5 ms·

In almost 30 years of software engineering, I came to the conclusion years ago that most deadlines were entirely arbitrary. The business isn't going to fail if

In almost 30 years of software engineering, I came to the conclusion years ago that most deadlines were entirely arbitrary. The business isn't going to fail if you slip a week, or frequently even six months and if it does its probably not the fault of engineering unless its so dysfunctional as to repeatably blow through its own delivery estimates. Sure there are regulatory deadlines, customer POC or delivery deadlines. But, truly hard deadlines won't pop up as "we need this done next week" unless there is something fundamentally wrong with the business/sales side. A good functioning engineering/business unit will know far in advance of those kinds of deadlines and it becomes a question of whether the existing engineering org is capable of delivering the functionality within a heavily padded schedule. Put another way, I don't think i've ever seen a customer walk from a product because it wasn't released on a cadence, in fact many aren't actually happy about upgrades unless they are fixing a problem they have requested. And sales/POC cycles are such that frequently the customer may not actually have an alternative solution, or if they do, the chances they will switch to your product simply because you have X this year instead of next year isn't something that hinges on small factors that can be done by the engineering team in short cycles. You win because the product is better/etc and that doesn't happen overnight. So, stressing and working endless 80 hour weeks likely is going to have the opposite effect and creates a low quality product that people are trying to get rid of because it keeps breaking.

