[HN Gopher] New software sells new hardware, but not forever
       ___________________________________________________________________
        
       New software sells new hardware, but not forever
        
       Author : redbell
       Score  : 46 points
       Date   : 2023-01-11 13:34 UTC (1 days ago)
        
 (HTM) web link (www.theregister.com)
 (TXT) w3m dump (www.theregister.com)
        
       | wintermutestwin wrote:
       | I think that anyone interested in HW<>SW product models should
       | look to this amazing company as a case study that flies counter
       | to current "wisdom:"
       | 
       | https://www.fractalaudio.com/
       | 
       | They make the highest end guitar audio processing unit and are
       | constantly providing free SW updates that greatly enhance the
       | unit. No subscription BS.
        
       | sirjaz wrote:
       | If you want to sell more hw actually sell software or more
       | specifically allow software to be local to the device. We have
       | plenty of space and we need to utilize that hw.
        
       | bitwize wrote:
       | And this is why you build checks for recent model CPUs into your
       | software, to keep the treadmill going.
        
       | 1vuio0pswjnm7 wrote:
       | "If it is not already obvious, these are somewhat conflicting
       | goals. The people paying for the bulk of the work are not the
       | most numerous consumers of it, and the desires of the billions of
       | end-users are to a degree opposed to those of the people paying
       | developers to work on it."
       | 
       | Having projects become dependent on commercial entities is not
       | always a good thing for everyone.
       | 
       | Those getting paid by corporations to work on FOSS might argue
       | it's a good thing.
       | 
       | The corporations paying them might argue it's a good thing.
       | 
       | The vendors to those corporations might argue it's a good thing.
       | 
       | Developers relying on FOSS for commercial purposes might argue
       | it's a good thing.
       | 
       | Tech journalists and bloggers might believe it's a good thing
       | based on what they are told.
       | 
       | But what about end users. What is their self-interest.
       | 
       | It is possible that for some this may not be a good thing.
       | 
       | Thus, for end users' benefit, some non-commercial projects are
       | needed.
       | 
       | When people online start arguing that FOSS projects need
       | corporate support maybe not everyone reading thinks, "Yeah, that
       | would be a good thing."
       | 
       | "For a while, things like NetBSD will come to the rescue, but it
       | can't do so forever."
       | 
       | Forever is a long time.
        
       | scotty79 wrote:
       | This reads like incoherent rambling around one thought without
       | ever finding any justification for this thought.
       | 
       | There are exactly same dynamics between hardware and software in
       | free as in proprietary. The only difference is that free software
       | is not sold directly. But it still is created and still acquires
       | new features that need better hardware.
        
         | sbierwagen wrote:
         | >This reads like incoherent rambling around one thought without
         | ever finding any justification for this thought.
         | 
         | That is the house style of The Register, yes.
        
         | AtlasBarfed wrote:
         | I think the overarching point, if there is one, is that with
         | Moore's law and Dennard scaling running out of real-world
         | impact between generations, the churn of new hardware is
         | becoming more forced-consumption / deck chair rearrangement on
         | the software side rather than delivering value / efficiency
         | that utilizes new hardware capabilities.
         | 
         | The software industry has been slowing down with worse and
         | worse runtimes, It used to basically be compiles C/C++ type
         | stuff in the 80s/90s, 2000-2010s was JVM/dotNET, and now it is
         | Javascript, although with Typescript the industry can probably
         | get VM performance back to the JVM days. (very coarse
         | paintbrush there of course).
         | 
         | At some point the industry will need to wring out optimization
         | from the software stack to get closer to hardware again. The
         | good news is that hopefully dependency churn will reduce,
         | because interfaces may finally become more stable, as will
         | implementations. Because ye gods is dependency churn a headache
         | these days. As I get older it just gets more annoying.
        
         | livrem wrote:
         | I did not really understand what they were trying to say, but I
         | think the dynamics for free (especially non-commercial,
         | community-driven projects) vs non-free are different The former
         | do not have the same strong need to keep adding stuff beyond
         | what makes sense. I am increasingly happy to see small tools
         | that does what they need to do and can almost immediately move
         | to maintenance-mode forever and just provide important bug-
         | fixes and every now and then add support for new platforms. But
         | what about progress? A fork can do that. Or some competing free
         | tool.
         | 
         | Also the link to the article by Wirth at the end of the
         | article, A Plea for Lean Software, was great and I never read
         | that before, so I am happy that The Register linked to it
         | (https://cr.yp.to/bib/1995/wirth.pdf).
        
       | twobitshifter wrote:
       | It's interesting that he notes the power of a core2duo being
       | adequate. I ran a core2duo laptop for 10 years! When I first got
       | the laptop I was running 3d compositing desktops with compiz. By
       | the end of it's life I was running xfce to make it usable. Same
       | hardware but the software changed to the point where the computer
       | fell behind.
        
         | pjmlp wrote:
         | My Asus 1215B (from 2009) also seems to have reached its end,
         | even with XFCE it requires quite some patience and being quite
         | spartan in open windows.
        
       | CyberDildonics wrote:
       | I think this might have been seen as insightful a few decades
       | ago.
        
         | criddell wrote:
         | It's still insightful today. Look at all the great applications
         | out there that can't simply be "done" and go into maintenance
         | mode. Instead, there's a need to continue to grow so features
         | are added and it seems that it's almost inevitable that
         | original fans leave and the software gets fatter and slower and
         | less fun to use.
         | 
         | Evernote is the classic example and I think Obsidian is headed
         | down that same path starting with the addition of canvases.
         | There are exceptions though. Photoshop is still pretty relevant
         | and it's been growing for decades now.
        
           | CyberDildonics wrote:
           | I hate bloated software too, but I don't understand how this
           | connects to this post.
        
             | criddell wrote:
             | The part about software being a gas and expanding to fill
             | whatever container you put it in.
        
           | ilyt wrote:
           | The _software_ doesn 't, not for probably 70%+ of user base.
           | The company needs to, because they need to have excuse for
           | selling new versions.
           | 
           | How you excuse selling new version if users are using same
           | old features and their use case doesn't need the new niche
           | gadget new version is adding ?
           | 
           | (you don't and just sell app as SaaS or other subscription so
           | everyone have to pay you monthly)
           | 
           | That's why I think we see the occasional "redesigns for sake
           | of redesigns", designers need to excuse their employment so
           | they make "new" ui, don't even test whether current users
           | area actually faster doing stuff the "new" way, copy some
           | recent design trend and now you have app with "new" UI
           | everyone actually using for longer periods needs to waste
           | time to relearn only to get to same (or worse) efficiency.
        
       ___________________________________________________________________
       (page generated 2023-01-12 23:01 UTC)