Hacker News new | past | comments | ask | show | jobs | submit login
Maintainers make the world go round: Innovation is an overrated ideology (projectm-online.com)
378 points by ForHackernews on May 8, 2017 | hide | past | web | favorite | 169 comments

You could say the same thing about any non-glamorous/lucrative position.

"Garbagemen make the world go round. Without them, we would drown in our own filth"

"Nannies make the world go round. Without them, half the workforce would be stuck at home"

"Auto mechanics make the world go round. Without them, we would have no way of getting places"

Ultimately, all such arguments are inane and pointless because every single job that exists in society

A) is important to the people paying for it

B) has wages that are based on both the importance of the job and how easy it is to find someone capable of doing it

C) The idea of glamorizing any job, and allowing yourself to be influenced by a job's glamor-rating, is just superficial drivel. Don't judge yourself or others by the job listed on their business card. If you feel the need to judge someone at all, judge them by the impact that they, as an individual, are making in the world.

I think there is a major flaw in the analogy.

Basically, maintenance of complex systems or assets, are intellectually demanding tasks. All your examples, however, are completely the opposite. I am not saying that the jobs do not require skills, it's just that these skills are not that valuable.

Also maintainer adds value in the positive way. Where all your example is to prevent value loss. The 2 are fundamentally different.

Contrary to common knowledge, maintainer is not needed to make the system running, if nothing changes. Maintainer is necessary to make the system running in changing environment. Real-world software or any valuable assets require change to keep up with the world surrounding it. A great mental bias is to implicitly think maintenance is about "keep it running", where at the same time conveniently omit the part about changing environment.

>maintainer is not needed to make the system running, if nothing changes

Something tells me that you haven't maintained badly designed systems. Some actually do require manual attention from time to time, even if nothing changes. A single non-rotatable log file becomes too big and a restart has to be scheduled or an application starts behaving weirdly because a buggy and rarely used bit of code suddenly got used once and it destabilized the entire thing... I'm sure there's a lot of other ways in which applications can require maintenance even if it hasn't been updated for months.

> Basically, maintenance of complex systems or assets, are intellectually demanding tasks.

Not as hard as making those systems in the first place.

> I am not saying that the jobs do not require skills, it's just that these skills are not that valuable.

I'm not convinced that an average network engineer does a more challenging (or valuable) job than an average car mechanic.

It's often easier to build a system than it is to maintain it, because you have a free choice across the solution space and you don't have to maintain any existing commitments.

Maintenance (physical or virtual) can often be made extremely difficult or expensive because of the need to avoid disrupting working systems, and the need to integrate with what's already there.

You've never had to add a new small feature to a highly customized Wordpress site someone else built I think.

Fair enough!

>I'm not convinced that an average network engineer does a more challenging (or valuable) job than an average car mechanic.

Mechanical work on cars isn't rocket science, in the same way that IT work isn't. It's difficult if you don't know what you're doing, and not hard if you do.

Both things most anyone could do (at least at a basic level) themselves, but most people don't care to learn.

Come and work on our system. Poor database design and way too much complexity in simple tasks.Maintaining it is far harder than a rewrite would be.

> Poor database design and way too much complexity in simple tasks.

This actually supports my point that making a decent system in the first place is hard. And if you have a reasonable estimate that a rewrite would be cheaper than incurred maintenance costs, it might be worth it.

I was speaking generally though: an average quality system and average maintenance work. Some systems are terrible and fall apart from an angry glance, some are so good they require almost no maintenance. In my personal experience systems maintenance can be menial and time consuming, sometimes clever, but it ain't no rocket science.

> "Nannies make the world go round. Without them, half the workforce would be stuck at home"

For my entry to Improbable Ideas day.... why don't we bring children to work? After they're no longer infants, its' plausible for children to simply run around the office, sitting quietly in the same rooms as their parents, and perhaps doing odd-jobs.

It'd require a more holistic attitude to education than is presently allowed, but it is certainly possible to implement.

If I'm working from home, even if my 4yo has every toy in the house at his disposal, after a while he will inevitably come into the home office and spin around on the spare chair saying "Why do you have to wooooork so much? I need you to help me buuuuuiiiild something. Why can't you help me? I can't find (a particular piece of Lego). Why is Mama at wooooork? Why are you working?"

The child is asking a valuable question.

Why all of this working? How can the work be explained, made interesting and relevant. Is there some way of involving the child that is educational rather than abusive? (Also, if a 5 year old can understand how to correctly use the UI for the application, why can't your client?)

I think it'd be a valuable question if it were asked by someone sincerely interested in the answer, but at that age children are really just complaining.

Could you involve the child? Maybe, but training someone for work is non-trivial. When I'm just starting to train someone, our collective output is usually less than mine working alone, and that's for someone with the requisite skills. How much work could reasonably be assigned to children in a way which isn't just a waste of time? How much are they really going to benefit from that?

I agree as regards four year-olds. That said, the usefulness of older children, even ~8-10, is frequently underestimated. There was a time when they made useful apprentices and assistants, though admittedly at crafts that were probably less abstract and specialised.

Still, I started programming and using UNIX when I was 9. I imagine I was probably capable of some amount of useful output by 11-12, though it would have to be carefully directed by a pragmatic mentor with a view to bottom-line impact.

