Thread by @marcorus

This morning I read the phrase "legacy Power BI" on the site of a "modern" tool, which uses SQL as a query language.

With all respect, marketing hype should understand when their words become counterproductive.

#powerbi #sql #mdx #dax #sequel #legacy #modern

1/19

I know hundreds of issues and limitations in Power BI, but when it comes to comparison, I can hardly understand why "moving data" should be considered "legacy".

2/19

How do you call any "caching" mechanism?
What is an "index" if not a copy of data?
Seriously, do you know business users who really care about how many copies of the data are around the world?

3/19

No, I will not mention products I don't really know. I will not be involved in a discussion about which tool is more modern than others.
I like to know about real advances in BI tools (including better implementation), I am happy to ignore the rebranding of good old stuff.

4/19

My starting point is that #powerbi is a model-based tool. Actually, it is based on Analysis Services, which debuted in 2005 (it was called OLAP Server in 1998, but because of the limitations it had at that time, let's ignore it).

5/19

Marketing is very good at finding acronyms to generate hype, even though few innovations come by name-based hype cycles.
Examples: OLAP, DSS, EIS.

en.wikipedia.org/wiki/Online_an…
en.wikipedia.org/wiki/Decision_…
en.wikipedia.org/wiki/Executive…

6/19

Analysis Services Tabular is the engine used in Power BI and it was launched in 2012.
Power BI launch date is 2015.
Technologies used in Analysis Services and Power BI have been launched by other products before those dates.
en.wikipedia.org/wiki/Microsoft…

7/19

Columnar databases: first ideas in 1969, first commercial implementations in the late 1990s.
One of the most known server products based on this technology is Vertica, launched in 2005.

en.wikipedia.org/wiki/Column-or…
en.wikipedia.org/wiki/Vertica

8/19

SQL language: the first formalization in 1986 by ANSI, but it was designed in the 1970s.

If you are curious about the different pronunciations ("ess cue el" or "sequel") is because the original name was SEQUEL, indeed.

en.wikipedia.org/wiki/Edgar_F._…
en.wikipedia.org/wiki/SQL

9/19

MDX language: first spec in 1997, launched in 1998, adopted by many products also outsider the Microsoft ecosystem.

en.wikipedia.org/wiki/MultiDime…

10/19

DAX language: Microsoft proprietary language, first released in 2010 (Power Pivot for Excel).

It aims to solve the same business problems faced by MDX with a more intuitive language.

Hint: if you think DAX is complex, try MDX scripts.

sqlbi.com/blog/alberto/2…

11/19

Is Power BI "new"? Not in a single part.

It collected different existing Microsoft technologies in a single product, improving and optimizing performance and user interface.

12/19

Is Power BI modern?
Of course.

It is a model-based tool.

Was it the first one?
Not even close.

sqlbi.com/articles/power…

13/19

What was the first model-based tool?

I'm not sure, but probably Business Objects (now SAP-owned), originally named Skipper SQL, launched in 1990 and widely adopted in the industry in the 1990s.

It's SQL-based and real-time, 30 years ago!

en.wikipedia.org/wiki/BusinessO…

14/19

What did we do in the last 10 years, though?
Implementation and optimization.

Which are important: ideas without implementation are just dreams.

New ideas don't appear every year. Not even every decade.

New ideas come in minutes, but their implementations need decades.

15/19

We just like to change names for the same technologies.

Don't worry, it appears everywhere.
Many machine learning / artificial intelligence products are just rebranded data mining algorithms of the past century.

I know that.
I play the game.

16/19

So please, marketing departments: if you don't want to be ignored, please just find another name.

Avoid adjectives like "revolutionary" or "modern" when you are talking about a technology you did not invent.

Avoid "legacy" when it could be used against your technology.

17/19

You can do a good job with a great implementation, though.

Emphasize "fast", "intuitive", "easy", "frictionless". These are good terms, users love them.

Just don't throw a boomerang with your words.

18/19

Age is not necessarily a defect in your SQL-based technology.

SQL is not new, but it is in good shape and everybody loves it.

For the record, I pronounce it "Sequel" just because it seemed more fluent.

Peace and love.

#powerbi #sql #mdx #dax #sequel #legacy #modern

19/19

Addendum: I see your replies, please understand I’m not going to debate about Power BI vs other tools or Power BI features.
I wanted to talk about marketing words.
That’s it. No flames.
Thanks!