Subj : Ubuntu replaces core
To   : Ky Moffet
From : Barry Martin
Date : Tue Nov 04 2025 07:31:00


Hi Ky!

 >   KM> Yeah, obviously. But it wasn't to keep schedule, it was whiney
 >   KM> brats going "You can't make me!" They've set their hearts on
 >   KM> swapping out C/C++ for Rust, and when numerous actual programmers
 >   KM> pointed out the problems with Rust, the whiney brats went "DO IT
 >   KM> NOW!" and the question of ready or not went out the window.
 > Not knowing who the 'whiney brats' are I'd guess they're powerful
 > enough to be able to chop heads but as with a lot of leaders don't know
 > diddly-squat about how the stuff they're overseeing actually works.  IMO
 KM> Oh, it's not ignorance.... I'm starting to think it's malice
 KM> aforethought.
 
It almost would seem so: someone (could be a group -- probably is)
forcing through the usage of Rust.  Rust itself might not be the problem
but rather the conversion to Rust.  (Gee,this almost sounds like X11 and
Wayland!)  How the incomplete product got released seems to be the big 
question, but we have seen this sort of instance all over, not just 
software.  Middle management tells upper management "it'll work and make
us money!".  Upper management looks at the one prototype -- "looks 
good!".  


 > good to try something new but then also needs to be tested and found to
 > work.

 KM> It gets worse.

 KM> https://lunduke.substack.com/p/debian-adding-hard-dependency-on

 KM> As I comment under the vid:
 KM> ===
 KM> Watch what Canonical does next, and see if your thoughts go where
 KM> mine did:

 KM> My cynical little voice wonders if the real "unintended
 KM> consequence" is to (for all practical purposes) kill off all the
 KM> projects downstream from both Ubuntu .... Ubuntu being the only
 KM> distro that has both jumped on this bandwagon *and* has the
 KM> resources to work past the problems Rust will cause.... and is
 KM> for all practical purposes Commercial Debian. ===

I could see both possibilities.  I'm leaning more towards the 'resources
to fix' option as it seems there are a lot of Ubuntu users out there.  
Would not be good to annoy a ton of users, but if they could somehow 
restrict the annoyance.  The 'restict' means something like run 
everything under the old/it works way and slowly move (and so test) a
utility.  Let's say test Pithos (plays Pandora, the music stream).  If
Pithos doesn't work it's going to be quiet here but I can easily get
around the error (access via a browser).  The Rust people would know
(how, I don't know -- quite sure there are ways to monitor without
grabbing too much personal data) it didn't work.

Quite sure super-overly simplified.  Ideas start somewhere.


 >   KM> The theory is that Rust is "memory safe" by default, but turns
 >   KM> out you still need to have programmers who actually know what
 >   KM> they're doing, and Rust, being the cool kid on the block, has
 >   KM> attracted a lot of rather novice hands at the task, and it wasn't
 >   KM> just a matter of swapping out functions.
 > Never is.  There are always basic steps one must know.
 KM> Yeah, and with foundational utilities, all the thousands of
 KM> time-tested interactions that suddenly are bent sideways and ALL
 KM> need to be tested again, to make sure one or the other hasn't
 KM> tickled some show-stopper bug.

Whis is sort of the basis of my Pithos example: start small and
restricted, and probably would be a good idea to start at the beginning,
or at least what is thought to be the beginning -- will find out!  If 
this test is made to work, great, and go on to the next.  If not, well, 
only one thing went down and reverse it to get it working and try to 
figure out what went wrong.  

Going to give a tangent.  One of the TVs here is a Vizio (brand). 
Approximately October 30 we couldn't receive local stations: just
'spinning' (process loader spinner icon).  All the other inputs worked
fine.  Ended up watching local TV though the MythTV input.  (This is
starting to sound like my Pithos example!)  Movies and other Internet
accesses worked fine.

October 31 still down, or at least first thing in the morning as didn't 
watch live TV the evening because of Halloween.  November 1st all working
properly again.

So apparently they did an update, it didn't go quite as expected, maybe 
tried a few other options, of which one worked or they rolled back to 
what did.

I'm thinking the Rust project could do something similar for live 
testing.



 > Might look like the right thing to do employment-wise to jump on the
 > RUST train as is the mode of software travel into the future, but it
 > still has to work: a train with a powerful engine sitting on shiny
 > aluminum foil wheels isn't going anywhere.
 KM> That's pretty much where it's at. People tend to either love or
 KM> hate Rust, but those who love it tend to be evangelical about it,
 KM> rather than pointing out where maybe it works better.
 
Blinded by the light!  If one gets too engrossed then one either doesn't
see the flaws or is convinced the flaws are due to something outside.



                     ¯                                 ®
                      ¯  BarryMartin3@MyMetronet.NET  ®
                     ¯                                 ®


.... Do electricians listen to AC-DC or something more current?
--- MultiMail/Win32 v0.47
 þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA
 þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
 * Origin: ILink: CFBBS | cfbbs.no-ip.com (454:1/1)

.