makk 2y ago
> most deadlines were entirely arbitrary Yes. As are most tasks, in my experience. Looking back on your almost 30 years, what percent of tasks could have been skipped entirely and, in the long run, it wouldn't have mattered? For me, it's easily >50% ... catch me on a grumpy day and I'd say it's closer to 90%.
- cmrdporcupine 2y ago
  Honestly, most companies and projects fail completely. In my 25ish year career I can only think of a handful of lines of code that I've written that are still in production (or that I think might still be in production) and only a few employers that still exist. When you pan out far enough it all seems a bit silly.
  - teaearlgraycold 2y ago
    Depends where you work. Code I wrote for the government may live on for far too long…
  - jahewson 2y ago
    But for a beautiful moment in time we created a lot of value for shareholders.
    - kunley 2y ago
      Pleasing shareholders is overrated
  - strken 2y ago
    I think this is confusing code with output. If you work for a company that builds medical software, then your output is something like quality-adjusted life years for patients and has nothing to do with the code you incidentally use to get there. Taking the view that none of the software you wrote is still running is a bit like an automotive engineer who thinks none of the car models she worked on at the start of her career are still in production. Well, no, they're not: but does that matter? Millions of kids got driven to school by cars she helped design.
    - cmrdporcupine 2y ago
      Even if you measure that way... again... most software industry businesses fail/disappear from the market. Outside of the largest companies, their business solutions tend to be transient.
      - strken 2y ago
        I agree, but that's the human condition, not some software industry peculiarity. As per Shelley: Look on my Works, ye Mighty, and despair! Nothing beside remains. Round the decay Of that colossal Wreck, boundless and bare The lone and level sands stretch far away.
  - 2y ago
- gavmor 2y ago
  Sure, but that inertia carries you over the myriad alignment traps strewn like potholes over the face of your production–possibility frontier.
  - MichaelZuo 2y ago
    If anything that's overstating the prospects of the median person in software engineering. Probably even the 99th percentile HN user couldn't do anything lasting and noteworthy if they somehow were blocked from relying on inertia, sheer dumb luck, obscure chains of fortuitous happenstance, etc... e.g. probably no one reading this can even hold the full Bluetooth spec in their head and understand all the tricks and pitfalls and obscure corner cases. Let alone do any useful work based on that. For the vast vast majority, progress happens in the form of millions of monkeys bashing typewriters in the hopes of getting a gem.
nox101 2y ago
It depends on the industry. Video games are still sold at retails stores (Target, Best Buy, ...). Those stores have to coordinate what is going to be on their shelves months in advance (6+)? So you promise your game will be ready and in a package and at their loading dock by November 15th. If you miss the deadline their shelf is empty of product since its spot was reserved for you. Other customer electronics have similar issues. Promises are made to coordinate. They're generally made before the product is actually finished. Now a deadline exists. It's not trivial to change it. The same is true for ads or was (TV ads in particular). They have to be scheduled well in advance. You pay for ads to run 6 months in advance because if you don't buy them in advance all the time slots will be sold out, no ads for you. But now you've added a deadline. If your product is not ready to buy when the ads run you've just wasted a bunch of money. I agree that for many standard IT projects, deadlines can be kind of BS but I can also see why they tend to exist in retail products. A solution could be, don't even start negotiating for shelf space before the product is 100% finished. Unfortunately few if any companies seem to be able to do this. For one, they need the money from the orders to make payroll. For another, if they waited until the product is done, then order ads and shelf space which won't be available for 6 months, what do they're employees do during those 6 months? They have no customers to fix bugs for yet. If the switch off the product to the next product there will be a big cost to pay to try to switch back the previous product they stopped working on 6 months previously when it actually ships and they need to actually support it.
- earthling8118 2y ago
  You dismiss that these are arbitrary, but I'm not seeing a whole lot in your examples that aren't also arbitrary. Sure, this is how a lot, if not all, companies work. That does nothing to disprove that this is arbitrary at the cost of mental health.
  - elevatedastalt 2y ago
    What is your point? We are not isolates who just subsist by ourselves. The modern world is a massively complex interconnected and interdependent system and we rely on the glue of capitalism to ensure that everyone still looks out for themselves but makes it all work for the system. Were the massive chip shortages causing a massive reduction in car availability and increase in prices "arbitrary"? Was the Suez canal blockage causing weeks of shipping delays "arbitrary"? If companies promise to pay for a stall at the next expo, but then don't actually make it because they don't ship their products in time, and now the expo loses revenue because they can't fill the stall at the last moment, is that also "arbitrary"?
makeitdouble 2y ago
> I came to the conclusion years ago that most deadlines were entirely arbitrary. To add to this, many manager/leaders assume some kind of deadline is needed for every tasks. They themselves manage their schedule by timeboxing and setting arbitrary due dates, feel extremely productive doing so, and expand that insight to the whole business. The base assumption is a task won't be done in a timely manner if there isn't a threat with a time limit on it. On a human level, I also understand the appeal of having a set date for something to be done: there's a warm and fuzzy feeling of a well oiled machine churning stuff on a predictable schedule. But it's just cruel to subject so many people to a forced schedule just to manage someone's anxiety and scheduling OCD. Reality is complex, longing for making it fit into neet boxes is delusional and unproductive.
- redrove 2y ago
  As an engineer I agree, but I’ve had the misfortune of managing poorly performing teams composed of unfit individuals that I couldn’t get rid of and they would do absolutely nothing if I didn’t arbitrarily give them some deadlines at least now and then. It goes both ways.
  - makeitdouble 2y ago
    I hear you. I see those more as checkpoints than deadlines though. Even with highly self managing people, it's always safer to have some idea of how much time it could take, and check for some kind of status midway to know if it's too different from the rough estimate (sometimes you learn they didn't touch the task for various reasons)
    - JohnFen 2y ago
      A checkpoint is a flavor of deadline.
      - makeitdouble 2y ago
        It's more of a camouflage I think: the theatric "management" role is fulfilled by having a random event on the calendar, but actual work doesn't have to abide to any set date and will take whatever time it needs to finish.
iamflimflam1 2y ago
Businesses aren’t that bothered about deadlines - provided they have sufficient notice that a deadline is going to be missed. What frustrates business is when there is no warning and engineering suddenly announces out of the blue that an imminent deadline can’t be hit. Or even worse, the constant drip of, it’s almost done - just one more sprint… The knee jerk current reaction of “we’re just not going to do estimates anymore” and “it’s done when it’s done” really don’t help in any way.
Tainnor 2y ago
I wonder where everyone is working, because in 11 years and several employers I've rarely had deadlines, if any, and if so, they usually came from customers and were announced in advance. Then again, I also never worked a 80 hour week. Maybe that's a US thing (I'm in Europe)?
- JohnFen 2y ago
  I'm in the US and the only time I've ever worked 80 hours/wk is when I was working for my own companies. That said, I've had deadline in literally every place I've ever worked (including my own companies). Even if they're artificial, deadlines are important. Without them, procrastination can all too easily take root.
cchi_co 2y ago
[dead]
tomcar288 2y ago
Yup and as a user of things like virtualbox and docker and numerous other softwares I really hate updates and so called "upgrades". I'd rather have a stable version that gets updated far less frequently.
- JohnFen 2y ago
  I couldn't agree more. I'm old enough to remember when software updates were something people anticipated and enjoyed. Now, they're a burden to be avoided or procrastinated.
jalapenos 2y ago
All deadline-based management does is turn the code to mush, so that tasks in 3 years time from now take twice as long as they would have, with even less delivery predicability (which of course they'll respond to by imposing deadlines more aggressively, and so on goes the downward spiral). It's a marathon, not a sprint.
- Viliam1234 2y ago
  > It's a marathon, not a sprint. Unless you do "Agile", in which case it is an endless sequence of sprints.

