The Stack Fallacy (2016) | Hacker News Reader

The Stack Fallacy (2016)

techcrunch.com

66 pointstomasz_wro28 comments

"The bottleneck for success often is not knowledge of the tools, but lack of understanding of the customer needs."

This is an issue I constantly encounter. Engineers are terrific problem solvers, but the problems to be solved are often not communicated effectively to the engineers capable to solve them.

Often the customer does not understand their real needs either, and if they don't properly understand their needs it can be hard to communicate to the engineers.

Also, something I have had happen to me is the use of words can have different meanings to different people. So the customer can think they have clearly described their problem but the person listen to them may think they want something different.

That's why showing customers actual software quickly, getting feedback directly from them and iterating rapidly on that feedback is so incredibly important.

No need for actual software!

A rather crude mock-up which shows the logic, the steps, the data involved, when discussed with the real prospective users, helps immediately find gaping voids in understanding, both to the developers and to the customers themselves. "Hmm, we never thought about it" is a rather frequent reaction.

If somebody ever wondered what product people do, this is it. They research the needs of the customer, and research, discuss, and achieve agreement on ways to solve them.

This is exactly why I love Example Mapping.

https://cucumber.io/blog/bdd/example-mapping-introduction/

From the site:

> By continuing to browse, you consent to our use of cookies

That's not how the GDPR works. [0] Here's an alternative summary. [1]

[0] https://ico.org.uk/for-organisations/guide-to-pecr/cookies-a...

[1] https://automationpanda.com/2018/02/27/bdd-example-mapping/

better to do the right thing poorly than the wrong thing well

100%

except for one thing. You have to be careful about building things "poorly". There will always be someone who will judge your engineering capabilities by your "poor prototypes". Usually someone non technical from management or waterfall thinking colleagues with mindset of releasing only when it doesn't embarass them.

Or people whose careers have been a train of following behind people who built the right thing poorly, made a quick buck and moved on.

ok, so "do the right thing mvp, ..." :)

Stack Fallacy - it seems it's easier to innovate up the stack, because you know the building blocks, because you own them (e.g. Apple doing apps). But it's actually easier to innovate down the stack, because you know the customer needs, because you are the customer (e.g. Apple doing chips).

Are there examples of company who have successfully gone up the stack?

Microsoft?

They started with the OS, branched to languages, added Office (against much better competitors), started to control the browser and then antitrust happened.

Actually, they started with languages, specifically BASIC, and then FORTRAN.

Also, their gaming industry story. They used to be just an OS that games ran on, then released one of the leading consoles and now are one of the major publishers and studio owners.

Also some of their hardware. For example lots of people love their mouse. On the other hand I know people who hate the Surface Pro.

As part of the commoditization of the pc, in the eighties and nineties, Asian pc hardware vendors went from basic manufacturing to simple assembly to motherboard layout to full system design.

They basically went from subcontractors for Dell and the like, to almost completely replacing Dell and the line by moving up in the chain until Dell had very little value left to add.

Red Hat? From a mass market Linux to enterprise Linux maintenance to middleware (jboss) And from hardware enablement to virtualization to lightweight virtualization, cloud management and container orchestration?

On reflection I think the BEST possible example must be Pixar. They are a shining example of going up the stack in a phenomenal way. They started with graphics rendering hardware:

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

And then they wanted to sell software:

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

And then they ended up making movies! This is described in Ed Catmull's book and I think there's a documentary on Netflix too.

---

I also think it's interesting that George Lucas started out as a special effects guy and then he made Star Wars. So the effects really do help the storytelling.

Like Amazon going from selling books to selling compute and network and warehouses and shipping.

Amazon is different,as all they sell is what they learned throughout the years: infrastructure.then they learn more things and start selling solutions to those problems.

and whenever they get into a new space like tablets, grocery delivery, or game dev it mostly doesn't go well... Alexa is probably the exception

Hm interesting... so if you consider Google's original products search and ads, then Android, Chrome, ChromeBooks and other hardware are down the stack. Google Cloud is down the stack.

YouTube is lateral, but it was an acquisition. Minor business, but being a DNS registrar is down the stack. So it seems that Google did go down mostly.

I would say Apple went "up" in at least one phenomenal way. They started out with computers, and then 20 years later their big success was iPod + iTunes. iTunes is probably 2 levels up the stack, being mostly a media business, not really software even.

Apple also wrote a lot of apps like Keynote, and acquired apps like Logic, etc.

From it's inception Apple made technology useable.

The Apple ][ was an appliance, the Mac even more so. That was happening during an era where a lot of computers still shipped in kits.

The iPod was a better UX on top of existing mp3 players Asians OEM were making (5 buttons and a tiny liquid crystal screen).

The iTunes store was Jobs spending an afternoon downloading music on the internet and figuring out he was saving less than minimum wage doing so.

The common theme when moving up the 'stack' (whether it be technology or academic disciplines as in the graphic) is that complex systems (whether OS or biology) are not just a sum of their parts and have emergent properties you cannot predict just by looking at the building blocks.

Lots of coverage on technical aspects but what's often missing is the sales and marketing functions and their ability to pull the whole thing out of water. They mentioned Salesforce and Oracle. Salesforce was pushing sexy company image from day one, while Oracle has a reputation of a death star. Salesforce community screams off rooftops how great it is, while Oracles customers fighting their legal teams. Could more companies build successful CRM system? Yes, and many did, but I can't recall MS Dynamics fans walking out of the conferences with big smiles over their face as they'd just won the lottery. It's not necessary hard to build the product, but how it gets pushed to the market is what makes or breaks it. Google cloud,anyone?

https://archive.fo/KsD50

Reply on news.ycombinator.com