On 02/07/2025 18:06, Dawn Wolthuis wrote:
> If so, I’m happy to be wrong about this. It is likely the case that only
> those who have previously worked with MV would think of putting it in
> their stack for new software, but if it is cost effective for new,
> contemporary apps to use MV, then that would good to know.
I'm struggling to get OpenQM into my company as a new stack component.
There's no opposition to it especially, just big company bureaucracy. I
did try ScarletDME, but the company is less and less inclined to use
Free Software. It's gutting - if I'd tried a year earlier I would
probably have got away with it ...
You'll know who I work for from LinkedIn - and I've got back into my
preferred role which is a programmer with a completely unrelated job
title! :-) working in a very-much-end-user role.
Our current software stack is mostly a massive data lake, which we're
drowning in with no information. And the system I'm working on is just a
morass of Excel spreadsheets, with loads of BigQuery data feeds, none of
which (I might be exaggerating - SLIGHTLY) with any primary keys.
As a result we have multiple sources of truth, a variety of different
ways of calculating the same derived data, missing links that stop us
pulling things together ...
My job is to take all the dollies of shopping coming out of the
warehouse/picking sites, putting them on lorries, and getting them to
the distribution centres where they go on vans for final deliveries. Our
new all-singing all-dancing system couldn't tell us which van routes
were being picked in which warehouses! And our botch/workaround was to
find routes that had shopping (which told us the warehouse), and apply
it to similar routes. This worked well until ... they introduced "same
day" routes, which only got shopping on delivery day ... seeing as we
need to know at least a week in advance, we were up a gum tree, which is
our usual position at the moment :-(
Which is why I want a simple, easily understandable system like QM, into
which I can slurp all this data, and where I can make the system scream
blue murder if it's missing necessary data.
I'm sure, once I've got the system working, they're going to try and
slurp it into our new all-singing all-dancing system, but I'm going to
be sure to point out (repeatedly) "how can I - as a one-man crap
programmer - design and build a system in six months that they're
planning a multi-man-year project to replace?"
Our fancy new system "was two years away for five years". I take great
pleasure now if I get the opportunity, pointing out that MV has the
reputation of delivering before time and under budget. My only problem
is that if I get given the chance, I've GOT to deliver. Okay, retirement
may only be a few years away, but I have no intention whatsoever of
cutting and running ... :-)
And given that we're a week away from shutting down our old system, I
might very soon be called upon to deliver! Yay! At least I've already
earned in this new job, my typical reputation in previous jobs of
identifying and implementing automation that speeds things up massively.
A macro that got me the compliment "You've turned a two-day job for an
experienced analyst into a half-day job for a rookie helper". Or "you've
reduced this on-call job (ie under a lot of time pressure) to one third
of it's previous time". I'll be pushing that, for sure ... even if I'm
just picking off low-hanging fruit - there's just so much of it!
Cheers,
Wol