This note from TI presents a practical and non-defensive discussion of what to avoid doing in decoupling: http://www.ti.com/lit/an/sloa069/sloa069.pdf

That's a very good article. I ran across it some years back and the thing that struck me the most was the discussion on how capacitor orientation is so significant, particularly for SMD devices.

 

Bad advice.

Same as someone who says,

"I never use ESD protection and I've never experienced a problem".

"I never wear a bike helmet and I'm still alive".

"I never wear a seat belt, why worry now".

Re: bad advice
I agreed earlier that this was bad advice, but if there is no problem why apply a remedy?

 

Re: bad advice
I agreed earlier that this was bad advice, but if there is no problem why apply a remedy?

So if someone that has never worn a seatbelt has never had a problem, then they should not apply a remedy and keep on not wearing a seat belt?

A better question -- why should anyone ever follow what is acknowledged to be bad advice?

 

So if someone that has never worn a seatbelt has never had a problem, then they should not apply a remedy and keep on not wearing a seat belt?

A better question -- why should anyone ever follow what is acknowledged to be bad advice?

Re: "why should anyone ever follow what is acknowledged to be bad advice?["
Economy. I have a lot of LED flashlights that are nothing but three AA or AAA batteries and LEDs. Not even a current limiting resistor. Bad design? Of course. Does it work such that it is? Yes.
When I breadboard a small circuit I seldom use such filter caps. No problems, but for advising the youngsters; if you are going to give advice, give good advice. For others advice can vary from proper designs.
So what have you actually done? Have you tried to build stuff not using such capacitors?

No, I don't always tie up unused gates.
I seldom use a wrist strap.
No, not all my parts bins are static proper.
No, I don't always simulate my circuits before I build them.

Okay, I'm a barbarian.
Your theory is correct ... for children.

 

Re: "why should anyone ever follow what is acknowledged to be bad advice?["
Economy. I have a lot of LED flashlights that are nothing but three AA or AAA batteries and LEDs. Not even a current limiting resistor. Bad design? Of course. Does it work such that it is? Yes.
When I breadboard a small circuit I seldom use such filter caps. No problems, but for advising the youngsters; if you are going to give advice, give good advice. For others advice can vary from proper designs.
So what have you actually done? Have you tried to build stuff not using such capacitors?

Sure. The bad advice that is being referred to is not that there are applications where bypass capacitors are not needed, it is the advice to first try the circuit without them and then, if nothing bad seems to happen, assume that you don't need them. This approach will almost certain get you bit on a regular basis.

The "good" advice would be to DESIGN the circuit such that you have a reason to include them or not include them. Circuits like your LED flashlights don't need them -- so don't use them in those circuits. If the internal resistance of the batteries are sufficient to limit the LED current to a level that meets your requirements, then there is no need for a separate current limiting resistor. Whether or not either of those constitute "bad design" depends entirely upon the metric by which design quality is measured -- if a major element of that metric is cost or component count, then including components that are not absolutely needed would tend to qualify it as a "bad" design. But those decisions should be done as part of the design process (which can certainly involve tests to help define the limits of the design options) instead of designing via happening (through something together and hope that it happens to work).