178 comments

[ 4.4 ms ] story [ 219 ms ] thread
I let my boss watch me work for a day and I’ve never been more productive
I'm excited to try this. I do my best thinking alone but work best with someone watching. I know some people can't perform even basic tasks with someone watching, but for me, it's easier and sometimes even necessary. If I'm stuck coding I'll imagine explaining it to someone (or ask someone to do it with me). If I can't nail a movement in the gym, I'll pretend to demo it.
That's interesting and I have no doubt it works. However, I think that you might simply get used to the pressure and start procrastinating either on the service itself or simply procrastinating using in the service.

The other problem is that any kind of outside force that pushes you to work harder also creates stress. This stress isn't entirely regulated by yourself and it's possible for it to backfire.

It’s about making it a bit easier to get into the groove. If you’re determined not to work, you’ll find ways around it. P
This is so insane it just may work. Countdown to camgirls ruining it is now ninety seconds.
That's cute. But did you consider hiring a stranger from craigslist to slap you everytime you access facebook?

https://news.ycombinator.com/item?id=4661940

Or perhaps testing websites... drunk? I love and loath these articles. They are the equivalent of "and you won't believe what this madlad did!!!!" of hackernews.

(comment deleted)
I thought you were kidding. You weren't!
(comment deleted)
We used to do this on LessWrong back in the day; people would write or study math in a video chat room. It was surprisingly effective.
+1 for surprisingly effective. Such a good word for my experience of this.
Anybody else get an immediately negative visceral reaction from this? If I'm understanding this correctly, the aim is to manipulate our sense of shame/guilt to boost productivity.

After sitting with the feeling for a bit, here are some ideas that come to mind:

1. Maybe we should ask why we feel shame/guilt in the first place. Is it "normal" to feel this? If it isn't we should not rely on it for our happiness (or productivity).

2. What is the value of productivity? Why does it make us happy?

For me, guilt/shame is something to be overcome, not used as a tool. And the value of productivity is something that has been handed down to us by a culture which we've been thrown into. And (at least for me), it's our duty to question these ideas instead of merely giving into them to self-reinforce themselves.

Do you feel shame/guilt when pair programming? Is the outcome still useful?
No, I'm not referring to pair programming. That is a dialogue between two people working to get something done (no shame involved).

I'm focussing more on the aspect of having someone looking over your shoulder. Just a pair of eyes to watch you work (which I don't equate with pair programming)

I personally think I would be more productive as a pair, in large part (but not entirely), for the same reason as somebody just watching me.
Yes, I'm not questioning whether one would be more productive (I think actually someone would be more productive!). It's the value of this productiveness in the context of manipulating shame. Maybe for you shame is not a factor here, and that's great. I can only know me, and for me, the act of having someone watch over my shoulder (and only watch over my shoulder) as I work would have some level of shame/guilt involved.
Right, me too. I'm just saying that the shame/guilt applies via your coding partner as well. Maybe more so, since they're a peer.
Yes, my immediate gut reaction to the title was to think, maybe next we can have an app to promote the creation of super workers that produce outputs far in excess of their quota, we could call them Shock Workers. Then I read it and some of the language leaned eerily in that direction.

OTOH I've seen similar things used in a different way which worked well, in places that have lots of ops centers to link them and make them feel like one - you can look up and see the other ops center with your colleagues there and chat etc. But that was less about applying a work/focus pressure and more about enabling communication and keeping the teams connected.

> For me, guilt/shame is something to be overcome, not used as a tool.

Wow, I think I quite strongly disagree.

What do you think it exists _for_ ?

It is good for society that people have the capacity for guilt and shame.

If we did not, I cannot see any other recourse other than fear!

Okay, so shame exists because it is good for a society that is built on shame? I don't mean to be derogatory in anyway, but the fact something is stabilizing doesn't mean it is good. I believe there are better ways to build relationships with others
I think OP is saying you should feel guilt and shame if you treat others badly, steal or break their stuff, murder or rape them etc.

