Code the right way… and your start-up will probably fail
If you’re in a start-up and you’re spending a lot of time coding – “the right way” – the way of cargo cults and W3C standards…
I’ll wager that your start-up will run out of money and that it will probably fail.
Check out my tech startup podcast TechZing. Maximise your downtime by listening to us while you code, commute, work-out or do the dishes!
Comments closed
Comments
Amen. Most programming books/resources ignore or skim over the time/budget constraints which are the main factor behind most things coded “the wrong way”. Deciding when and where to code the “wrong way” has been by far the biggest hurdle in my career, so far.
OH: “If it’s not tested, it probably doesn’t work”. It’s not the coding that matters, it’s what you are testing against. If your testing aligns with what your users are doing, it doesn’t matter if you are coding the “right way”. I’ve see sh**** coding that just “works” for users.
[…] This post was mentioned on Twitter by Justin Vincent and Jordan Keats, Hacker News. Hacker News said: Code the right way… and your start-up will probably fail: http://bit.ly/d0yDMX Comments: http://bit.ly/d25RnS […]
Hmm…to some degree it may be true actually. But normally “bad” code will hit you sooner or later, and that’ll most often hurt more than if you’d have done it the “right” way straight from the beginning.
Got any statistics to back this up? Failed startups you can point to as examples? Anecdotal evidence even?
In my experience you can launch a working app with sloppy coding, but you will eventually have to pay the technical debt that has accumulated. It’s usually when you start adding new features later and the codebase grows when the sloppy start will begin to get in your way and cause problems unless dealt with.
@Jani Hartikainen – You describe a much better scenario than zero technical debt and no cash in the bank because you ran out of money making it “the right way” 😉
This is probably very true for fledgling companies who dont grow rapidly. These companies can afford to do things the wrong way, with plenty of time to fix it in the future. ‘Shipping’ the product is more important for these companies.
But if you have a service that grows rapidly (read: if you are successful), then you better make sure things are coded right, otherwise you should expect to be hacked, lose credibility and experience plenty of downtime.
if you call “the right way” cargo cult, i can only agree with you, but that’s a very strange kind of right way….
I used to work for a startup that failed BECAUSE nothing was done right. Every prospective customer’s requirements took forever to implement and we were endlessly fixing bugs because the thing was put together with duct tape and bubble gum.
The question is, how bad is it. If you’ve got coders wringing their hands over if their solution is optimal, then yeah, you’re going to run out of time and money and never launch. Especially if you take the position that it needs to be perfect to launch.
If you have solid developers writing code they may put a hack or two in there that doesn’t accommodate all the edge cases and is less than elegant, but it isn’t too bad and the impact is minimal.
Individual features can be done right or wrong, but if your whole application is done “wrong” it will be tough to do anything with it and you’ll spend so much time fixing bugs you’ll never sell anything. It’s best to release quickly with a small, well tested feature set than trying to get every feature in the book in there and only half of it works.
As a long-time technologies and a developer for more than three decades, I must respectfully disagree. Most startups fail because they don’t really have anything other than a feel good idea, bad management, a tendency to over-promise and under-deliver. The technology is built to get there fast, but it typically an unreliable, non-performing, steaming pile that requires constant hacking. It’s the constant hacking, downtime, and lack of planning that blows through the funding. Most technical guys really don’t know how to run a business, market their product, etc. The “build a web app/site and they will come” fantasy is a really lousy business plan. For every Zuckerberg, DHH, etc., there are ten thousand wannabes that never get anywhere.
This is true. I can definitely vouch this from experience. Premature optimization is always a killer when you have only a few bucks in your pocket.
@Justin What I’m saying is that you’ll eventually have to deal with it – Not right away, not until it becomes a problem, but at some it will become one.