I saw someone's solution to a problem I've always struggled with and felt dumb for not figuring it out myself

Posted by haddock420@reddit | learnprogramming | View on Reddit | 14 comments

I'm a hobbyist programmer, I run two sites, one for small personal projects, and another site that finds deals on Pokemon cards which I make money on from affiliate links. I just wanted to share this experience I just had.

All the links on my Pokemon deal finder are to eBay listings for Pokemon cards, and I had the idea of adding links to TCGPlayer for each card.

The problem is that the page for each card on TCGPlayer is based on their own product ID for each card. I found a website which had CSV data from TCGPlayer with all the cards and their product IDs. The next problem was that TCGPlayer used slightly different naming conventions for cards than Pricecharting (the site I was using to get my card values).

I wrote a script to try to match the TCGPlayer names to the Pricecharting names, but the script only covered about 40% of the cards, and there were too many edge cases for me to add. So I removed the TCGPlayer links from my site.

Today I was looking at another Pokemon card affiliate site, and they had such a simple solution that I can't believe I missed it.

For their TCGPlayer links for each card, they just linked to the TCGPlayer search page for "[set name] [card name]". That's it. No messy matching, no product IDs, just a link to a page that has exactly what the user needs in a line of code.

I've been programming for a few years now, and this has really made me realise that I should look for simpler solutions to problems where possible instead of trying a complicated solution that doesn't even work.

Just thought I'd share because I'm sure this is something other people have experienced as well.

[-]

Acceptable_Pen1696@reddit

Oh man, I feel this so hard. I do similar thing all the time in my work projects where I'll spend weeks trying to build some complex matching algorithm when there's usually just a search endpoint sitting right there

Few months ago I was building feature for one of my indie games and spent forever trying to calculate exact collision boundaries for irregular shapes. Then I saw someone else just use simple radius checks and it worked perfectly for their needs. Sometimes the "dumb" solution is actually the smart one

The worst part is when you finally see the simple solution, it seems so obvious that you wonder how you missed it in first place. But I think it happens because when we're deep in problem-solving mode, our brain just defaults to "this needs to be complex" instead of stepping back and asking "what's the simplest thing that could work"

Your Pokemon card solution is brilliant though - users probably don't even care if they land on search page instead of exact product page as long as they find what they need

doshka@reddit

when you finally see the simple solution, it seems so obvious

A good idea is something that anyone could have thought of, but nobody has.

Statcat2017@reddit

Sometimes the solution is just outside the parameters you thought it would be in

Ok_Chemistry_6387@reddit

It is really hard to see other solutions when you have found one. I always at least try to build the prototype quickly. Then use that to think of another solution now that my brain has had some time to think about things. It is amazing how often a simple answer jumps out.

SheepherderSavings17@reddit

Reminds me of the famous XY Problem

Dry_Button_3552@reddit

But really you're solving two different, but related, problems. Your solution would pull specific details about the card and simplify the interface for the user. The search page link puts the burden of matching the cards on the user.

The second solution might get well be good enough, but the first solution is still the "better"solution.

The question is: does the time and trouble you spend solving for solution 1 result in enough gain over solution 2 that it makes sense to pursue?

The real lesson here is you learned to assess your requirements. It's not necessarily that the solution is simpler, you just didn't have a very clearly defined requirement for what this feature needed to accomplish.

inspectorG4dget@reddit

Hah! Classic story. Don't let it get you down. I was stuck on some really tough edge-case problems for my grad thesis. My dad (not at all technical) solved most of them with common sense and logic. I was kicking myself for not having thought of "forget the Wi-Fi network"

patternrelay@reddit

This happens a lot. You optimized for perfect mapping when the system only needed "good enough" resolution. It’s a classic case of overfitting to data inconsistencies instead of stepping back and redefining the requirement.

Imaginary_Camel_4187@reddit

This is one of the best lessons to learn early on. Whenever you run into a problem, the tendency is to dig further down the rabbit hole rather than to ask whether the problem itself is the right problem. Experienced programmers do more thinking upfront about whether there is a simpler way to state the problem before writing any lines of code. Your script that matched the string wasn’t bad, it helped you see the simpler problem once you had a chance to look at it.

TseehnMarhn@reddit

I design and build many different things, from programs to machines.

One axiom that has held true, no matter the subject: if your solution is getting complicated, stop and reevaluate before going any further. Just about every design I've had that got lost in the weeds needed more time to cook.

The counter to this comes from Patton: A good plan violently executed now is better than a perfect plan executed next week. Make it work now, you can make it good later.

Which is exactly what you did, so don't feel stupid. This is how engineering works.