You were probably thinking of completely different contexts.

>Okay, so shame exists because it is good for a society that is built on shame?

Yes. Shame exists because it gave some evolutionary benefits to the primates who had it, like being able to live in a society and cooperate and don't cheat/kill/fuck over each other as much (if you think we do bad stuff too much, wait till you see what we can do without shame).

>I believe there are better ways to build relationships with others

The problem is that made-up (e.g. of our own making, requiring us to think about them and follow them rationally), not instinctual, ways, are none effective at all compared to innate, evolutionary, feelings like shame...

Good points! I agree the feeling of shame is pre-reflective. At the same time, I think we have the possibility of comporting ourselves in a way-of-being that better copes with shame. For me, By reflectively using your pre-reflective feeling of shame as a tool for productivity feels wrong (and this is personal and I don't have the best words right now to expound that feeling)
>* For me, By reflectively using your pre-reflective feeling of shame as a tool for productivity feels wrong*

Well, with that I agree.

For one, it cheapens the quality and utility of shame.

Following this to the end, could end up with cheating on one's spouse or killing someone feeling only as bad as checking your Facebook page when you should be working...

> Okay, so shame exists because it is good for a society that is built on shame?

I think so, yes.

Evolutionarily, as humans, social inclusion was incredibly important. Being removed from a tribe was likely a death sentence. As such, we have some very strong social drivers like shame and embarrassment that people feel quite keenly.

But these social drivers caused people to do things that were good for society. You felt shame if you didn't contribute to hunting or gathering, cooking, planting and harvesting, building structures, raising children, or other crucial societal factors.

I think you're on pretty solid ground to say that shame and embarrassment go a long way toward the foundation of the society we have now.

I do think shame and embarrassment are useful tools that we shouldn't want to get rid of entirely. That being said, I think it's totally fair to question whether our current-society over-relies on these kinds of things, and asking if it would be healthier to scale them back rather than expand them.

Love, kindness, empathy, etc. You dont have to build your society as a goad to make people do what you want.
You’ve tapped on the fallacy of assuming it’s all about shame. Accomplishment is very rewarding to plenty of people.
Do you know of any example of societies that were based on love, kindness, empathy and so on, and were stable for a prolonged time?
Your argument that it exists _for_ anything is a bit hollow. What do superstitions exist _for_? What does religion exist _for_? What does the shame of being on the LGBTQ spectrum exist _for_? What does fat shaming exist _for_?

That something exists and that people are using it doesn't at all mean that the use is still valid, that it ever was valid, or that there are not much more fruitful alternatives.

Moreover your argument that you need guilt and shame to have a good society is the same sort of argument that Christians use against atheists and their lack of religion. The reason I don't go raise hell is simply because helping other people is what has been burned into my mind as a child. I don't hand a homeless person money because I'm worried about the shame of not doing it, nor do I feel any real happiness from when I do it; I do it because I believe it is my duty (which is the same reason I rarely miss deadlines).

> what has been burned into my mind as a child

How was this done?

With my father doing the complete opposite by being abusive as possible without physically touching me.
:( I’m sorry to hear that,

I don’t mean to suggest that shame/guilt aren’t sometimes (often?) overused.

I just think that attempting to entirely excise them is a mistake (and that they are at times useful in moderation for certain things).

It sounds like you are basing your life on negative emotions only. Just so you know, there is another universe out there, based on positive ones.
By “recourse” I meant “recourse for bad behavior”.

Society must have a way to discourage bad behavior, not only encourage good behavior.

It reeks of "scientific management", a scrupuleless way of exploiting the slaves^H^H^H^H^H^H employees.

Ruthless con named Frederick Taylor looks for a way to make himself a ton of money by "consulting" on improving worker productivity.