makk 2y ago

> most deadlines were entirely arbitrary Yes. As are most tasks, in my experience. Looking back on your almost 30 years, what percent of tasks could have been skipped entirely and, in the long run, it wouldn't have mattered? For me, it's easily >50% ... catch me on a grumpy day and I'd say it's closer to 90%.

cmrdporcupine 2y ago
Honestly, most companies and projects fail completely. In my 25ish year career I can only think of a handful of lines of code that I've written that are still in production (or that I think might still be in production) and only a few employers that still exist. When you pan out far enough it all seems a bit silly.
- teaearlgraycold 2y ago
  Depends where you work. Code I wrote for the government may live on for far too long…
- jahewson 2y ago
  But for a beautiful moment in time we created a lot of value for shareholders.
  - kunley 2y ago
    Pleasing shareholders is overrated
- strken 2y ago
  I think this is confusing code with output. If you work for a company that builds medical software, then your output is something like quality-adjusted life years for patients and has nothing to do with the code you incidentally use to get there. Taking the view that none of the software you wrote is still running is a bit like an automotive engineer who thinks none of the car models she worked on at the start of her career are still in production. Well, no, they're not: but does that matter? Millions of kids got driven to school by cars she helped design.
  - cmrdporcupine 2y ago
    Even if you measure that way... again... most software industry businesses fail/disappear from the market. Outside of the largest companies, their business solutions tend to be transient.
    - strken 2y ago
      I agree, but that's the human condition, not some software industry peculiarity. As per Shelley: Look on my Works, ye Mighty, and despair! Nothing beside remains. Round the decay Of that colossal Wreck, boundless and bare The lone and level sands stretch far away.
- 2y ago
gavmor 2y ago
Sure, but that inertia carries you over the myriad alignment traps strewn like potholes over the face of your production–possibility frontier.
- MichaelZuo 2y ago
  If anything that's overstating the prospects of the median person in software engineering. Probably even the 99th percentile HN user couldn't do anything lasting and noteworthy if they somehow were blocked from relying on inertia, sheer dumb luck, obscure chains of fortuitous happenstance, etc... e.g. probably no one reading this can even hold the full Bluetooth spec in their head and understand all the tricks and pitfalls and obscure corner cases. Let alone do any useful work based on that. For the vast vast majority, progress happens in the form of millions of monkeys bashing typewriters in the hopes of getting a gem.

nox101 2y ago

It depends on the industry. Video games are still sold at retails stores (Target, Best Buy, ...). Those stores have to coordinate what is going to be on their shelves months in advance (6+)? So you promise your game will be ready and in a package and at their loading dock by November 15th. If you miss the deadline their shelf is empty of product since its spot was reserved for you. Other customer electronics have similar issues. Promises are made to coordinate. They're generally made before the product is actually finished. Now a deadline exists. It's not trivial to change it. The same is true for ads or was (TV ads in particular). They have to be scheduled well in advance. You pay for ads to run 6 months in advance because if you don't buy them in advance all the time slots will be sold out, no ads for you. But now you've added a deadline. If your product is not ready to buy when the ads run you've just wasted a bunch of money. I agree that for many standard IT projects, deadlines can be kind of BS but I can also see why they tend to exist in retail products. A solution could be, don't even start negotiating for shelf space before the product is 100% finished. Unfortunately few if any companies seem to be able to do this. For one, they need the money from the orders to make payroll. For another, if they waited until the product is done, then order ads and shelf space which won't be available for 6 months, what do they're employees do during those 6 months? They have no customers to fix bugs for yet. If the switch off the product to the next product there will be a big cost to pay to try to switch back the previous product they stopped working on 6 months previously when it actually ships and they need to actually support it.