I'm not sure why you're getting downvoted, perhaps because people don't want to imagine modern child labour.

From my side, in the company where I work there are a small handful of people who could easily be replaced with fully unskilled labour (ie. an 8-12yo) with no negative effects.

I wasn't proposing that we employ children! My statement was made in the context of the discussion about the practicality of engaging and occupying children of certain ages in higher-order cognitive tasks.

Except for negative effects on the children? How do you schedule that with school and homework?

Of course. I was only showing it as a comparison from the adult side to show that some adults are easily replaceable with children.

I wouldn't want to actually implement it.

Even some adult employees require non-trivial amounts of effort to keep them working, just so anyone in a role with oversight can actually try to get something done.

That's like every employee with a child taking on a completely incompetent intern. It will just suck down the productivity of them massively without really gaining anything useful.

Perhaps it's the irregularity of your working from home may contribute to this.

My parents were artists who both worked from home, it was generally established that I didn't bother them in their offices when they were working on their projects.

I am grateful to have grown up in such a situation and if in the future I decide to have children I will definitely switch to a primarily or entirely remote position.

You would probably enjoy the book "Iron John", as written by Robert Bly. Your entry for today in the Improbable Ideas forum, was how things happened in the past. Children would go to work with their parents, in the pre-Industrial Revolution era.

Father was a blacksmith? Children would go with him, and sweep the floor, pump air into the furnaces, and organize things. As they matured (both in age and skills), their tasks would grow. Father's brother was a leather worker, so it was easy to send your kids to work with your brother, so they would see if they wanted to be blacksmiths or leather workers.

The Industrial Revolution destroyed the father-son relationship. Fathers would leave the house at 6:00am, and come back at 8:00pm. Young boys would have a blackhole regarding what their fathers did during the day. This blackhole, Bly argues, would get filled with "demons". Not voodoo-demons, yet "demons" in the sense of a "lack of knowing what my dad does every day".

In part, we as a society in the West have been suffering the effects of this destruction ever since. In the 80s and 90s, with the increase of mothers in the workforce, the same thing is now happening to the mother-daughter relationship. We have to wait a couple of generations to see what effect this has on women.

> sitting quietly

...what does this mean? Is this something that children can do?

yes, for minutes at a time!

Of course.

Both my wife and I have taken our kid to our respective offices and it was fine. She just sat there drawing and playing on her tablet. This was when she was 6-7 years old.

Not all kids are brats.

And not all kids are content with sitting inactive, quietly, in a chair for 8+ hours. The fact that they are energetic, inquisitive, and have a need for attention from their parents does not make them "brats".

Not all kids are 6-7 years old. And I assume you're talking about occasional visits rather than a daily replacement for childcare or a nanny.

Are you drugging your children? What are you doing to suck the energy out of them that they can sit in an office for hours without recess to run?

You've never tried to work around small children, have you? I work from home and let me tell you, if our nanny has a day off it's absolutely impossible for me to get anything done beyond a couple of emails if I'm lucky.

... and the few e-mails I manage to send sound pretty ludicrous and ill-considered.

The thing is, for people with kids, at least half of their motivation for showing up to work is getting away from the kids for a couple hours. (Exaggerating, but still).

As someone with kids, I never understood why people say this. Oh what I'd give to spend every waking moment with them as they discover the world.

Not all parents, children and situations are made alike, I suppose.

Personally I enjoy spending a portion of my day not interacting with humans, let alone my children.

Some days I even enjoy my commute, because even though I'm on a busy line and we're all jammed into each others armpits, my mind is my own and nobody wants anything in particular from me.

Not to be too pointed about it, but you don't have kids, do you?

Just a few reasons that come to mind:

- They'd be yet another distraction

- Who pays for the resources (space, nutrition, time)

- What if you don't work in an office

- Work environment might not be a great place for a kid

The typical work environment isn't a great place for parents either. Adding kids might be an opportunity to improve that.

I agree with you. All people here are complaining with status quo bias in control of their minds.

I think work can be much more interesting to a child than school or a nanny and the playground.

Also, it would be so much cheaper to spare school/nanny costs that a lot of money could be spent to make the environment better for the child.

We're talking about children older than 7 or so, right? I think children from 7 to 20 could be taken to work and actually learn much more.

All depends on how you raised the children before 7-10. They don't magically become a certain way at 7-10. ;-)

I mean, to some extent they do. Kids are born with an amazing amount of cognitive bootstrapping and initrd. But by elementary school years, the difference between children who were reared in intellectually stimulating ways and ones who weren't become patently obvious and to a larger extent entrenched.

The glamor is fake, the dollars are real.

(OK, fine the dollars aren't real either, but what they can buy is.)

I subscribe to the sentiment!

As a semi-philosophical aside: in what capacity are dollars not real, but the things you can buy with them are?

Since everything seems to be a means to an end, from glamour (gets you money), to money (gets you food) to food (provides energy to your cells) to species (reproduction, evolution) to more efficient use of resources, all accelerating the universal march toward the heat death of the universe.

At which point do things in this optimization chain become "real"? What makes a specific segment "not real"?

I think you overestimate both the market saturation and lucrativeness of nannies.

They didn't say anything about how much nannies make.

Yes all roles are important.