"How did Taylor arrive at forty-seven and a half tons for Bethlehem Steel? He chose twelve “large, powerful Hungarians,” observed them for an hour, and calculated that, at the rate they were working, they were loading twenty-four tons of pig iron per man per day. Then he handpicked ten men and dared them to load sixteen and a half tons as fast as they could. They managed to do it in fourteen minutes; this yields a rate of seventy-one tons per man per ten-hour day. Taylor inexplicably rounded up the number to seventy-five. To get to forty-seven and a half, he reduced seventy-five by about forty per cent, claiming that this represented a work-to-rest ratio of the “law of heavy laboring.” Workers who protested the new standards were fired. Only one—the best approximation of an actual Schmidt was a man named Henry Noll—loaded anything close to forty-seven and a half tons in a single day, a rate that was, in any case, not sustainable. After providing two years of consulting services, Taylor billed the company a hundred thousand dollars (which works out to something like two and a half million dollars today), and then he was fired."

Interesting! Shame and awareness seem to have always been pretty tightly linked[0]. I think it's normal.

[0] Genesis 3:7

What if it’s pride and not shame? One may feel like they get to show off how talented or productive they are.
It sounds like you're questioning the whole idea of accountability. The mechanism here is the same as if you have a friend check in on you to help you quit smoking or exercise more or eat better. It's true that you're adding pressure/shame/guilt into your life, but I don't think it's particularly sinister.

The way I see it, the mind is extremely complex, and the decisions you make in the moment may not be the decisions you'd like to make in life. In the moment, you might end up eating a tempting ice cream sandwich, or you might get distracted by Facebook when you meant to be working on a meaningful project. The sort of accountability from the article is an example of understanding your goals, emotions, and habits, and harnessing that understanding to better achieve what you really want. The pressure/shame/guilt here is a tool to be used with care, and if it negatively affects your life, then certainly you should stop or scale it back a bit.

I find happiness from productivity because I try my best to work on projects that I find meaningful (inside and outside of my job), and I find pride/meaning to be one of the most satisfying forms of happiness. People who don't find meaning in their work might still feel that productivity helps them achieve their goals by getting a raise or keeping a job, thus providing money to use for other goals (like happiness).

Thanks for the writeup. I think you're right: the idea of external validation / accountability is definitely involved here as well.

And I think it's easy to read what I wrote to mean that "shame is objectively bad," but that was not my intention. The intention was more to question (and possibly reevaluate) our own relationships to it.

> I find happiness from productivity because I try my best to work on projects that I find meaningful

In the end, it comes down to what makes you thrive, and only you can answer that. Productivity, in its most general sense, can be a way to achieve that. At the same time, for me, it is healthy to question these assumptions every once in a while.

>What is the value of productivity? Why does it make us happy? [...] And the value of productivity is something that has been handed down to us by a culture which we've been thrown into.

I think it's more that the business-speak label of "productivity" bothers you rather than its underlying idea of efficiency of input effort in relation to desirable output.

The concept of "productivity" doesn't have to be a Peter Drucker style management guru propaganda. Productivity makes us happy because it's an intrinsic human desire to improve our lives. For example, consider a prehistoric hunter in Africa that's running barefoot with a spear in hand and chasing after antelope to kill and eat it. He doesn't need modern McKinsey consultants to tell him he wants to do the least amount of running for the most amount of food. The better that ratio of expended effort to food quantity, the better the productivity. It's just that the hunter didn't label that concept as "productivity". If he has a sprained ankle, his chasing ability will decrease and his productivity will also decrease. He becomes unhappier. If the hunter uses his brain and notices the paths the antelopes use to the watering holes and takes advantage of those patterns to intelligently intercept it, his hunting productivity increases, and he becomes happier.

In the case of this thread's article, it looks like the author is a freelance journalist and so "productivity" to her is writing articles faster and/or writing more articles.

Good point. I agree the word "productivity" is bundled up with many meanings, and it could be a whole study in-itself to unpack it. Preliminarily, I'd say productivity is pragmatic towards-which one identifies oneself (a hunter hunts, a writer writes, etc).

