I have a Rigol and an Agilent over 10 years old now and going strong. Honestly though, my Agilent overheated once and destroyed its own SMPS. This was terrifying given the price of the scope (and at the moment, just knowing -- it was dead and smelled like an electronics fire). I now have a circulating fan on my scope shelf moving air behind and across all my bench equipment.
I replaced that Agilent SMPS for $150? and then got a 2 year service plan for another couple hundred dollars.
The latest Rigol prices are pretty expendable, $299!! but I suspect they will be running for quite a few years.
Also like I'm not sure? A lot of the discrete component and analog stuff on the old models is way more flimsy that the new system on a chip things on the signal path of the new instruments, specially for RF.
Plus if you need data logging it's not even worth to have the discussion on if older gear was better.
Unless you need traceability for contract or regulatory reasons, oscilloscopes don't need regular calibration. Most of the time, you're probably using them for rough visual confirmations (e.g. is this signal behaving like I expect it to) more than precise measurements. In any case, it's rare for an oscilloscope to go seriously out of whack without a physical insult (like being dropped or exposed to major overvoltage).
My main problem with the old analog scopes is that the controls get flakey over time. Also, I do like the measurement and data logging features.