If elephants get wiped out, there will be some consequences.

If insects or algae gets wiped out, the world is apocalyptically screwed.

yet you are comparing the impact of wiping out two species of mammals with possible impact of all insects or algae being gone...

Well no the point is that while all roles are important, some are so foundational that if they break down, the whole system breaks down. Innovation doesn't happen all the time, and it needn't to, as long as the system is stable and works for everyone (in theory.) We still need it when it's required, otherwise maintainers are indeed honourable roles.

(Here's one source for my counter-analogy: http://www.livescience.com/52752-what-if-all-insects-died.ht... )

> I’ve actually felt slightly uncomfortable at TED for the last two days, because there’s a lot of vision going on, right? And I am not a visionary. I do not have a five-year plan. I’m an engineer. And I think it’s really – I mean – I’m perfectly happy with all the people who are walking around and just staring at the clouds and looking at the stars and saying, “I want to go there.” But I’m looking at the ground, and I want to fix the pothole that’s right in front of me before I fall in. This is the kind of person I am. - Linus Torvalds @TED[1]

[1] https://www.ted.com/talks/linus_torvalds_the_mind_behind_lin...



In response to the notion that we need to break userspace in the name of progress/innovation.

"If the President had picked me to predict which country [in postwar Europe] would recover first, I would say, 'Bring me the records of maintenance.' The nation with the best maintenance will recover first. Maintenance is something very, very specifically Western. They haven't got it in Russia. If I got in there in the warehouse, let's say, and I saw that the broom had a special nail, I would say, 'This is the nail of immortality.'" - Eric Hoffer

> Maintenance is something very, very specifically Western

Not really. The need for maintenance occurs when a specific "Thing" has been around for long enough, and when the institutions, schools of thought that sustain that Thing persist for long enough. When there are radical, rapid political/economic/political changes, "maintenance" and the need for maintenance goes away. This article focusses on maintenance of industrial artifacts, and nowhere in recent times have industrial artifacts (boats, lifts, software), the companies that produce them, and the living standards they create, and the governments that encourage them been continuously existent longer than the "Western countries" (since the early 1800s at least). If you were to look outside the sphere of industrial artifacts, for example at old religious buildings - churches, mosques, temples - you'll see solid examples of good maintenance all over the world. The Koran, a literary artifact, for example, has been well-maintained for over a thousand years across the world by a mostly pre-industrial society.

TLDR: Keeping something running isn't just a "Western" thing - but keeping industrial products running is certainly Western, because only the West has seen a continuous and sustained industrial movement for over 200+ years.

There used to be major US companies that were really into maintenance, and pverbuilding for low maintenance. The Bell System. The Pennsylvania Railroad (the Standard Railroad of the World). The King Ranch (the King Ranch line was "we don't fix fence, we build fence.") Teletype Corporation. IBM. Xerox.

There's something to be said for that.

Reading the quote, it seems the east/west divide is along first/third world rather than the traditional hellenestic divisions.

Nonetheless, it does diminish the quotation.

Hoffer is wrong. He's confusing different military philosophies with this mixed up idea about national identity regarding "maintenance".

I don't think anyone would suggest Germany didn't perform a minor miracle by rebuilding as quickly as they did (with massive support), but the Russians went from having half their country burned to the ground, to a world superpower (also with a large amount of support), in a shorter time span than it took West Germany to return to pre-war industrial levels.

More than half. The vast majority of Soviet population and industrial base was in the burned-down part.

Who supported Russia after the war?

The conquered countries that were stripped of a good part of what they had, largely.

would appreciate a reference.

p.s. many parts of eastern europe has been destroyed during the war as well; and was getting rebuild themselves.

Most of what was left of factories in the Soviet occupation zone (the later GDR) was dismantled and moved to the Soviet Union as reparation in the years after WW2. Here's some info in English about this: https://en.wikipedia.org/wiki/Economy_of_the_German_Democrat...

When I was young I heard a lot of crazy stories about the Russians in Germans (I grew up in the Erzgebirge mountains where the Russians mined the Uranium for their atomic bombs, think of an Eastern 20th century gold rush), some of those stories are how entire production lines were moved onto trains and to the Soviet Union nearly over night (they had a lot of experience with this since a few years earlier the Soviet Union basically moved their entire military production a few thousand kilometers eastward, out of the war zone).

Interesting, thanks.

Here's a reference for you https://en.wikipedia.org/wiki/Gulag People all across occupied lands were enslaved (on paper sentenced for some nonsense and send to workjails)

I was referring to the massive lend-lease support, and the continued industrial support provided to the Soviet Union during the war, not after war. I should have been more clear about that.

But this measure, Japan proved to be utterly Western. They also recovered brilliantly, as predicted.

Er, can you explain that last sentence?

A nail as in its own rack to hang on. Because they value that broom and its mission enough to go through the trouble of creating a rack for it, you could imagine that they put some higher value in that mission (maintenance) and its role in accomplishing it.

At least, that's what I got out of it.

It's quite real. TechShop once had racks for their brooms and dustpans. After two moves, the mid-peninsula TechShop lost their broom racks. Maintenance went downhill about that time.

That's about when the LSD started kicking in.

>Maintenance is something very, very specifically Western.

Yep, no other cultures maintain anything.