I've been reading Being and Time recently, and Heidegger makes an interesting distinction between what is "ontic" (a writer writing, for example) and "ontological" (a writer investigating the state of being behind writing). In this context, productivity interestingly can be in both the following ideas:

Ontically, productivity writing a book. Ontologically, productivity is understanding of the way of being a writer comports oneself to be a writer.

I think both ideas are equally important. I'm not sure why I'm writing this; maybe to address the difficulty with pinning down a consensus of what productivity means. but happy saturday!

It really isn’t like that when using it. Culture matters a lot and the community is just focused on actively working in a collaborative environment. It works wonders. I call it “flow state on demand.”
I had some time between my graduation and my job and I streamed some software development on Twitch, I don't think I had that much fun and was never so productive when others were watching me. It's such a weird effect but when someone is watching you your whole behavior changes ... and for me, in a good way. I'm naturally kind of lazy so this was really awesome.
Sounds effective but weird. Maybe just go to a coffee shop.
If you did this everyday with a new stranger you'd stop caring eventually.
My girlfriend tried one of these recently. It mapped to her "Obliger" personality, per The Four Tendencies by Gretchen Rubin.

Me..? I'm a Questioner. But I'm happy for her to find whatever makes her self-employment day-to-day life more manageable and uplifting.

This seems to be one of the major drivers for some folks who do pair programming.
Tangentially related: is there any sort of dashboard service that would let me create a dashboard showing things like:

* email inbox count (differentiated by email inbox)

* unmoderated comment queue on website

* Facebook unread message count. Similar ones for other social media inboxes

* and maybe some custom metrics or ones triggered by a spreadsheet or todoist or something

I think this would be immensely useful for my own purposes, in seeing where I wasn’t on top of things. But also useful for sharing with someone else and being accountable. Would allow analysis too, such as eliminating a certain inbox entirely, or delegating some sections, etc

No, but that sounds like a great startup idea. Sign me up!
I looked into it since commenting. Klipfolio might be able to do this, but it wasn’t clear how to get a gmail inbox count.

Then I remembered I use Numerics to put monthly sales on my apple watch. I checked, and it can get email inbox counts at least! They have a ton of data customization. Will see what other services I can add.

Ios only for now unfortunately. May come to mac once Marzipan happens: https://cynapse.com/numerics/

Edit: doesn’t have messages beyond email unfortunately. Though there is perhaps a way to jerry rig something using zapier and one of their supported services such as Trello or Todoist. Klipfolio does support Slack.

Legends say that long ago outside of the bubble where people working constantly monitoring each others and there even was people who's position was literally supervisors.

This is called "office". Somehow absolute peak of my procrastination happened in those human filled dedicated work places with supervision, from bosses, peers and subordinates.

I was going to say. Go work in an office?
Eh, it’s less about monitoring and more about shared effort and focus. Take what works, leave the rest. For a lot of us, this is a life saving and wonderful service.
Now do that for a month. Or even a week. You may be more productive bc you don't want to be seen as slacking off with someone watching, but maintaining that kind of pace is untenable. Did you ever get stuck and check your email or Facebook just to give your brain a little break? Did you ever sit, unsure of your next move? Or did you do the first thing you thought of, even if it was wrong, so you didn't look like you were wasting time?

Intense productivity with someone looking over your shoulder is great for a day. After a week you'll want to quit. After a month you'll want to kill yourself.

That’s fine. Having a tool I can use a few hours a few week to get me through things that are the highest product of important and annoying is super valuable. Doesn’t have to be a lifestyle.
You won't be the one choosing when and for how long to use the tool...
Normal office environments already have managers that are plenty capable of watching and critiquing everything you do all the time, but at least in my experience, that's almost never how it goes. Good managers know that freedom and autonomy are empowering. I'm sure others have had much worse experiences, but I think it's mostly well-established that micromanagement is a bad thing.