earthling8118 2y ago
You dismiss that these are arbitrary, but I'm not seeing a whole lot in your examples that aren't also arbitrary. Sure, this is how a lot, if not all, companies work. That does nothing to disprove that this is arbitrary at the cost of mental health.
- elevatedastalt 2y ago
  What is your point? We are not isolates who just subsist by ourselves. The modern world is a massively complex interconnected and interdependent system and we rely on the glue of capitalism to ensure that everyone still looks out for themselves but makes it all work for the system. Were the massive chip shortages causing a massive reduction in car availability and increase in prices "arbitrary"? Was the Suez canal blockage causing weeks of shipping delays "arbitrary"? If companies promise to pay for a stall at the next expo, but then don't actually make it because they don't ship their products in time, and now the expo loses revenue because they can't fill the stall at the last moment, is that also "arbitrary"?

makeitdouble 2y ago

> I came to the conclusion years ago that most deadlines were entirely arbitrary. To add to this, many manager/leaders assume some kind of deadline is needed for every tasks. They themselves manage their schedule by timeboxing and setting arbitrary due dates, feel extremely productive doing so, and expand that insight to the whole business. The base assumption is a task won't be done in a timely manner if there isn't a threat with a time limit on it. On a human level, I also understand the appeal of having a set date for something to be done: there's a warm and fuzzy feeling of a well oiled machine churning stuff on a predictable schedule. But it's just cruel to subject so many people to a forced schedule just to manage someone's anxiety and scheduling OCD. Reality is complex, longing for making it fit into neet boxes is delusional and unproductive.

redrove 2y ago
As an engineer I agree, but I’ve had the misfortune of managing poorly performing teams composed of unfit individuals that I couldn’t get rid of and they would do absolutely nothing if I didn’t arbitrarily give them some deadlines at least now and then. It goes both ways.
- makeitdouble 2y ago
  I hear you. I see those more as checkpoints than deadlines though. Even with highly self managing people, it's always safer to have some idea of how much time it could take, and check for some kind of status midway to know if it's too different from the rough estimate (sometimes you learn they didn't touch the task for various reasons)
  - JohnFen 2y ago
    A checkpoint is a flavor of deadline.
    - makeitdouble 2y ago
      It's more of a camouflage I think: the theatric "management" role is fulfilled by having a random event on the calendar, but actual work doesn't have to abide to any set date and will take whatever time it needs to finish.

iamflimflam1 2y ago

Businesses aren’t that bothered about deadlines - provided they have sufficient notice that a deadline is going to be missed. What frustrates business is when there is no warning and engineering suddenly announces out of the blue that an imminent deadline can’t be hit. Or even worse, the constant drip of, it’s almost done - just one more sprint… The knee jerk current reaction of “we’re just not going to do estimates anymore” and “it’s done when it’s done” really don’t help in any way.

Tainnor 2y ago

I wonder where everyone is working, because in 11 years and several employers I've rarely had deadlines, if any, and if so, they usually came from customers and were announced in advance. Then again, I also never worked a 80 hour week. Maybe that's a US thing (I'm in Europe)?

JohnFen 2y ago
I'm in the US and the only time I've ever worked 80 hours/wk is when I was working for my own companies. That said, I've had deadline in literally every place I've ever worked (including my own companies). Even if they're artificial, deadlines are important. Without them, procrastination can all too easily take root.

cchi_co 2y ago

[dead]

tomcar288 2y ago

Yup and as a user of things like virtualbox and docker and numerous other softwares I really hate updates and so called "upgrades". I'd rather have a stable version that gets updated far less frequently.

JohnFen 2y ago
I couldn't agree more. I'm old enough to remember when software updates were something people anticipated and enjoyed. Now, they're a burden to be avoided or procrastinated.

jalapenos 2y ago

All deadline-based management does is turn the code to mush, so that tasks in 3 years time from now take twice as long as they would have, with even less delivery predicability (which of course they'll respond to by imposing deadlines more aggressively, and so on goes the downward spiral). It's a marathon, not a sprint.

Viliam1234 2y ago
> It's a marathon, not a sprint. Unless you do "Agile", in which case it is an endless sequence of sprints.