This is pretty laughable. To me, almost every Western country is half-broken. When I lived in the US, I was surprised how frequent power faults were there. Streets are full of bumps and corroded pipes were everywhere. And in Europe. Do they have any concept of a decent public bathroom over there? Oh, and don't get me started on how badly their trains are operating. I bet that many train accidents are due to the poor maintenance. It's probably true that maintenance is less paid an attention in poor countries, and there are more non-Western countries that are poor. But it's funny that people can get away with a claim like this. It almost sounds like a white supremacist to me.

TL;DR: Japan is nice.

The GP quote contrasts "the West" to Russia, which is itself overwhelmingly White. It's hard to see a white supremacist angle on a comparison between two majority-white blocs.

In the taxonomy of white supremacists, Russians are "slavic". This was literally part of Hitler's justification for invading eastwards.

("Whiteness" is not simply about skin colour. There's a fascinating/ugly history of how various immigrant groups to America became "white")

Yeah, Japan was my first thought when I read that. You can't tell me with a straight face that they don't maintain everything to within an inch of its life.

Ironically except for the single most maintained item in the west, which is a house.

In Japan they are treated as basically disposable.

Gotta have a gold sink in your economy somewhere to fight inflation.

> TL;DR: Japan is nice.

True, but was also devastated by a war that allowed rebuilding almost from scratch.

Maintaining things coincidentally, and thinking in terms of there being such an activity as "maintenance" that you might want to start doing on any given thing if you aren't doing it already, are different things. I can believe that some cultures have no concept of "Total Cost of Ownership" distinct from purchase price, for example.

You might say I've observed this many times in my career. I think the best move career-wise is to be one of the people who makes the new thing. Those who clean up after them, bear the brunt of their design flaws and careless mistakes, will never be recognized, appreciated, or remunerated as well. At least in my experience.

Ask yourself this: who is the most famous maintainer you can think of? (Not someone who devised an innovation and then maintained it - pure maintenance)

At the same time, a maintenance position can offer benefits as well. With the exception of emergency bug fixes, maintenance is likely to have a more stable work load which might appeal to those who want more work-life balance. Some engineers may prefer to spend their time on lower level tasks that aren't showy but are nevertheless important to get right. Working on threads/locking/synchronization may not be glamorous, but it's critical for the operation of everything else and difficult to do right.

I think we should be putting more effort into recognizing/compensating those involved in maintenance work rather than steering folks away from it. The raw technology invention is important, but the maintenance work is what makes it ship and continue to work for customers. There are also plenty of widely-used projects remain relevant only due to maintenance. Projects like OpenSSL or the Linux Kernel come to mind here.

It's a risk/reward trade-off. You're never going to be famous being the maintainer, but you'll also never be viewed as crazy, nor do you have to face outright failure. You hear about the innovators who succeed; you kinda shake your heads and pity the ones who don't.

I'd be very curious to know why this is getting downvoted, and if it has anything to do with the fact that YCombinator is a hub for the 'innovators' the post (and this comment) are talking about.

Don't downvote because someone hurts your feelings - that's not what this site is for, guys.

Aw, it's just fake internet points. I don't worry about it too much

It's also an issue of visibility. Downvoted comments are seen less and don't get to contribute to the zeitgeist. That's sort of by design and I'm not trying to have that conversation, just saying that it's more than "aw, just internet points".

This is the situation where the reason behind downvotes would be appreciated. He/she said the truth - especially for startup environment.

It is easy to agree.

It is difficult to tell if not-agreeing is due to a difference of opinion or the statement being off topic.

Thus while it may seem natural to cast to two directions the implicit difference between agreeing or not is actually more nuanced. The /reason/ is important in the latter case.

> Those who clean up after them, bear the brunt of their design flaws and careless mistakes, will never be recognized, appreciated, or remunerated as well.

Next time I have a say in hiring I'm not even going to bother looking at people who have only done green fields work. They've never had the opportunity to learn from their (or others) mistakes.

> Ask yourself this: who is the most famous maintainer you can think of? (Not someone who devised an innovation and then maintained it - pure maintenance)

What do you consider innovation? Two I'd point to are Bram Moolenaar (vim) and Linus Torvalds. They both wanted to clone existing systems, not to innovate. The same goes for many other open source creators, they had a need to solve, it wasn't a creative exercise.

>Those who clean up after them, bear the brunt of their design flaws and careless mistakes, will never be recognized, appreciated, or remunerated as well.

It's a tradeoff. Innovating employee #10 at a big succesful startup will be better-rewarded than your average maintaining employee at, say, Facebook today, but that "maintainer" is going to be far better rewarded than innovating employee #5 at some-startup-whose-new-thing-proves-unwanted.

Aren't software as a service and cloud computing examples of businesses that are primarily about maintenance?

It seems to me that the best businesses are mostly about maintaining and profiting from some technology. Startups are advised to keep costs very low until they have "product-market fit", that is, until they have innovated enough to have made something that people want. After product-market fit has been achieved, marketing and sales begin to dominate; the technical innovation is over except for maintenance and incremental improvement.

Most businesses don't require innovation at all. McDonald's, for example, didn't invent the hamburger, but it has grown large because of its ability to grow while maintaining high standards.

