XPost: alt.comp.os.windows-11, comp.os.linux.advocacy, comp.sys.mac.advocacy   
   From: BD@hotmail.co.uk   
      
   On 21/11/2025 18:22, Paul wrote:   
   > On Fri, 11/21/2025 10:39 AM, Brock McNuggets wrote:   
   >   
   >> I get what you're trying to say with the "chained curation," but that   
   doesn't   
   >> really address the point I made.   
   >>   
   >> Sure, there's an upstream structure. Debian feeds Ubuntu, Ubuntu feeds Mint,   
   >> etc. That's packaging lineage, not user-facing curation. The existence of a   
   >> family tree doesn't help an average user figure out which distro they should   
   >> pick, what tradeoffs they're signing up for, or whether the maintainers of a   
   >> given project are making choices that will affect them a year down the road.   
   >>   
   >> Most of the differences aren't obvious from the outside. You have things   
   like:   
   >>   
   >> - Ubuntu leaning hard on SNAPs   
   >> - Mint avoiding them   
   >> - upstreams with conflicting philosophies   
   >> - different release cadences, different patching approaches, different   
   tooling stacks   
   >>   
   >> None of that is clear until you're already using the system, and it's not   
   >> explained in any unified, beginner-friendly way. That's the "paradox of   
   >> choice" part: plenty of options, very little guidance unless you already   
   know   
   >> the ecosystem well.   
   >>   
   >> So yeah, the distro family tree is there, but it doesn't fix the actual   
   >> user-experience problem. A car lot still gives you a salesperson, brochures,   
   >> trim levels, test drives, and a guided funnel toward a decision. Linux   
   distros   
   >> mostly give you a giant chart and tell you good luck.   
   >>   
   >> That's the gap I was pointing to.   
   >>   
   >> To be clear, this does not mean I am against Linux. I have used it myself,   
   >> have set up labs in schools, have set it up for users, and helped them set   
   it   
   >> up. I mostly used Mint. No list of distros for them. No options. Just   
   >> installed Mint or gave them media with it for them to do so. Before that I   
   was   
   >> doing the same with Ubuntu. I "curated" the choices for them.   
   >   
   > I'm not sure that we can expect a single human to do those tradeoffs   
   > in an intelligible way.   
   >   
   > Imagine if this diagram had a third dimension, with some properties listed.   
   > That would be totally unmanageable.   
   >   
   >     https://upload.wikimedia.org/wikipedia/commons/1/1b/Linux_   
   istribution_Timeline.svg   
   >   
   > I doubt even an LLM AI analysis would yield a result, as the LLM AI would   
   > succumb to the marketing text and not concentrate on the numbers or whatever.   
   > For example, when Zak Wallen writes an article, the article is   
   > the fluffy kind and not substantive. That would be an advocacy or marketing   
   > type article (where we compare the pastel shades of the background picture,   
   > as an important aspect of distro choice).   
   >   
   >     Paul   
      
   Paul,   
      
   I think you're actually making Brock's point for him here.   
      
   You're right that no single human can reasonably evaluate all those   
   tradeoffs in the timeline diagram - but that's precisely the problem   
   Brock identified. The fact that the choice space is "totally   
   unmanageable" even for knowledgeable users is exactly why newcomers   
   struggle.   
      
   Your solution - picking Mint/Ubuntu for others and skipping the decision   
   entirely - is essentially what Brock meant by needing "curation." You   
   became the missing salesperson. That works great when you're there to   
   help, but it doesn't scale to the millions of potential users who don't   
   have a knowledgeable friend to make the call for them.   
      
   The LLM comment is interesting but misses the mark a bit. The issue   
   isn't that analysis is impossible - it's that the information isn't   
   organized in a way that maps to user needs. A newcomer doesn't care   
   about "packaging lineage" or "upstreams." They care about: Will this   
   work with my hardware? Will it stay stable? Can I install the software I   
   need? Will I get updates that break things?   
      
   Those questions have answers, but they're buried in the ecosystem   
   knowledge you already have. That's the gap Brock was pointing to - not   
   that Linux is bad, but that the on-ramp is unnecessarily steep.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)