles, we try to follow a few guidelines.
Ideally, an “article” does not have a short expiration date and does not repeat information already available elsewhere. An article should cover a topic and provide educational content. There are different types of articles: solving a specific problem, explaining an abstract concept, or teaching a particular technique. Any of these scenarios translate an experience (from past work, from research and development, from studying) into something that can be helpful to the reader for personal growth or to solve specific challenges.
When something is “new”, it’s hard to have accumulated any significant experience. Sometimes, we are lucky enough to test a new feature in preview, and when this allows us to get enough knowledge to translate into something helpful, we are glad to share that as soon as possible. This was the case for calculation groups, which were released much later than we hoped, causing a few months of delay in publishing the second edition of The Definitive Guide to DAX – on the positive side, we had months of experience to share when the feature was made public.
However, this is not always the case. For different reasons, we started studying DirectLake and Visual Calculations recently, not long before they became publicly available. Thus, it is absolutely not surprising that we have not published any articles about these topics yet! This doesn’t mean we are ignoring them – actually, quite the opposite. Let me share more details about our plans.
For DirectLake, we will certainly create content to explain the differences and similarities with VertiPaq, and we will invest in tool improvements that could also help understand what is happening with large models. However, DirectLake is not at the top of our priority list. We want to improve the tools first, and we are already working on other features of VertiPaq Analyzer (more on those when they are ready). Because we are discussing free, open-source tools, any additional help and support would be welcome. Otherwise, be patient! (there is a joke that says that “P” in Power BI stands for “patience” – we could say the same for the “P” in “Phabric”…)
For Visual Calculations, we are preparing a complete whitepaper for SQLBI+ subscribers in addition to several articles that will be made public on the SQLBI website. They are just not ready yet. We want to provide a good conceptual description of the model behind the visual calculations. If you tried to write CALCULATE in a visual calculation, you might wonder whether you are using a different type of filter context or row context… well, actually, you’d be dealing with neither of these. Enter a new concept: the visual context. Do I have your attention now? Well, this topic definitely has our attention at SQLBI, and we do not want to write anything wrong. Hence the need for a little bit of patience. But what’s a few weeks in the grand scheme of things? Nothing, especially if you consider that visual calculations are in preview and you should not use them in production!
Do not worry: we will cover these topics. When the time is right. Soon, but not too soon.
Evaluates an expression in a context modified by filters.
CALCULATE ( <Expression> [, <Filter> [, <Filter> [, … ] ] ] )