Good Software Takes 10 Years to Make (2008)
joelonsoftware.com
joelonsoftware.com
Mistake number 1. The Get Big Fast syndrome.
Mistake number 2. the Overhype syndrome.
Mistake number 3. Believing in Internet Time.
Mistake number 4. Running out of upgrade revenues when your software is done.
Mistake number 5. The “We’ll Ship It When It’s Ready” syndrome.
Mistake number 6. Too-frequent upgrades (a.k.a. the Corel Syndrome).
The sum of points doesn't support that it takes 10 years to make good software, but rather these pitfalls may slow you down. Did the Macintosh have good software? How about Windows 3.0? Of course, these were derivative. VisiCalc? Maybe too early to be representative. NeXTSTEP, iPhone, Google Chrome, YouTube (even before Google bought it)?
If we're talking about VC-funded software, then that certainly has a list of problems akin to this one. It would be great if VC's were like well-informed penny-stock investors investing in many companies giving the most chances rather only taking big swings. Is there an investment pool like this, that knows what they're doing, or at least earnest tries to do this?
I would link most software today as the Starbucks or McDonalds of food. Low-ish implementation quality, but reaching as many consumers as possible at the lowest possible cost and time to market and maximum profit. There's few companies that are in the market for 3 star Michelin style software.
I’ve been in the industry for almost 20 years, and been part of startups and worked at FAANGs, but I’ve never experienced a company have that mindset.
My experience has been that individual engineers are the ones that have that mindset and if two or more coincidentally get to work on the same project, the strive for excellence flourishes across the team and sometimes beyond. But it’s a very fragile ecosystem that can be disrupted by an eager management.