Even businesses that we would think of as very innovative don't make their profits on continual innovation as much as growing and maintaining a standard process. YCombinator, for example, has a process of selecting great companies and nurturing them. The companies do a lot of innovating, but YC itself is basically maintaining and incrementally improving the innovation of Paul Graham and the other original founders.

Profit comes during the maintenance stage.

Most businesses don't require innovation at all. McDonald's, for example, didn't invent the hamburger, but it has grown large because of its ability to grow while maintaining high standards.

McDonalds and its competitors have been continually innovating on the details of how to provider hamburgers and chips to customers quickly and cheaply.

That's not innovating, that's optimizing, an important aspect of good maintenance.

I completely agree. Maintenance is important, but I feel I learn more and get more career growth out from jobs that involve building new things.

> who is the most famous maintainer you can think of?

The pope.

> Ask yourself this: who is the most famous maintainer you can think of?

I can think of a few, actually. Pretty much every BDFL in every open source project ever. Most famous is probably Linus Torvalds. He invented the first version of Linux, yes, but he's been primarily a maintainer not inventor ever since.

> (Not someone who devised an innovation and then maintained it - pure maintenance)

Everything you've mentioned doesn't count.

Not saying you are wrong, but who is the most famous inventor you can think of? No no, from a startup, similar to the scenario depicted. That's right, none (+- a handful) - regardless, we aren't doing this for fame anyway (I'd imagine).

The most famous maintainer(s) I can think of are the people who paint the Golden Gate Bridge.

Name an innovator who didn't spend any time improving, maintaining, or cleaning up their work? Name an innovator who only ever innovated? Pure innovation.

Jon Postel perhaps?

This is like saying LeBron James is less important than Cleveland's role players, because you need five people on a side to have a basketball team.

The question isn't who's "necessary", since everyone who is necessary, no matter in what way, is necessary; necessity is a tautology.

The question is whose contributions are more replaceable.

Had to reply here given how special it is to watch LeBron play right now. Even if you're not a basketball fan, watching LeBron is like watching an extra rogue queen run a chess board. He should be past the peak of his career, but he's adapted his game to become a better all-around player. Reminds me of Mega Man suits where he can co-opt certain abilities given who's on the floor. He can kind of make his teammates necessary, if that makes sense. And he's more than aware of everything that's happening on the floor and on the benches. Him pulling Powell back on the court was a fun, albeit silly example of this: https://twitter.com/wenotsocks/status/860900323275264002

I'm not sure there's much relevance here, but there's something to watching someone maintain competitiveness as they age.

I'm not so up-to-date with Basketball, but certainly in Soccer and Rugby you need both. You need the disciplined engine-room of the defence and centre-field to maintain control of the field and prevent concession. At the same time however you won't get anywhere without the big-name points-scorers.

Similarly, businesses need the sales guys to go out there and get the customers. But without disciplined design, engineering and maintenance you're going to end up leaking all your revenues in support costs, not to mention intangibles such as reputation.

I disagree strongly. Innovation is a replaceable concept. Anybody has innovative ideas.

Putting in the years of work to implement them, on the other hand, is rare.

There is probably a distribution here that matters - without maintenance there is no foundation for innovation, without innovation there is no motivation to maintain - man wants to produce and consume newer ideas, materials, tools, items etc.

There's a great piece in the Lapham's quarterly about maintaining NYC's infrastructure and how without any maintenance, NYC would be replaced by forest cover within 200 years. Can't find it right now but it is a great read trust me.

This reminds me of a great quote, attributed to Thomas Eddison:

"Opportunity is missed by most people because it is dressed in overalls and looks like work"


You know, if Hollywood made the blockbuster movie "Infrastructure" [1], maintenance might be views as a "good thing."

[1] https://www.youtube.com/watch?v=Wpzvaqypav8&t=17m14s

No, it's precisely the opposite! Maintainers (using whale oil for energy) would have driven every whale to extinction and left the world without an energy source a century ago. Maintainers (using horses for travel) would have drowned Manhattan in horse shit a century ago.

Innovation is if anything under-rated and under-funded and under-supported. The homes of hundreds of millions of people and energy itself is threatened by depleting fossil fuels and global warming...and some of the major efforts to stop this have depended on effectively "insane" entrepreneurs like Elon Musk...not a smart system! All the while hundreds of billions of dollars in health care costs for just say unnecessary tests flows to negative value addition maintainers.

Maintainers mostly either conservatively follow & accept or exploit the current system. It's innovators who've driven down the cost of lighting your home to a few hours of income or the ubiquity & cheapness of books & information (perhaps to the detriment of wisdom but that's another story) to stopping war through protest to ending non-man-made famine.

Wouldn't maintainers just be in small social groups, eating fruit and whatever else they happened to collect?

That's hunter-gatherers! That's more than 10,000 years ago before agriculture developed...

Sure. How do you get from there to a city without any innovation?

How do you even get to a sharp stick without innovation?

>How do you get from there to a city

Enclosure https://en.wikipedia.org/wiki/Enclosure

Basically peasants, once kicked off the land end up herded into cities.

Presumably the thinking at the time (beyond "lets make ourselves rich") was that agricultural production is more efficient when undertaken as a large-scale concern rather than by scattered small community holdings.

So how is the step of consolidating the land not an innovation then?

