What does this have to do with "Art pieces"? In the real world we know that what you are saying just isnt true. You can substantially upgrade capability with the same airframe, or the same chassis or the same ship hull. Weapons upgrades, sensor upgrades, avionics upgrades, computer upgrades, communication and connectivity upgrades, defense system upgrades, engine upgrades, range, etc.
That's all true TODAY. It wouldnt be true in the 23rd or 24th century? This is silly.
Indeed, I've often found this viewpoint a bit strange myself. In the context of how some Trek technology works, there's no reason to waste resources on the next major ship classes if the goal isn't to upgrade them as needed. I personally have no huge issue with seeing Excelsiors and Mirandas, along with Constellations, still active in the TNG+ era, and I think it's far more likely that Starfleet would upgrade them as needed than the Klingon Defense Force might for their older ships.
The TNG Technical Manual suggests the Galaxy class is intended for a potential lifespan of up to a century, with eventual refits.
There is of course a logical cutoff point to when older designs might be less ideal to keep refitted, of course, but that's not inherently the same as older technology being stagnant. How many of us today drive cars that are a decade or more old, because they're still reliable and much cheaper to maintain than the cost of a new modern car?
One of my favorite non-Trek examples, Battletech, has a lot of examples of designs being built for centuries, and sometimes individual examples of those designs being maintained for that length of time by successive generations. The
Crusader heavy mech is one such unit, now nearing the 400th year of overall production (the game's current timeline is now in the 3150s, and the Crusader first appeared in 2752). There's more than a dozen variants in the series across the centuries, not counting how individual pilots have modified their Crusaders as a result of damage or personal preferences.
It's generally less common for individual units to survive for a relatively long time, versus the total length of time that the class has been in service, since it's a military RPG with lots of stuff blowing up.
Capital ships and warships are the interesting exception, as the huge expense in building and maintaining them means it's more likely a warship dating back to the time of the First Star League is a few centuries old. The Clan navies were originally made up of many such vessels from the era of the Exodus, when the bulk of the Star League Defense Force jumped into uncharted space. Warships became a lost technology during the Succession Wars and even civilian jumpships - the only means of interstellar travel - became more rare, so it was an informal rule not to target them or else lose any form of travel. The technology has since been regained in the Inner Sphere but the writers have tended to get rid of large fleets for plot reasons. They want to keep the traditional focus on ground combat.
The Clans also made a number of impressive engineering breakthroughs after the Exodus kind of fell apart (the SLDF had hoped to restart the League elsewhere but suffered their own civil war, leading to the rise of the Clans), in particular the ability to create omnimechs and vehicles with equipment bays that entirely modular and interchangeable. This means with an omnimech, like the
Nova, you can have a single chassis that can handle a wide variety of mission profiles, as well as easing repairs considerably. The Clans also use this to a limited degree for other units like fighters and tanks, but they traditionally regard non-mech forces as being inherently "inferior" despite how well that unit actually performs.
Despite its many advantages, the omni technology has limitations as well. It's considerably more expensive than producing a more standard mech design, and components between them are not compatible even if the omni version is functionally much the same parts. During the time of the original League, vehicle designs were typically built around specific mission profiles with the occasional later variants, so the flexibility realized by the Clans wouldn't have been seen as essential. It also required extensive work to get the original omni designs to become practical, due to the engineering challenges of building a computer that could manage multiple configurations and changes without potentially going haywire. The Clans still use a large percentage of standard units as second line and reserve forces, and some are more open minded about the use of combined arms versus strictly mechs.