Whether it's in-person or with a tool like this, if your employer is creating an unreasonable amount of pressure/stress, then that puts them in a weaker position in the job market than companies that treat their employees well. Give them feedback, and if they're not receptive to feedback, see if you can ask for a raise or find another job.

> Good managers know that freedom and autonomy are empowering. I'm sure others have had much worse experiences, but I think it's mostly well-established that micromanagement is a bad thing.

This is really important. If your manager or PM is constantly bothering you, you won’t be able to get into your flow state, which is important in software engineering.

that's what we use daily standups for. i announce my plan for the day, and the next day i report on how well it went. if i procrastinate to much i have nothing to report.

it doesn't impact my freedom and autonomy because i decide what to do every day, so i set my own goals and only report on how well i achieved them or explain why i didn't (run into a bug etc)

True and depending on how much your performance improves it may be freeing time for other stuff.

If you get ten times more stuff done than usual, it would be enough to do this one or two times a week.

That's one way to see it. Another way is "I've shown that I got some unleashed productivity potential. How can I make it sustainable rather than an exhausting sprint?". In other words "how to make it a habit", habits are easy to sustain

I'm not arguing for having a stranger watching you all the times or that you should only think about productivity, just that it shows that there's a margin for improvement that can't be dismissed

It's not sustainable. Your brain needs downtime, just like you can't run everywhere you go.
This is not black and white, many of us procrastinate in ways that are not always a healthy break and finding ways to curb some of that is not bad.
I don't think this approach necessarily helps with that distinction of good breaks and bad procrastination. I think it would be interesting to distinguish those, but I would look for another method if that were my goal. That also isn't to disregard the interest of the experiment above, just to say it doesn't help here.
but can't we just allocate those downtimes to the time when we are not on the clock?
I'm not so sure this is true.

Yes, being "on" takes energy, but it's counterbalanced by the buzz of productivity and working at your peak.

I'd expect to be more tired at the end of day, but also to feel more accomplished.

I'd imagine that after a week you'll start checking email (because that's still work, right?) and after a month you'll be on Facebook because really, who cares about some stranger knowing you're procrastinating?
Maybe. On the other hand, the human animal has a powerful capacity to decide that whatever it’s doing is the good and right and normal thing to be doing. After a month you’re just as likely to be completely acclimated as you are to be suicidal.

There’s also the element that when a stranger (or a subordinate) is watching you don’t just act busy — you act like you enjoy your job, you show off a bit. This is the kind of performance that soaks into the skin and becomes real.

I find that pair programming has the best of these benefits at the same time reducing the probability of feeling suicidal etc. Personal experential bias of course.
I don't believe this is the case at all. When I was in university, everyone I know (including myself) organized into study groups precisely to combat procrastination, and not for any intellectual benefits. It worked. At the end of a day of solo studying, you would feel horrible and ashamed because you got so little done. But at the end of a group study session, you actually feel fulfilled.
I don't believe studying with a group of people and letting stranger watch you what you are working.

Also, did your study group have stranger or at least you kind of a bit familiar with them?

Some were friends, but there were members I only knew by sight. Of course, by the time of the tests, they were friends.
this is not much different from pair programming. even pairing with a junior developer where i do most of the work myself makes me more productive.

pair-programming is also exhausting. after an 8-hour work day i am beat. however i can't think of any better way to work, and it it is practiced in several companies successfully.

if people would want to quit after a week or kill themselves after a month of pair programming we would see a lot of backslash against it.

>> if people would want to quit after a week or kill themselves after a month of pair programming we would see a lot of backslash against it.

That's been my experience. Even at the few places I've been were some do pair programming it's a very rare occurrence. Occasionally upper management will stump for it, likely because they (a) don't actually program all day and (b) see it as some sort of kale-like super habit. None of my IC have ever pushed hard for regular sessions.

