What do you have against MUMPS (i.e. ANSI M)? It was a great technology of its time, and there are still many critical applications running with it today, and internationally recognized companies like Intersystems whose products still run ANSI M inside.
back
2 comments
> It was a great technology of its time
It was. And then that time came and went and after that it became harder and harder to justify its continued existence and development of applications.
I've worked on a moderately complex mumps system for a bank and the amount of arcane trickery to work around the limitations of the core MUMPS system were mind-boggling. The whole thing had to interact with other systems via trick comms links and the fact that it worked at all was a small miracle.
There are some other systems that are still in use even today that should have probably never been built in FoxPRO, Filemaker and other pieces of software of a certain age.
Well, some MUMPS implementations had adverse limitations, but there were also others, and companies like Intersystems managed to gradually expand and continue using the technology successfully. What becaume know as "NoSQL" was already available in MUMPS 40 years earlier. MUMPS as a language has many disadvantages, but the concepts are still amazing today.
NoSQL databases came before SQL, irrespective of MUMPS. But back then they just called them databases.
Read the link.
Companies that develop their own idiosyncratic dev systems are generally making a mistake. Developers who endeavor to master those systems are using energy and time to add a skill to their resume that has zero value on the open market, which is a bad career move.
I read it. The article is superficial and one-sided. And there is nothing wrong with companies building their own tools. There are many such companies, including SAP, Microsoft and Google.
Most companies are not SAP, Microsoft or Google, and would do well to stay miles away from developing their own tools. Because the tools that you develop then become the millstone around your neck in the longer term.
Unless a tool gives you a major advantage you should never go near that. Your competitors will all happily use COTS, drop projects and adapt new tech while you will be stuck in the hole with your sunk cost and impossible to migrate toolchain. That sort of situation has sunk more than one company.
AND more than one career, too.
Folks who invest their careers in single-company toolchains tend not to have very remunerative or rewarding careers. It creates a monstrous imbalance between employer and employee, too -- all the "lifers" who stayed at TCK were being paid well below market, but couldn't command better elsewhere because their whole professional life had been spent on single-site tools.
That's MUMPS for you.
That's an excellent point. Working on a single company toolchain when that company isn't a household name is a career dead end.
....which is the whole point of the MUMPS link to DWTF, so it's weird that the other guy is dug on on an alternate interpretation.