A programmer who works 55 hours a week will produce twice as much code - but for how long? Once fatigue settles in (a few weeks in), you're back at the 40-hour productivity levels, just paying it at 55-hour rates (and organizational overhead for overtime, oooh boy!), plus burning out the developers (and organizational overhead for employee churn).
Your assumption is that 55 hour weeks are sustainable: just change this number in the spreadsheet from 40 to 55, and voila! And while we're at it, why not 80? Why not 200? Very PHB-ish, in my unhumble opinion.
200 hour weeks would be quite unsustainable, indeed. Forget not sleeping, where do you get the other 32 hours?
This is most true for tasks that are more intellectually demanding, like the development of large-scale software (which is what Brooks was writing about.) The more routine a task is, the easier it is to divide it up (though that is rarely to the benefit of the worker.)
By my experience I would say around 18-24 months.
I've tried the steady working, but I just do not operate that way. When there is work to do I do it, and simply have a hard time sitting still or concentrating on something else.
I'm fortunate to be in a field and with a company that is flexible in this way. I also really like this method. When I'm on vacation that is my 100% focus. I'd much rather work a lot and then completely unplugged instead of always working even on vacation.
If, on the other hand, you are only interested in the product, and/or you have other stuff to take care of, that would be Hell on Earth for you.
if i want to shortcut communication i simply call mates via dialthrough or skype
The issue I had was with this as the answer to life, universe and everything: "thou shalt not have life outside of work."
There's a reason why 80-90% of the wealthy/developed countries make it illegal to overwork people (even if the worker would be willing to work)
Rather you're back at far worse than 40-hour levels. Once fatigue settles in the code will be much worse.
Study-after-study has shown that knowledge workers are useless after 40 hours a week, yet every body seems to think they are the exception. I'll tell you why: Because we all know what it's like to work more than 40 hours a week on something we truly love and are passionate about, and we've sustained insane levels of work for a long time on such things.
The problem? Your bog-standard social-media/consumer app is not something your developers are going to be passionate about. They may find the odd project here or there in your company that they truly care about, but for the vast majority of applications your dream/company is work for them that happens to align with their skill set, it's not something they'd prefer to be doing.
I know there are exceptions to this rule, and I've even gotten a chance to land a dream job once. Do you know what a dream job turns into pretty quickly? Something you've mastered, and find boring.
Company runners/founders need to get over the idea that their employees should be as passionate about their company as they are, it's such a BS idea. Your workers, should be loyal, do what it takes to keep your dream up-and-running, and coming up with original ideas to make it great, but they should not be sacrificing their lives for your dream.
All-in-all I think you can still design a great product with two people in less than 40 hours. Hell...I think you can do it in 25.
Which team would you bet on? The team of great devs at 32hrs a week, or the team of decent devs at 40 hrs a week? You hear about the 10x developer, but if a 1 in 10 developer is even just 1.3x they end up being a better deal at 32hrs a week than avg devs at 40 hrs a week.
[0] Think of an average developer as the avg developer you work with, and the great developer as the best developer you work with.
I think that usually in these discussions, the money saved (the 20%) is intended to be used on other things than development.
You can't ship a product that isn't finished, and programmers with unfinished projects can't start new ones.
Let other programmers fix the bugs and charge it to support after you release it. Send your main programmers onto the next project. That's how commercial software works.
That's how some commercial software works, but less and less. It's a dying model.
That strategy represents two bets: 1) You can shift a lot of the costs from before release to after (in the form of tech debt), and 2) people will pay you a lot up front before they figure out that your product is buggy and that you can't innovate any more.
That could work with packaged software where you can market the hell out of it and get a lot of people to buy it once. It doesn't make sense for internal software, though. And it mostly doesn't make economic sense for software as a service, where people pay you month by month. With a service, it's better to start small with a core market and then innovate continuously, so people always feel like it's getting better. You can't do effective continuous innovation with high levels of technical debt and your "main programmers" off on the next project. Unless your market has a very high barrier to entry, somebody will spot your weakness, jump in as a competitor, and quickly lap you.
We'll still be dealing with the bad habits from that era for years. But the era itself is over.
https://www.gov.uk/overtime-your-rights/compulsory-overtime
and after that the employer must obtain signed consent of the employee, otherwise it is illegal. So the original point of "shaming into OT" stands.
