Which still seems like a short period of time, considering timelines in engineering are to support a version for 60 years.

>considering timelines in engineering are to support a version for 60 years.

In the vast minority of cases.

Not in most electrical engineering (except power plants), nor in computer engineering. 60 years ago was 1959. What software/computer project from that time is still running?

Even moving outside of the electrical domain, how many physical products outside of civil engineering is expected to last that long. I certainly can't expect support for my car for longer than 25 years.

If there's a difference it's not time scale necessarily, it's that CPAs are better at amortizing support/maintenance costs across decades. CPAs have amazingly detailed depreciation charts for non-software engineering lifecycles. If you build a civil engineering project, you figure out asset depreciation versus maintenance schedules, and you budget accordingly.

I've yet to see a CPA adequately depreciate software assets, anywhere. Maybe we should help them out by building better depreciation schedules. There seems to be a lot of CPAs that don't believe software depreciates over time, and maybe that's the largest disconnect in labeling it "tech debt", because accountants hear that term and think they can ignore it on a balance sheet they don't care about, but "tech depreciation" might actually scare them straight (until they find the tax advantages in calling it that).

They really need our help :-)

In a previous job that was mostly in an engineering/manufacturing department, but with a lot of Perl/Python automation scripts, we had an internal conference. One of the keynote talks was when not to automate using SW. It went into the cost of maintaining SW over the long term - including the fact that authors leave, and people who understand their scripts require higher salaries. Most people who write these scripts are not hired for SW roles, so their replacements likely cannot debug/extend.

Classic cars can still get parts after 25 years, but a lot of times its custom machine shop stuff. There is a market for it. Really there's also a market for restoring old computers (look at The 8-Bit Guy, LGR, The Living Computer Museum (Seattle) and all those other people who restore old hardware for fun and education.

But I get your point, those are special cases for preservation. For mainline things, especially with today's processes for continual integration, dependency checking and advance build tooling, dependency rot is something that should be accounted for in all software project plannings. If your dependencies are a few months out of date and you don't have the time to update them and re-run tests (people write tests right?) things are just going to hurt more and more later.

You can keep running Python 2 just like you keep your COBOL systems running, nobody's gonna stop you except common sense.

IBM does quite a good job of making sure that COBOL is supported long term with all needed updates for many many years to come. Python 2 is not in that position.

You mean, IBM does quite a good job of making sure IBM COBOL is supported long term. They are maintaining their compiler, which is exactly what PSF is doing. They are maintaining their interpreter, which is Python 3.

The GP post missed the fundamental difference between keeping COBOL running and keeping Python 2 running. Python 2 was also PSF's interpreter. IBM handling a COBOL upgrade like PSF handling the 2-3 transition would be unacceptable.

> IBM

> PSF

These are two very different kinds of institutions!

People who need COBOL support from IBM are paying a lot of money. Giant piles of money can get you many kinds of help that people won't volunteer to do for free... among them, maintaining ancient software in amber.

If you need Python 2 support and you are willing (and able) to pay the kind of money that IBM's customers pay for COBOL support, you'll be OK. For a start, Red Hat (aka also IBM!) shipped Python 2 in RHEL 8, which means they'll be supporting it until 2029 at the earliest.

To circle back to the original point, no COBOL committee would break commonly running COBOL programs like the Python 2-3 transition. Using COBOL in an example with Python is just wrong. Maybe the break is justified, maybe it isn't, but some languages do a lot of work to make sure things continue to work.

Would PSF exponentially increasing support and maintenance costs for Python 2 into multimillion dollar contracts and bundling over-margined hardware in with the bundle to make it more of an IBM-like transition help?

Yep, someone has to pay in one way or another like Red Hat customers on 7, but to say Python has near the life cycle of COBOL is just disingenuous. Old COBOL still runs, but Python 2 programs will not. It really shows what the achievement languages like COBOL, RPG, and Fortran are in terms of longevity and migration.

> but Python 2 programs will not

Er, why not? It's not like there's some kill switch in Python 2 that will make it stop working after January 1st, 2020. If it works now, then it'll still work, you're just not guaranteed fixes anymore. At least, not for free. As stated in the article, paid support options exist from several vendors.

Any software in the modern era that isn't upgraded for new OSes or security patched is dead.

Ok, sure. To you, maybe. But that's a far cry from saying they won't run anymore.

You can still run 1960's code today for both COBOL and Fortran - I suspect some Fortan 2 oddities might not work I am thinking of the sense commands

Probably, but the care the Fortran folks take to not break anything but still evolve a modern language is amazing.

Right, but in engineering it's kind of expected to get support for a version for at least 60 years. Software engineering is just really weird in that it moves so fast and nobody seems to care to break things.

They (in engineering general) aren't expected support something free though. Putting it another way: You can get you support with Python2, if you pay.

What free things are supported for 60 years in engineering applications?

Name one.

Fairly ancient Fortran compiles OK in open source compilers. Not quite 60 years old, yet, but it'll be there soon.

It's already >65 years old. The 65th birthday was this summer 2019.

We sell welding systems to weld stainless steal, copper pipes, etc. We always give a warranty of 24 month, guarantee paid support for 10 years, and support older machines only if possible. I am not sure which industry you are talking about, but 60 years is the exception in my experience.