Or the invention of the various tools to actually make the agriculture viable and progressively more efficient.

It is an innovation. I was presenting this as an alternative antecedent to the development of cities.

Also provides an interesting corollary on the consequences of innovation.

> How do you even get to a sharp stick without innovation?

Evolution? That's how agriculture and many of our earliest tools were created. I don't think there were many/any single steps that we would consider innovative.

I don't see how something like intentionally sharpening a stick would not be an innovation over occasionally finding one.

It may not have been intentional, it could have been someone rubbing it against a rock out of boardom or trying to start a fire with it then realising it was pointy. In that case it was a discovery, not an innovation.

We may have had a very suboptimal way of doing it for centuries while the process evolved. I'm sure a lot of people here have stories of their pets confusing cause and effect.

Great point!! Yes, exactly!!

So you agree? Innovators are responsible for depleted resources and global warming.

Well, perhaps it's the previous generation's innovators and the current generation's maintainers...if you want to trace out lines of responsibility but that's tricky...whoever discovered petroleum probably never wanted so much burned up to heat up the whole planet! The maintainers (energy companies) do...they're going to die (those firms) and they're ready to take the whole planet and all of us with 'em.

On a lighter note, Pump 6 from "Pump 6 and Other Stories" (https://smile.amazon.com/dp/B0071CX7V4/) is a fun take on a world that has become too good at making things that don't need maintenance.

In contrast, in David Brin's The Practice Effect objects improve with use instead of wearing out. Rich people hire others to "practice" their possessions.


Innovation changes the world, maintenance keeps it going.

Obviously the thesis is true. Maintenance is crucial while innovation rarely changes anything substantive...

I think that is the reason that we hold innovators in high regard. People are very bad at it and it happens so infrequently. We rarely get the right person when we hand out credit so such idolization is usually meaningless, but I suppose in the long run that is not important. We have this irrational need to attach a person to the idea.

So we end up failing to properly credit any of the people that make and keep our civilization...

> People are very bad at it and it happens so infrequently.

Is this really true? Plenty of innovations occur in parallel. Perhaps the people who desire to innovate dwarf the opportunities for innovation.

"We" hold innovators in high regard because of technology and the way capitalism neeeeeds technology. We live under capitalism so we are taught to love innovation. By maintenance alone a capitalist (person who has so much he no longer needs to work) cannot out run the herd, he needs to innovate because the other capitalists are also innovating. Since labour has already been squeezed to the max, they seek technologies that increase the output of labour, or even "better", make labour obsolete.

Workers should be weary of innovation, but they have been taught to love it, and thus they do :)

Wouldnt maintenance lead to innovation analogous to necessity being the mother of invention?

Most of the inventions came up as an easier or better way of doing something which was the current maintainer(to be innovator?) decided to rework/create .

And if the author talks about only maintenance where no development can be done(even those that make the life of maintainer easier). IMHO , Maintenance procedures should also be constantly improved and hence that would lead to innovation.

I found the thesis interesting and plausible: that innovation is fetishized.

But on a slight tangent I wondered whether the "innovation" that they complain about is a particular variant, one that we're all familiar with here: a pseudo-libertarian start-up variant

  "Innovation ideology is overvalued, often insubstantial, and preoccupied with well-to-do white guys … in a small region of California"
It seems easy to argue against this tired representative of innovation.

By contrast there are those that would argue that most of the major technological and scientific gains have arisen, not from these VC hype-machines, but from large-scale state planning and investment. One of the best expositions of this argument is from economist Mariana Mazzucato: https://www.youtube.com/watch?v=yPvG_fGPvQo

Or as I like to say, "incompetence makes the world go round". Extracting little glimpses of functionality out of a chaotic mess is a challenging, at times satisfying and definitely valuable exercise that keeps many people at work...

Maintenance is not the opposite of innovation, it is the opposite of good design.

One could also view good maintenance as the antidote to bad design. Or the antidote to "not 100% perfect" design and execution. Nothing's ever perfect. Innovation is to maintenance as inspiration is to perspiration.

I'm not sure I understand your point - the best built system in the world still needs maintenance. Parts go rusty. Requirements change. Those are not results of bad design, but the inevitable consequences of the real world.

The better the design the less the maintenance. There might not be a perfect design in the real world, and we would agree on that. But at least the ideal perfect design would imply zero maintenance.

What happens if requirements change? What happens if a related technology is invented which would make your current design more efficient? Do you class changes like that under maintenance?

And how do you check your theoretically 'no-maintenance-required' design genuinely doesn't need any maintenance? And how do you get round the catch 22 that you can't build a perfect system without perfect knowledge about what to build, but you can't have perfect knowledge without feedback from an existing system?

Instead of building a product that theoretically never needs maintenance, I'd rather build a product for which maintenance is easy.

Well, maintenance is what you do to avoid (negative) change. Changing requirements or improvements are not maintenance.

There's an analogy here vis-a-vis tradition/progress. In order to be reasonably sure that a change is an improvement, you must understand what you're changing and how. To borrow a Chestertonian example, if you encounter a fence and you don't know why it's there, find out why before you remove it. Maintainers are in the best position to understand the impact of making changes, and because of that, they're able to function as either advisers or as "innovators" by knowing where improvements can be made and having the knowledge to understand why they're improvements.