We pair programmed more than 90% at my last job, and it was great. Together we would typically come up with architecture, design, tests and names which were better than either of us would have done on our own. It is hard work, but very rewarding, and it has many pitfalls and nuances which we tried to work out at our ~fortnightly review sessions.

More important than pair programming itself is the willingness to look at what works for the team and how you can improve the experience and productivity together. Most of the time we did traditional pairing, with two keyboards and mice on a dual-monitor PC, and simply signalled each other when we wanted to take over or cede control. Towards the end of the four years there we very successfully "mobbed" on some complex problems, and split up when we had worked out what to do next. We would typically work in different pairs every day, which resulted in everybody knowing the entire system to a similar degree — we didn't even need a handover when someone left, and onboarding basically meant that the more senior employee had to explain a bit more of the context while pairing with the new hire. I found it did wonders to my understanding of how other developers think and especially how to explain my thinking to someone else. But YMMV, and good luck!

The article is just wrong. Social pressure makes you less productive and hampers cognitive health while working.

I can’t believe anyone even feels this is worth discussing or disagreeing. From the mountains of evidence refuting the communication conditions of open plan offices, we know that knowledge workers require privacy to be productive.

The case is closed. This essay is arguing against the tide. Move on.

Pairing or close one-on-one supervision hasn't much in common with being in an open-plan workspace.
They are even more severe and intrusive forms of disruption than ambient noise and impromptu conversation of open plan offices, so they would exacerbate the negative effects already known.

Pairing in particular requires you to schedule time for it, host it in a private conference room for those two people, and have adequate video conferencing. If pairing happens at someone’s desk in an open plan setup, looking at the same laptop screen or monitor (instead of a projected version), then it universally falls apart, and is significantly disruptive to other around as well.

Once the novelty wears out and your subconscious processes that the stranger has no power over you I believe you will relapse into your old habits.
Can turn out to be quite a creepy experience.
I particularly enjoyed the follow-up article "I burnt out and lost my sanity/health/job/gone to opioids/alcoholism/etc after pressured to maintain this 'more productive' pace for months on end as if I'm a hamster on a wheel".
Yeah. I've experienced this working to project deadlines at several jobs. Like everything else, "crunch" and "pair programming" are good in small doses for getting productivity. I couldn't handle it long-term. I have seen some people thrive on it with no ill-effects, but I have never had an extremely-focused, very driven work personality.
My first month of pair programming was utterly exhausting, but once I got into the pace, it was the best thing I've ever done.

The mutual focus is definitely a big part. You're in a social context, and there is no place to stray from the mutual task, and that discipline is very liberating, weird as it may sound, for someone prone to procrastinating like myself.

There are some pre-requisites to Pair programming. That you need to have a well defined task, that you need not be bogged down by insane deadlines and can afford to have 2 people work on the same thing and both need to be well versed at the task at hand to live up to the promise of productivity. My experience pair programming for a year was sh*tty enough to never consider it again unless I have these pre-reqs met. I'd much rather split tasks, work independently and then touch base often.
> you need to have a well defined task, that you need not be bogged down by insane deadlines

Aren't these pre-requisites to all work? How does working alone solve them?

The thing that's often overlooked is that pair programming is a skill, and you need to learn it to do it well. Just putting two novice pair programmers at a computer and asking them to "just do it" will usually result in a lot of frustration.

I was fortunate to be pairing with a lot of seasoned pros in my first time.

