One of my biggest pet peeves is when someone tries to explain their modular or componentized system as being “just like Legos”. That is almost never true.
One of the first misuses of the term was back when I first started working on web services hosted in Tomcat. It seemed like anyone who made a tiered system or one that talked through SOAP liked to describe it as being ” just like Legos”.
Often it is just a lazy way to report on a new technology. Like this article, which prompted this rant.
http://www.engadget.com/2014/10/03/google-x-giant-modular-displays/?ncid=rss_truncated
Now while I’m a huge AFOL, I’m not so picky that not using the official term “Lego brick” instead of the common “Legos” bothers me. So what is the problem?
First off, Lego bricks, with a few Mindstorms exceptions, don’t care what they are connected to. If the stud fits in the hole, that’s all it needs. Contrast that to a web service, where the format and intent of the connection is critical to the function. Or contrast to the displays mentioned in the article. While they don’t give details, even in the most modular of display applications, you need to know which display is connected to which port and what orientation so you can render an intelligent image onto the combined layout. Lego bricks don’t care what the brick on the other side of it is – no information passes through them.
With a Lego model, studs are studs. You can connect them together however fits, and whether it is good is evaluated externally by the viewer (art is in the eye of the beholder). A web service can’t have its display layer and its database layer swapped and still function.
Yes, I’m probably over analyzing it. But it dies bother me and I wish people would just put a little more effort into accurately describing systems.