The maintainers! I know a couple people connected to this group - heard great things about their 2nd conference last month.

The premise is great. From Russel's article on Aeon:

"We organised a conference to bring the work of the maintainers into clearer focus. More than 40 scholars answered a call for papers asking, ‘What is at stake if we move scholarship away from innovation and toward maintenance?’ Historians, social scientists, economists, business scholars, artists, and activists responded. They all want to talk about technology outside of innovation’s shadow."

What about Improvers? Not a word about us? You can innovate while maintaining.

Ridiculous. Maintenance is momentum. Innovation is boost.

Each boost is miniscule, but our momentum is enormous after thousands of years of human development so of course maintaining our momentum gets us incredibly far.

I'd say there's no fine line between maintenance and innovation. Many innovations arise in response to the pains of maintenance.

Related video by Jordan Peterson, on how liberals and conservatives need one another because liberals (high trait openness) innovate, but conservatives (high trait conscientiousness) maintain things:


Interesting topic, lots of sweeping generalizations being thrown around.

Personally, I've found my politics have not changed substantially as I've aged. As a 20 year old, I was liberal, and in my mid-30s with a mortgage and a family, I'm still quite liberal.

However, I've found my professional proclivities have changed. I value maintenance and conscientiousness much more than I once did. I think that's due to a few reasons. For one, I've been through a few big rewrites, and inevitably, I've seen them fail or be much harder than anticipated. But also, my focus has shifted from using cool new technologies to delivering value through working software. It's often much less effort to maintain and slowly improve software systems than it is to innovate your way out of a mountain of technical debt.

It's hard for businesses to understand this. Often, the expensive "A team" builds new software, while the cheaper "B team" maintains it, but the reality is that in short order, that software will effectively be written by the "B team". Software would last much longer and just function better in general if you kept the "A team" on to maintain, upgrade, refactor, etc.

Of course, this isn't totally business's fault, because maintenance is seen as a career killer. So if you try to have the "A team" maintain, many will leave for greener pastures where they get to build new systems in new tools.

As an industry, I wish we cared more about the longevity of systems, rather than what new tools are used. We seem to be impressed when we hear someone talk about building a new system in React, Elixir, and Kafka (or whatever's new and shiny at the moment), but nobody wants to hear someone talk about a system they built eight years ago in, say, Java that is still working well, still helping users, still meeting the needs of the business.

Maintenance and green-field development are different skillsets. Green-field development requires that you can envision what doesn't exist and hold it in your head long enough, and precisely enough, to build it. Maintenance requires that you take what already exists and improve it. Green-field development tends to be more intuitive, and it requires the ability to balance many trade-offs and consider many possible solutions in your head. Maintenance programming requires less creativity and to some degree less short-term memory, but more attention to detail, more logical thinking, more communication skills, and more ability to read & make sense of existing code.

I've worked in both situations, but I'm better at one than the other. It's very rare to find someone who is excellent at both. You do get the benefit of accumulated domain knowledge when you keep the initial developers on to maintain the system, but oftentimes they don't have the temperament or personality to be happy and successful in that role.

So, I think I agree with you, in general, that people can be better suited to green-field or maintenance, both from a skillset and personality perspective. But I also think good, active maintenance can have a lot of creativity and require, as you say, envisioning what doesn't exist.

If we're just talking about bug fixes and level-2 support or something, sure, that's mostly about attention to detail, etc. However, maintenance is much more than that, in my experience.

Maintenance involves keeping a system up to date and functioning in the face of changing business requirements. So, you have a new customer come onto the system or a department wants to change a business process, and now you need to change the system. What was once a single implementation of some feature now makes more sense as an abstraction with extensibility points. Or maybe just the system has more data than anticipated, and a good developer might realize that instead of putting a bandaid on the Postgres search, it's better to integrate a dedicated search solution like Elastic.

The point is that software is never done, as long as it's being used. Drawing a line between development and maintenance is often just an arbitrary distinction.

I guess I should've said "a different kind of creativity" rather than "less creativity". When you're doing maintenance work - even the sorts of large scale rewrites that totally change the architecture - your requirements are known. You can write down the goal, ideally write automated tests for it, and then you're done when the system passes the tests. Figuring out the solution may require a lot of creativity and intuitive leaps, but at least both the starting and ending points are expressed in rational terms.

For green-field software products (contracting is different, and much more akin to product maintenance), your requirements are unknown. Your ultimate acceptance criteria is "does the user like it?", which is an emotional one, not a logical one. This requires that you connect the rational part of your brain, which speaks the language of the machine, with the emotional part, which speaks the language of the user. Being able to exercise both parts in tandem is a very rare skillset, which is why market rates for truly innovative programmers range into the millions. (This ability to fuse emotional & rational thinking together is also a skillset required of CEOs, executives, politicians, and pop entertainers, all of which are also professions with compensation reaching into the millions.)

Yeah, that's fair. I'd agree green-field product development is different from maintenance. It requires great UX thinking, really understanding the users, product-market fit, maybe some behavioral economics, etc.

I was thinking more about enterprise or line-of-business green-field development. That still benefits from all of the things we're talking about, but it's a different set of problems to focus on.