Not all problems are well defined. Generally the more business impact you have, the more undefined your work is.
a lot of this depends on the dynamics of the team.

i don't think pair programming necessarily means typing 100% of the time. I get a ton of value in tackling a very ill-defined task with another member of my team, especially in the design / architecture phase where I find that a lot of engineers tend to over-think and over-engineer a task.

i don't want to discount your experience tho - i def. have moments where i absolutely do not want to pair, but i do think it's a practice that is super context (team) sensitive

If that is how you define pair programming, how is pair programming different than just working with someone on a problem? Legit question.
Pair programming is literally working with someone on a programming problem.
yeah i think that's a fair question. i think while the extreme programming community may define it as a practice with a particular set of guidelines, ultimately it's about working effectively with another programmer.

sometimes that means bucking the rules of whatever the methodology prescribes and doing what works for your team.

for example: xp people tend to suggest frequent pair rotations in order to spread knowledge (faster?). I know half my team, though advocates of pair programming, would be pretty unhappy with that given the costs of context switching.

maybe some people will point at that and say we're not really pair programming /shrug

I haven't had a job that required pair programming, but I've found it can be a pretty useful tool in certain circumstances. There have been a number of times when a colleague and I have naturally fell into pair programming for tasks, normally starting with us just sharing ideas and one of us eventually jumping in front of a keyboard and the two of us trying out several ideas live. It certainly isn't as formal as "pair programming" normally might imply, but as someone who's hasn't been in the industry all that long, I've found the times that one done it with more senior teammates to be insightful, as well as pretty effective for certain types of tasks.
I'd rather have a design session and a code review in that case.

What I've found pair programming useful for is tooling and workflow.

When working that closely you'll notice when they hit a few keys and are suddenly in the right file, and you go "wait, what did you just do?" and they tell you about this great new Vim plugin.

Inversely, they start doing something really manual and you can say "hey, I do this a lot do I wrote this utility script that automates it"

(comment deleted)
I think this has mostly to do with the tasks at hand.

I know a bunch of people who hate their jobs and would feel pretty bad if someone would supervise them 40h a week, even if they may perform better under such supervision.

Lol no thanks. Hire a robot
Came here to say this. When ever I'm lagging on putting down some code. I can grab another developer that seems like they are just in that afternoon haze too. A lot of times just talking about things much more elegant and dry solutions come too.
Now you know what foremen do.
I can't really imagine a job in which someone is bothered to pay you to do work, but cares so little about the work you're outputting that they don't notice whether or not you're working efficiently/effectively.

I've worked from home full-time for just over 5 years now. I'm self-employed and answer to my client directly. I operate almost entirely autonomously and am given a heap of freedom, both tech-wise and with respect what hours I work. I've not once in 5 years reported my exact hours worked - but of course do report days off sick etc.

However, despite my client not being a developer (or overly technically inclined) I can assure you my client still notices if I'm being less productive than usual - which over the course of 5 years is somewhat inevitable e.g. when my daughter was born.

Actually, I find that not having someone knowing that I'm ticking over hours sitting at desk, hitting buttons, and making the screen flash, causes me to work more. For the last 5 years I've consistently worked more hours than I ever did in an office environment, not by a small margin either.

Sometimes you just don't feel "switched on", but if you're working in an office you're still tallying up hours. However, if I'm not totally switched on, I somewhat subconsciously know I need to work longer in the afternoon/evening. This is simply because I feel pressure (am motivated) to live up to the standards I've established with my client. I must add though, that my client is absolutely fantastic, any perceived pressure is almost entirely self-inflicted.

If anything, I need to start setting firm boundaries so that I work less. This means I need to start considering those less productive "hours worked" as genuine hours worked, just as I would if was working in an office.

When you take a walk and come up with clever solution to expensive problem wouldn't it be more valuable than proving you deliver keystrokes constantly?

I think you need to have some explanation, at least for yourself, what kind of value you deliver even when you are just around not so occupied.

Author might be even more productive if the watcher threatens to or inflicts pain on her if she begins to slow down.
I’ve had experiences like this with glassblowing. When I have an interested observer, such as a student, I often get enthusiastic and make several pieces in a row and use a wider variety of techniques.

Also, many glassblowers work while hanging out together on video chat.

Well in China they do that with cameras see if students are focusing on the classroom. I think the feeling may be the same.
A camera or knowing that someone can remotely view your computer screen at any time should do the trick.