It has to do with the buyer's ability to determine what the code quality is. I think people are limited in their ability to quickly assess code quality. Myself, I spent two days reviewing a system to give it a clean bill of health, but six months later my boss said that it shouldn't have been written at all and I agreed.
I’m just thinking now, turning on a tool like code climate [0] could be useful as it scans the whole codebase and gives each part of the code a grade (eg. A,B,C).
Not a perfect approach but a useful guide to find obvious code smells/issues.
Yes. It does affect. If I, the buyer, determine that your code needs to signifcantly rewritten to fit my needs/systems, I will be willing to pay less money. Instead, I may want to build it myself.
99% of the time. Nope, having tons of paying customers and profit is what will matter. If the "shitty code" was good enough to build that business, it will be a good enough to keep it running. I have seen companies acquired for tens of millions and 2 hundred million offers and the code was pure shit. It was the customers/business that was important. Guess what, those acquisitions were successful and the buyer is happy! The acquiring development team wasn't happy of course and won't have signed off on buying because of the code base, but that's why they are engineering and not M&A!
Thought experiment, imagine you build a social network that grows to be as large as twitter or Instagram, or you come up with an amazing deep learning tech that is far ahead of anything in the industry. Do you think all the companies that will come knocking at your door will care about the code quality? In the former case, if you have 100 million DAU, folks want the users not the code. In the later, if you have amazing tech, no one cares about the code quality, they want the tech, they can hire pros to fix and improve the quality.
This answer is spot on. User base is always more important than code base. ROI is what business people care about and code quality doesn't matter as much as does it work as advertised. Also there are always those coders who'd work on building a 2.0 for said project at a decent rate because they need the paycheck.
7 comments
[ 1341 ms ] story [ 1439 ms ] threadIt has to do with the buyer's ability to determine what the code quality is. I think people are limited in their ability to quickly assess code quality. Myself, I spent two days reviewing a system to give it a clean bill of health, but six months later my boss said that it shouldn't have been written at all and I agreed.
Not a perfect approach but a useful guide to find obvious code smells/issues.
[0] https://codeclimate.com/
Thought experiment, imagine you build a social network that grows to be as large as twitter or Instagram, or you come up with an amazing deep learning tech that is far ahead of anything in the industry. Do you think all the companies that will come knocking at your door will care about the code quality? In the former case, if you have 100 million DAU, folks want the users not the code. In the later, if you have amazing tech, no one cares about the code quality, they want the tech, they can hire pros to fix and improve the quality.