[HN Gopher] The Stack Fallacy (2016)
       ___________________________________________________________________
        
       The Stack Fallacy (2016)
        
       Author : tomasz_wro
       Score  : 60 points
       Date   : 2021-02-18 09:10 UTC (1 days ago)
        
 (HTM) web link (techcrunch.com)
 (TXT) w3m dump (techcrunch.com)
        
       | kragen wrote:
       | https://archive.fo/KsD50
        
       | tomasz_wro wrote:
       | 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).
        
         | debarshri wrote:
         | Are there examples of company who have successfully gone up the
         | stack?
        
           | froh wrote:
           | 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?
        
           | Areading314 wrote:
           | Microsoft?
        
           | compiler-guy wrote:
           | 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.
        
           | alex_young wrote:
           | Microsoft?
           | 
           | They started with the OS, branched to languages, added Office
           | (against much better competitors), started to control the
           | browser and then antitrust happened.
        
             | zabzonk wrote:
             | Actually, they started with languages, specifically BASIC,
             | and then FORTRAN.
        
             | golergka wrote:
             | 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.
        
               | ecpottinger wrote:
               | 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.
        
         | lanstin wrote:
         | Like Amazon going from selling books to selling compute and
         | network and warehouses and shipping.
        
           | cosmodisk wrote:
           | 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.
        
         | chubot wrote:
         | 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.
        
           | 908B64B197 wrote:
           | 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.
        
       | cosmodisk wrote:
       | 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?
        
       | francisofascii wrote:
       | "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.
        
         | m463 wrote:
         | better to do the right thing poorly than the wrong thing well
        
           | snidane wrote:
           | 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.
        
             | bluefirebrand wrote:
             | 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.
        
         | ecpottinger wrote:
         | 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.
        
           | mpweiher wrote:
           | That's why showing customers actual software quickly, getting
           | feedback _directly_ from them and iterating rapidly on that
           | feedback is so incredibly important.
        
             | nine_k wrote:
             | 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.
        
           | aslakhellesoy wrote:
           | This is exactly why I love Example Mapping.
           | 
           | https://cucumber.io/blog/bdd/example-mapping-introduction/
        
             | MaxBarraclough wrote:
             | 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/
        
         | Person5478 wrote:
         | This is exactly why I say for best results, let your engineers
         | and "customers" talk directly to each other.
        
       | AAmarkov wrote:
       | 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.
        
       ___________________________________________________________________
       (page generated 2021-02-19 23:01 UTC)