Thus, bugs and quality are secondary. Why spend $ on engineering and fixing bugs when you can spend that on marketing for much better ROI.
The trade off skills apply outside the engineering domain as well.
I don't think a blanket put-down of all marketing as negative value is really true.
Of course there are instances where marketing is dishonest or harmful, but by and large, how do you expect people to discover goods and services if there isn't someone who's willing to pay for the cost of dissemination of information, whether it's a marketer, or Google's search algorithm, or the consumer?
Marketing is by and large about gaming the way people select goods and services, so that you can sell more regardless of the quality of your offerings, or of whether the client actually needs it.
That may be the objective function of the marketer or salesperson, but it doesn't follow that marketing itself is a simple net transfer of wealth from consumers to producers.
> I think bringing up marketing-as-information-dissemination is a red herring, because it's only a small fraction of the whole thing.
The marketer is not paying to move bits to a consumer, the marketer is paying for the consumer's attention, which is much more limited than our ability to move bits around.
Do you have an alternative fair and equitable way of allocating that scarce resource? For example, if we went with your suggestion and restricted marketing to only classifieds and other forms of print advertising, then we would expect the cost of placing ads there to skyrocket as everyone competes for scarce space. Also, the same risks of manipulation and dishonesty apply just as well there:
https://www.google.com/search?q=snake+oil+ad&espv=2&biw=1440...
Yes. Not allocating it. What you call trying to "allocate" consumer's attention is, from the point of view of a consumer, people purposefully attempting to distract you from living your life in order to extract money from you. In all other contexts of life we would call such people "assholes". Classifieds have the benefit of consumers paying attention only when they want to pay attention.
Time is the most precious thing we have in life. Attempts to take it away from other people should be treated more seriously, IMO. It's never just one marketer - they're competing for attention just as they would compete for classified space.
I'm not sure what you mean by not allocating it.
Even if we restricted all advertising to classifieds, we still need to decide whose ads get shown, how many pages they get, etc.. That's what I mean by allocation.
And that would still be subject to the problems you identified: the advertiser might be dishonest, they might not have the consumer's best interests at heart, etc.
As for how to achieve this objective--have the government rigorously define advertising and ban it in all forms? That just shifts power from advertisers to bureaucrats then, with all the risks of regulatory capture that implies.
Let me pose the question in a different way: let's say there are a hundred different kinds of toothpaste. There is a cost to figuring out which kind is the best kind(s), and then letting everyone know the ordering. Who does the work of figuring that out, and then how do they get compensated for that work?
Advertising and marketing is one possible answer by creating a signalling equilibrium (https://en.wikipedia.org/wiki/Signalling_(economics)#Introdu...): all else being equal, the better toothpaste should have access to a larger marketing budget. If consumers know this, then seeing someone's ad is actually a (perhaps weak) signal of quality.
There are also other models, like Amazon reviews; but there are plenty of ways to game that system, and you're really depending on Amazon to police it effectively.
Or you can actually pay someone to review products, like the Consumer Reports model, where you only get access if you pay up.
But each of these models has a tradeoff; Consumer Reports works best when everyone subscribes, but then as it becomes more popular, producers will be tempted to buy them off; Amazon reviews are subject to bandwagoning and cheating; classic marketing is expensive (in time and cost) and unreliable.
Now obviously, humans being humans, there's no way to prevent companies from trying to influence that word of mouth mechanism. But there seems to be a benefit from keeping it a shady (or perhaps illegal) practice - the society in general will shun corporate shills and attempts at influencing people's opinions. Which seems to be a better option than what we have now, where marketing is actually a respectable occupation, and people get assaulted with advertisements.
> all else being equal, the better toothpaste should have access to a larger marketing budget
This doesn't work because the definition of "better" for the market is different than "better" for the consumer. A market!better product is the one that sells more than alternatives, while the consumer!better product is the one of greater quality/usefulness. Market!better gives you profits which increases your marketing budget which increases market!better - you have a positive feedback loop here that bypasses the consumer!better variable entirely - hence the observation that investing in marketing has better ROI today than making something that's actually useful for the consumer.
> This doesn't work because the definition of "better" for the market is different than "better" for the consumer.
I think these two points of view are incompatible.
The first quote implies that consumers can distinguish between two toothpastes, and moreover, they prefer the better one by being willing to pay more for it.
If that's the case, then given two initially otherwise identical companies, the one making the better toothpaste will have more sales, and then a larger marketing budget, etc. -- basically, the initial condition of producing a better toothpaste, combined with the positive marketing-sales feedback loop that you identify, leads to the better product winning.
Of course, this only works if consumers can tell which product is better, and there might be other factors that prevent product quality from being a strong enough initial condition (e.g. it's really hard to displace incumbents even with a better product).