Overtime of any type for any type of worker is frowned upon by the authorities here in Norway but I suspect that the UK and US are a bit different.
Because most countries require you to pay overtime for salaried workers, including in the US from what I see here:
You might also find this interesting: http://www.independent.co.uk/news/world/europe/sweden-introd...
This was my summary of it when it was last posted:
Regardless of the outcome (productivity -> more / less / equal), I would like to see the same experiment in IT or even other so called white-collar positions. I am sure things would be different.
It makes sense when all your employees are singles without a family or any other obligations outside work. You might want to take into account that your employees who are not, may have to pick up their kids from kindergarten/school, etc. They may not be able to work long hours without letting down their children or damaging their marriage.
You could make crunch optional, but there is some peer pressure ('Look, Johnny is not committed to our project').
These are waters to thread carefully if you want to have healthy/happy employees. Make clear rules ahead of times of what is expected, pay extra compensation for overtime.
(Sorry if I set up a straw man.)
A huge cost that is often misunderstood is the cost of replacing an employee who leaves because the working conditions (endless hours, regular extra time, bully environment) are not compatible with his/her work/life balance.
My previous workplace was founded upon extra time (I remember our CTO and CEO clearly saying that he expected people to work 50-60 hours per week. The CEO literally said multiple times "these people are young, I expect them to work 120%").
We constantly were fixing issues and problems, we never had time to properly plan and design. Once the CTO told me that he expected our SW to have problems, but since we had no time to fix them or to build better software, he also expected the programmers to be available so fixing production issues when they happened no matter if it was night or weekend. Once he told us that he expected senior developers to check email once a day while on holiday to check if there were no problems.
No wonder that in a few month 3 out of 4 of the senior engineers left the company.
They also say that overtime can make sense if its short and a rare occurrence.
Also offering help with child-care makes things easier for employees with families. Those are solvable problems if crunch-time is limited to short stints and if both sides are willing to work together.
I see unpaid overtime as one of the primary drivers of extreme crunch abuse. It aligns incentives in the worst possible way: More work gets done per week (though only marginally at best), project management gets to claim "we're doing all we can" and then, on top it's actually free time donated by the employees at no immediately obvious costs. Not surprising that this sounds attractive.
Nice! I'll remember that one.
Every tool has a fixed number of cycles before something gives and repairs need to be made (or the tool written off and a replacement obtained).
Depending on the price and availability of a tool, some businesses make conscious decisions to limit the number of hours they use a tool, so as to extend the life of the tool.
Now if we are to consider that we are the said tools, and once we wear out, the onus is on us to repair ourselves and bear all costs of doing so (because the business "writes us off"), we might have more clarity into pricing ourselves?
If we want to add to that, that we are actually human beings, some more clarity perhaps could be obtained.
I invite your opinions and feedback.
I agree however that adding more developers can slow things down but this is partly a result of how software is designed and work divided up.
That really depends on the architecture of the software. There are thousands of people creating and maintaining device drivers for the linux kernel. The idea that only two people could do that simultaneously, is a bit far fetched when the architecture actually makes sense.
But then again, most commercial software does not have that kind of extensible architecture. Therefore, it is probably true that a commercial software team will never really scale.
Here I will quote a HN post which quotes a secondary source which analyses mythical man month
"From a business point of view, long hours by programmers are a key to profitability. Suppose that a programmer needs to spend 25 hours per week keeping current with new technology, getting coordinated with other programmers, contributing to documentation and thought leadership pieces, and comprehending the structures of the systems being extended. Under this assumption, a programmer who works 55 hours per week will produce twice as much code as one who works 40 hours per week. In The Mythical Man-Month, the only great book ever written on software engineering, Fred Brooks concludes that no software product should be designed by more than two people. He argues that a program designed by more than two people might be more complete but it will never be easy to understand because it will not be as consistent as something designed by fewer people. This means that if you want to follow the best practices of the industry in terms of design and architecture, the only way to improve speed to market is to have the same people working longer hours. Finally there is the common sense notion that the smaller the team the less management overhead. A product is going to get out the door much faster if it is built by 4 people working 70-hour weeks (180 productive programmer-hours per week, after subtracting for 25 hours of coordination and structure comprehension time) than if by 12 people working 40-hour weeks (the same net of 180 hours per week). The 12-person team will inevitably require additional managers and all-day meetings to stay coordinated."
https://news.ycombinator.com/item?id=3547965