Having obtained an MSc in Software Engineering in the early 90s, I have to confess that my "three-month intensive research project" did not take three months and was not intensive.
Microsoft grabbed the headlines this week when it announced a Professional Degree Program at its annual partner conference. It starts with data science. Microsoft claims to have consulted data scientists and companies that employ them in order to ensure students the core skills for a job in this extremely hot field of …
Could not agree more. The author clearly has no clue about the workflows and challenges of analysing big data. He betrays his cluelessness with the "surely" qualifier when moaning about the fact that there's no Relational DB theory in the course. Relational databases were supplanted in this domain *precisely because* they're useless at big data. Nobody cares about the client/salesperson/sales order/HR table referential integrity that RDMS obsess about.
"Nobody cares about the client/salesperson/sales order/HR table referential integrity that RDMS obsess about."
In order to play about with your large scale data sets you need a working business to pay the bills. A business that sells goods and services. If there isn't someone to obsess about ACID qualities in the database* that supports that business then your big data is going to get its budget yanked from under it and will, in any case, be pointless.
*The database may well be providing the data sets in the first place.
There are several online courses with similar purposes, structure and costs -- I am sure Coursera offers one. There is no need to pay for any software mentioned on the syllabus (R, RStudio) so the investment is lower, and more people can take it.
Why not call it "Data Science with Microsoft Tools" and offer it for free? People would have to pay somehow to have access to the tools anyway.
"That’s OK if this was a training course, but if it’s meant to be training data science professionals, why not teach generic (or standard) SQL?"
As the late great Admiral Hopper said, the best thing about standards is that there's so many of them. I can imagine that it would be a useful exercise to have students design a set of queries that would work across multiple databases. But if you are going to touch on stored procedures and triggers, and I think you should, then I don't know what standard means.
One of the ways Microsoft achieved its operating system monopoly was to get software developers to do things the Microsoft way, using Microsoft ASP, .NET, Active X, and all the other features that lock one into the Microsoft operating system and the Microsoft (at the time) Internet Explorer. (Had they seen fit to give away all of their software development tools, too, we would not be having any conversation at all about OS X and all the Linux distros.) What Microsoft is doing for and to data scientists with this course reeks of the same odor. In the old days, they did not call their product managers evangelists for nothing.
- why would I want to pay money to a company with such low ethical standards or trust the content of a course produced by a company that has made quite so many monstrous cock-ups over the years?
And I happen to LIKE T-SQL (insofar as this lowly pleb uses it, anyway)!
Biting the hand that feeds IT © 1998–2019