> the society in general will shun corporate shills and attempts at influencing people's opinions. Which seems to be a better option than what we have now, where marketing is actually a respectable occupation, and people get assaulted with advertisements.
People who use aggressive tactics like cold-calling usually get paid minimum wage precisely because the signal-to-noise ratio is so low. In a sense, the market already strongly discourages blindly blasting people with noise.
Compare to the success of Adwords--advertisers only want to show ads to people who will be interested. The tradeoff here, of course, is reduced privacy for the consumer.
Engineering tradeoffs are very much about costs. I could (well once upon a time I could :-) design a bridge that's stable up to a (not very) nearby atomic blast; more stable than the bedrock it's built on. But that would require an unnecessary level of resources (unless atomic resistance is in fact a design requirement). I could design the most super-awesome web site that scales to 40 billion unique visitors...but it's unlikely to be needed (though if some of the uniques are robotic, perhaps it is). Simply choosing an alloy that's simpler to work with the tools on hand is a rescue tradeoff, and those resources are time and money.
Likewise, the plane (I assume you mean airplane) has only one valuable purpose (transport people or goods that someone wants transported) and that is denominated in money. Make it cheaper and more people can fly, or pay less for their gadgets or fresh shellfish.
It's incredibly naive to to say that marketing is some zero-sum game. If you have an amazing bandage that is cheap and saves lives you still have to get the word out.
The core topic in this subthread is focusing work on marketing the bandage instead of making it cheaper/save more lives, because marketing has much better ROI than actually providing value to people.
And the word can get out fine by itself, actually. The word of mouth is a pretty effective thing - and a lot of marketing is trying to game it, getting increased profits by destroying its usefulness.
1) Person A makes a crappy bandage (Y units of quality) with X units of budget and markets with 9X units of budget
2) Person B makes a decent bandage (5Y units of quality) with 5X units of budget.
If person B were to also market the bandage with 5X units of budget, they will take a larger fraction of person A's market share than if they don't, thus providing more value to people. Yes you can argue that if person A didn't exist, then they wouldn't need to spend this money and the bandage could be cheaper, but this is clearly an example where adding marginal marketing can be a net benefit to society.
> And the word can get out fine by itself, actually.
Umm, good luck with that. There are only a few, small, specialized businesses for which that is true.
In some situations, it is about 1 or 0, because the output of that continuous function at a certain time t is put through a thresholding function: either buy or don't.
I'm simply pointing out that it's more accurate to look at the world in terms of degrees.
"never" and "always" have, I believe, a high degree of built-in denial.
Companies that don't sell software, see it as a by-product and have zero worries about its quality, what matters is that at the end of the it delivers the outputs the actual business requires.
So beware of presenting prototypes in such industries, their are always the final product and code refactoring is not something that usually makes them sell more stuff from their actual products.
Also the code tends to be pretty horrible, because a lot is developed by external companies that usually tend to implement a just single set of features. It is almost impossible to write properly safe code, because no external is able to grasp the whole codebase in the required timeframes for delivering the new feature and leave to another customer.
And increasingly, because of the 'go fast a break things' attitude, this is extending to companies that _do_ sell software.
That said, as someone who highly values a well built product, this may be wishful thinking.
But the reality is that companies are getting pwned over and over and nothing bad happens to them. Actually looking at cost-benefit it still makes sense to get to market quickly.
What is the argument? If you start thinking about it, I think you'll find it's actually quite hard to create a coherent counter-argument.
Of course, as we all know, it never happens. BigCorp continues to churn out buggy, leaky, unoptimized code, and customers keep giving them money for it. I suppose you can handle this in a number of ways: 1. You can accept this and re-adjust what you consider to be "good enough to ship", 2. You can get jaded and dismiss all software customers as morons, or 3. You can start your own company that TAKES SERIOUSLY the proper use of namespaces, and see if that lets you outcompete the competition.
They just don't care.
Engineering quality really doesn't mean much to people, so stop worrying so much about it once the core works without killing the user.
Often, that requires the technical skills to reliably solve their problems.
Always, that requires the communication skills to pursuade someone you can solve their problems.
I think one way to do both of these is market your product well, yet at the same time develop it in a sustainable way so that this piece of crap is becoming something that eventually looks great on the engineering side of things.
To use the ever-present car analogy, shoddy engineering didn't harm British Leyland - until it did. And now there's no native British car industry to speak of.
But people don't care about engineering - they care about how your product will affect their lives today. If you can prove to them that it will solve that problem they have, they don't care that it's made out of biscuits.
Engineering or rather sometimes over-engineering things might make it super easy but not intuitive to use.
think of the extremes; you could develop a technical masterpiece and keep it under covers, making $0, having spent every single bit of available time/money/resources on it.
that's a hobby, not a business.