I've been an avid "fixer" of things for almost four decades. One of my earliest memories is of helping my grandfather fix a broken antique clock. (I don't remember what part I played beyond the fact that the job required my expert application of a little yellow plastic hammer.) We must have fixed it because it worked for another decade before it finally broke terminally.
As I grew up, I progressed onto bikes, then TVs and other assorted appliances (back in the day, when all you looked for was a valve that wasn't glowing like the others, which you replaced with another one that looked the same), and then cars (lots of cars), before coming to PCs and consumer electronics, which is where I've been for over 20 years.
Now, I'm not going to try to fool you into thinking that I'm some "fixing guru" or "tech ninja." I'm not. While I have a pretty high success rate when it comes to resuscitating things, I've had plenty of failures, too, and I've been responsible for letting the magic smoke out of a lot of devices by doing something daft.
But, over that time, I've built up a set of rules that I keep in mind when fixing things. I call them the "Prime Directives," not because I'm a huge Star Trek fan, but because they're important, and bad things tend to happen when I violate them.
I present them here in no particular order.
#1: If it ain't broke, don't fix it.
This one belongs right at the top.
If something is working, leave it alone. Don't mess with it. Don't try to make it better. Don't fiddle with it.
Just let it do its thing.
I've lost count of the number of times I've come across someone who read or heard about some "tweak" that would make their PC or router or network faster (most of the time pursuing some snake-oil nonsense someone made up), then they dove in with the best of intentions, and they break the thing they were trying to make better.
My philosophy nowadays is that when I buy something, I set it up and then leave it to do its thing until it needs updating or it breaks.
#2: Never say "I'll have that fixed in five minutes"
At least until you've troubleshooted the problem and know what the fix it. Even then, expect the unexpected!
Like Star Trek's Mr. Scott, I'd much rather give a pessimistic timeframe and go on to impress, than an optimistic one and go on to disappoint.
#3: Just wait a minute.
You would be amazed how many problems go away if you give it a minute or two. This is especially true of network and internet-related issues. Yes, it might be worth having a quick look through the logs to see what happened so you can perhaps stop it happening again, but unless you promise (or demand) high reliability, don't start chasing it.
Give the problem a chance to catch its breath and it might just fix itself.
#4: Don't keep fixing something until it's broken again
You know the routine. You start out fixing something, but then you see it as a good opportunity to upgrade stuff.
There are no end of examples. Maybe your network switch is dead, but while you're changing that you notice some of the cabling is a bit rough and start tackling that job. Or you're swapping out a dying hard drive in a PC and think that this is a great time to add more RAM and tweak the BIOS.
The problem with this is that if you have problems it makes it hard to know where to start. Is it a problem with your fix, or a problem with something else you touched?
Another example of this is where a PC is misbehaving, but the owner thinks that might be a good time to upgrade the operating system and drivers. I can't believe the number of PCs I've come across that have ended up in a worse state than they were in before they were "fixed."
#5: Don't trust the work of people who've already tried to fix the problem
A common thing I hear when troubleshooting is "oh, you don't need to do that, the other guy/gal already tried that." This is how you get dragged into the weeds and end up spending hours chasing down something simple.
Don't trust others who've tried to fix the problem. Why? Because they, for whatever reason, didn't fix the problem, and that might be because they missed something simple.
#6: Don't be a parts changer!
This is one from the auto repair industry that is also prolific in the PC repair industry. Someone tries to "fix" a PC by just randomly throwing new parts at it. Firing the "parts cannon" at something is a really expensive -- not to mention hit-and-miss -- way to try fix a problem.
I once came across a PC that had the motherboard, RAM, and power supply unit replaced before it was discovered that it was the processor that was the problem. Someone else replaced an iPhone, which would no longer charge, when the problem was the charger.
Make sure you diagnose the problem, and not just throw parts and guesses at it.
#7: Use (and contribute to) the hive mind.
The internet is a powerful resource for diagnosing problems, but it only works if people contribute to it in the form of blogs and forums (or the awesome repair guides over at iFixit).
If you come across an unusual problem, share it. If you solve it, update the post or forum with your solution, being as clear as you can so that others can follow in your footsteps.
#8: Document what you do
It's much easier to make a change than it is to undo that change, especially when days, weeks, or maybe months have gone by. I know it seems pedantic to document the changes you make to something, but when it comes to complex setups such as networks, it's really the only way to go.
Not only might you be helping yourself down the line, you're helping the next poor soul that might be tasked with fixing something.
#9: Know when to give up.
Sometimes a thing you're trying to fix just doesn't want to be fixed. You might have the tools and be armed with the know-how, but the repair is either too costly or complex or is just taking more time and effort than it's worth.
Then give up (or, if you are uncomfortable with that idea, think of it as a retreat).
This can be especially true of items that are cheap to replace. For example, you're not going to see me wasting my time fixing a cheap cable or power adapter, or something like a cheap consumer network switch or router. Even if you can get the parts, by the time you've put in the time and money, you're worse off than if you'd just replaced it.
Obviously, if the piece of kit is specific or customized in some way, this may not be the case, but for cheap off-the-shelf parts, there's no sense in wasting time with an attempted fix.
10: Take care.
The gadgets around you are home to a number of dangers, from high voltage waiting to zap through your body, lithium ion batteries that are just waiting for you to short them out so they can blow up in your face, to dangerous chemicals that can cause you long-term harm.
Take the time to be safe. I routinely wear safety glasses because I intend to leave this world with the same number of eyes I came into it with. Also, after years of not giving a damn about my skin, I'm now a convert to nitrile safety gloves. And I'm super careful around high voltage, especially now that the "hair that sticks up" look is out of fashion.
But all jokes aside, do take care. Repairing something is not worth your life, so if in any doubt, just don't. Get a professional in, or at least someone who values their life less. I've seen a lot of nasty things happen with CRT tubes and capacitors and high voltage and lithium ion batteries -- stuff that looks awesome in a YouTube video, but isn't so awesome when it happens on your living room table.
Only the other day I was fixing a busted Bluetooth wireless speaker, and I can tell you that I was much happier continuing with the repair once I got the lithium ion battery pack out of it in one piece. These things are in everything and are safe until you cut into them, bend them, stress them, or short them out, at which time they can become horribly unpredictable.
If you've got a cool tip or story to share, feel free to leave it in the comments section.
PREVIOUS AND RELATED COVERAGE:
Equip your iPhone with some the best accessories available without breaking the bank.
Dozen Raspberry Pi alternatives Here is a selection of single board computers for homebrew projects and automation, with prices starting at only $5.