What's the gnarliest challenge you've tackled while developing or launching your product?
Artur Monroy
12 replies
How did you crush it?
Replies
Igor Lysenko@ixord
IXORD
I think this is a solution with advertising and personal branding.
Share
Scaling from 100 to 10k users in a week almost nuked our servers, but we optimized just in time.
I didn't really think it would be that time intensive, but having to put together a 45 different color palettes and templates designed for each was much more of a task than I had imagined. Of course I could've done it quicker, but the part of me that refuses to sell something I don't see as quality wouldn't allow that.
It took me about a week. I would say that I focused on the system for getting it done first, as opposed to just working on each one. The system allowed me to complete it much quicker.
Back SEO Marketing Software
@bradley_w We've all had that one set of problems we create a system for, where we spend so much time creating the system, that just doing it without the system would have been much more practical.
@bradley_w Wow! It's evident that you put a lot of effort and dedication into your work. It's commendable that you prioritize quality over speed. I'm sure those 45 color palettes and templates are amazing.
@backseosoftware this is true. I believe much of it depends on the person. A system-based approach allows me to "organize" my thoughts and the overall concept much more clearly, so that I'm able to navigate forward in a way that is more optimal. At least, that's what I think currently.
Back SEO Marketing Software
I guess the hardest thing to create in my product was the Theme Editor.
Problem 1: there are 56 theme components, and I wanted ALL of them to be changeable in real time, saveable to an INI and loadable through an INI.
Problem 2: The only documentation that I found on those items was through each item having a math equation stored inside of a list of floats. So I had to copy all of them, parse them through the initial function that ran the equations, then save those to the ini, in non-math equation float values. I might've been tired when writing this section, but it was a lot harder than it sounds (dang ini wouldn't fully save, had to deal with so many crashes, ini deletes because I missed a decimal somewhere. It was a mess, but eventually figured it out,. Then had to redo the algorithm for non-math values again lol.
But, then loading & saving worked! But none of this would matter if I didn't solve the next problem:
Problem 3: Only the ini would get loaded, none of the real colors would change, since I was running the colors as a dict, rather than GUI objects. So that means, I need to develop formatted strings, and run them through a function call from string. Not that hard for save/load. But what good is save and load without
Problem 4: Updating the values in real time
This is a lot harder than it seems. Normally binding colors isn't hard, but that's not in real time while also storing it into multiple variables at the same time without the user noticing a delay. There was no documentation of this anywhere I could find, so I had to do a lot of guess work. There was a proof of concept done, but no code to be found of it. To be honest, I was close to giving up here. So much progress made, but if this didn't work, it would all be for nothing. Took a break from testing, decided to do some research in GitHub of other libraries and how they handled their themes (lots of default libraries/a few items changed), then I found something, a piece of inspiration. It wasn't my exact solution, but it did kinda work... Until you gotta do it in bulk for every component, then it kinda breaks and goes slow.
Problem 5: Putting it in one line
For some reason, having the function didn't work. So I had to put it in one line. This wasn't too difficult, but it was something I hadn't done in a long time and was in a completely different language. So figuring that out took quite a bit, then adding the theme plugin into the real source was another issue, but not too difficult.
Anyways, that was probably one of my roughest, yet most rewarding, challenges in this project. Took about 15 hours of coding nonstop to figure it out, but felt good when I did.
@backseosoftware That sounds like a real odyssey but also a masterclass in problem-solving. The Theme Editor appears to be a feat of engineering in itself. Do you have any advice for others who might be facing similar technical challenges?
Back SEO Marketing Software
@mammothmango Keep searching, keep learning. You're much closer to the solution than you think, and chances are that someone else has solved your problem before, so make good use of GitHub