Is there a best HA solution? Is there a best data type for sales data? Is there a best way to back up a database?

I see so many posts asking for the "best" solution to a problem. I suppose that's only natural since we want to be sure we are doing our best at work. However the "best" solution varies depending on the situation. What an expert might recommend for one company, might not be what they recommend for a different company.

Instead of asking for the best solution, what I'd rather see people doing is trying to learn more so they can make an educated decision for their situation. It seems that you ought to be able to just get an answer that would solve your problem, but it might not do so, even if it's what is recommended by someone you view as an expert.

If you plan on engineering things, or building them, or even just assembling parts, it pays to understand enough to make some educated guess about what is appropriate. While it might not be perfect, and might not be a good solution for your situation, if you at least understand why it was chosen, you should have a better idea about why it might not be working.

We ought to do the same thing in software, especially when so many of us act as architects, modelers, engineers, designers, and assembly people, often all in the same day! We aren't experts in all these areas, but we can learn to make better judgments and to question our assumptions on a regular basis.

Even if you come up with the best solution for your situation today, it might not be the best one tomorrow.