Generally I would agree. But this past election cycle turned a number of things on its head -- Trump was the "innovator" and HRC clearly the "maintainer". But your point is well taken - it would be so good to get back to two parties, opposed but working together towards a better outcome.

I don't think it's so clear cut that blue is left is liberal and red is right is conservative (or whatever permutation you want).

If you asked a retired Evangelical Christian whether HRC was a radical, what with her gay-love and abortions-for-all banner (I kid), I think they'd agree. But of course, they would've spotted that the "left" have signalled a stronger willingness to go in that direction and have cemented their position over the last 8 years under Obama. Does 8 years make you the establishment?

And of course Obama had helped push forward a trade policy concocted as early as the Reagan era. Does 30 years make you the establishment?

That just touches upon just two surface level ideas that fragment a person's perspective about these candidates and parties. Painting HRC as the maintainer and Trump the liberal is with too broad a brush.

So who are all those people going on about the inherent awfulness of government and how it should be replaced by private businesses?

The new nationalist/globalist split emerging/subsuming the fading conservative/liberal split is shattering this trope.

I don't see any particularly ideological reason conservatives (high conscientiousness, border and rules oriented) shouldn't be in favor of plenty of government. And good and hard, at that.

I have said before we are entering the era of post-modern politics, where values will invert (e.g. free speech) and maintaining coherent narratives will be increasingly difficult. Unfortunately, I think this means people will cling to whatever stable identity they can ever more tenaciously.

What's the demarcation of post-modern politics?

Post-modern politics can be contrasted with modern-era universalist rationalism: identity politics become dominant, inter-temporal logical consistency becomes less important, inter-group dialog becomes devalued, power structures become interesting for their own sake, etc.

A stepwise abandonment of enlightenment ideals for raw inter-group Foucauldian power plays.

Conservatives, of course!

Oh wait, now we've blown up the whole premise of innovation vs maintenance as regards left vs right. Hmmmmmm.

There is more meat to the argument than you are giving credit. Peterson has done quite a bit of psychological research on ideological differences and the behavioral manifestations of them. Jonathan Haidt has as well. Both are worth reading and watching.

It's interesting stuff and I would recommend approaching it with an open, non-ideological mind as best you can. (Certainly I'm no saint when it comes to leaving my own biases at the door: we are all all too human.)

This may be true for the original meaning of the words but I don't see how today's conservatives are maintainers and today's liberals are innovators. It's probably closer to the opposite.

I am using Peterson's psychological definition of liberal and conservative temperaments, which I think holds regardless of the current manifestations of (and abuses of) the terms. He has theorized that the core political, and in fact philosophical and perhaps even more deeply, religious, battle in the west is between people who think there should be borders (physical, e.g. between nations, conceptual, e.g. between sexes, etc.) and those who do not, which revolves around these two psychological axes.

Interesting to think about.

This makes more sense to me.

So-called "Conservatives" have been much more "egg-breakers" for the past 30 years - they've gotten more done, and most of that results in a mess (two examples - Reagan: outspent Russia, ballooned national debt/deficit, Bush2: started 2nd Iraq war - we're still paying for it)

If you remove politics from your point, I'll rephrase as "we need both maintainers and innovators" but that doesn't get the page-hits, does it?

OT, but this site looks just great with Javascript turned off (as I usually do with all the "trendy"-looking longreads, as they tend to be processor hogs). Even animations on the title screen. Awesome front-end job.

Does Edisons 1% inspiration (innovation) and 99% perspiration (maintenance) apply?

If maintaining these things is important, might we have to wonder how they came to be?

Maintainers make the world go round, innovators MAKE the world.

Meh, if innovators didn't, somebody else would.

As the frontier of knowledge expands, increasingly more low-hanging fruit appears on the horizon.

> Meh, if innovators didn't, somebody else would

And then this somebody would be an innovator.

Making the label near meaningless.

Innovation is inevitable; so too are people claiming credit for it.

Honestly, we need a better label for "anti-innovator" (conservative is taken)—they have an oversized personal impact on the world.

If the maintainers didn't make the world go around, somebody else would.

Either way those that would have to fill the spots would likely not be as competent in the doing.

Without maintenance things decline. Progress is exciting, decay banal. A recent example is the Oroville Dam spillway.

I just have to laugh at this title. It's delusional.

And what about devops? With OpenStack and et al, aren't gurus telling us that maintainance (adminitración) doesn't exist no more?

HAHAHAHAHA, it hurts. OpenStack. It needs maintenance. A lot. Maybe that's what you were after and I missed it, but OpenStack ONLY works if you have an in-house mechanic or 5. If you have less than an FTE to throw at ops, use ESX or qemu/KVM.

How exactly does that follow?

I'm a "devops engineer", and fully half of the concern around what I do is "how do we make this maintainable and know when maintenance is necessary?".

The death of devops/sysadmins/infrastructure engineers has been greatly exaggerated within the startup scene.

Agreed, though I think it somewhat depends on your definition. Most sysadmins at my clients have been uneasy at the pervasive sort of automation that we like; a lot of it is much more dev-heavy than a "modern" sysadmin (not your graybearded Perl wizard) seems comfortable with. Everybody else? Business is pretty great.

That sentiment has actually been stated? I honestly thought it was on the uptrend...

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact