gHosts

Definitely not true. Not even close.

Here some reasons I have seen why nobody has a clue about value...

* Often the financials of a company are "need to know" or obfuscated (Y-Axis unannotated)... * The financials are regarded as irrelevant to engineering (by management, finance AND engineering) * The estimates of where the value lies is often just a spreadsheet sucked out of the thumb of somebody in marketing who already has a eye on his next job. * Nobody _ever_ goes back and closes the loop and evaluate the value estimations and the value estimation process, so the company never learns to do better. * Let a drooling slob of an engineer talk to a customer? You're joking Right?

And yet engineer is all about tradeoffs, we make several per day under the illusion we know which are wise....

If I understand what they are looking at, they are looking at problems that are simple in the sense of having no local optima, but complex in the sense of having very high dimension.

ie. It's not a problem for the class of problems they are optimizing.

I've been out of that domain for awhile... but when I did a lot of optimizing / fitting in a high dimensional space with a function that had lots of local optima... I found the downhill simplex method with simulated annealing was most effective / robust.

This is why we need initiatives like https://en.wikipedia.org/wiki/Open_data

Actually, once you're over the addiction to sugar....

* Sugar tastes rotten. Seriously, it's tastes like somebody put something that's gone off in your coffee.

* Fat tastes sweet. I use cream in my coffee, don't miss sugar anymore.

* The supermarket sugar aisle smells like a gas station. Seriously.

I don't miss sugar, it wasn't really adding joy to my life. It was mostly adding dentists bills.

My brother once lived in a very hot dry climate and had access to unlimited Coca cola sugary drinks.... within in weeks he had fattened like a bloody pig. It was amazing.

He now has type 2 diabetes. :-(

I don't miss gluten either, pizza's used to cause excruciating pain.... I now just eat the topping and throw the base away.

All the taste, none of the pain.

Best illustration ever for the saying, "Give me 19 parameters and I'll fit an Elephant!"

Agreed...

..also "Zen Flash, Zen Bones" https://en.wikipedia.org/wiki/Zen_Flesh,_Zen_Bones

http://www.ibiblio.org/zen/cgi-bin/koan-index.pl https://terebess.hu/zen/101ZenStones.pdf

I regard engineering as a daemon that will magically give me anything, from a nice cuppa tea to a nuclear holocaust.

The daemon has a strong preference for nuclear holocaust.

The trick is to give it explicitly, concisely and precisely supply all the information it needs, out of the infinity of possibilities, to make a nice cuppa tea (preferably without creating a toxic waste problem).

I tend to construct my code.

I have the signatures of the functions at the boundary of the code visible, and then I create the data structures to hold the information required.

Then I wired the boundaries to the data model and uncover... holes. Gaps.

So I assume then my daemon will magically grant me all I need to plug that gap, so long as I tell it precisely what the gap is... and then test that what I have so far is "tea-like" and not emitting gamma rays.

And then turn myself into the daemon to create those parts I needed to plug the gaps.

So what is my mental model? A daemon with a preference for nuclear holocaust.

Especially automatic refactorings...

If you look at real code, often the main reason why this routine was added to this file... was that the programmer worked in this file last.

Is that a Good Reason? No it's a bloody awful reason.

An automated tool should be able to look at the network of declarations, definitions and references and work out which module each thing should be in.

Graph Theory in the presence of modules and dependencies between modules.

Every thing in computing is modular and multiple levels of modules and has coupling and dependencies between modules.

It's the hottest and hardest problem in real world computing... it's nicely representable in a graph theoretic framework.... but graph theorists aren't looking at it.

I feel...... spuriously invoked

Golf.https://code-golf.io/

Mathematical golfing

And on the subject of what most not to do....

For pity sake don't send a picture of a log or an error dump..... send the text.

Please.

Pretty please.

With tears in my eyes.

Always do both.

If it matters, do both.

The email is for reference, the walk and talk is for communication, awareness and motivation.

Consensus is NOT about everyone agrees on the solution... it's about everyone agrees that they have been heard and understood.

https://tools.ietf.org/html/rfc7282

1. Remember the person you are talking to does not have the same things in their head, and they are not looking at (or seeing) the same things you are....

So no matter how much you disagree with them, from their perspective it makes sense and is ethically correct.

2. Most people are in perfect agreement on most things eg. Motherhood and Apple Pie are Good Things.

Most conflict arises from differing priorities being assigned to various "Good" things.

3. There are precisely two things and ONLY two things you can change in this world.

What you do and how you react.

No matter how important it is, those are the only two things you can change.

Mostly the problem is that the better activities have become less readily available and/or more expensive.

It's finally happened.

Satire can no longer compete with reality.

Because they are written by people who never actually understood Design by Contract and class invariants.

Meh.

I worked on mainframes for years and cheered when minis came along.

I worked on minis for years and cheered when PC's came along.

I worked on DOS for years and wept when Windows came along.

I worked on windows for years and sang choruses of joy when Linux came along.

I haven't moved off Linux because it just keeps getting better and better.

Would I recommend anyone go back off Linux to ye olde mainframes?

Not on yer Nelly. Never.

In Africa children often have names like "Beauty", "Princess", "Precious" (yes, Precious, that's a fairly common boys name)....

I actually love it.

It says what the parents feel about the child.

Oh the day I realized there was more interesting stuff appearing on fidonet faster than I could ever read. That I could read all day and still get further behind...

Now I'm now so sure... sometimes I think the rate of interesting new stuff to _read_ is slowing down again and I may yet catch up again.

Are the Good Old days before computers became those things the secretary didn't quite know how to use.... it was downhill all the way from there.

Or "poke" into the interpreter's ram region to do crazy things like shifting the string store to overlay the graphics ram and clearing the screen faster than any other technique by creating new blank strings and then poke the original values you peek'd out of it back.

Actually had an office with a door you could close and a window you could open to actually think in.

And then walk down the corridor to the terminal room to turn thoughts into programs and/or ask the other guys or gals.

Strangely enough there were more ladies around in those days.

Miss that.

Or writing amazing CGI scripts in perl that did amazing things before I realized I had to maintain aforementioned perl scripts.

Turned to shit pretty rapidly, but for a day or three I was wondrously happy.

Did a course on a new fangled laser printer that printed _all _ the payslips for _all_ government employees of the time/place.

It took a fair number of pages to get up to speed and a largish chunk of a box of paper just to stop.

Postcript? Ha.

You're joking.

Ah, the good old, "Read this pile of books", call me when you done style of onboarding.

I'm an applied mathematician that worked in those days with chemists, microbiologists, hydrologists, civil engineers, electronic engineers, physicists and almost every darn flavour of critter _except_ computer scientists.

Learnt a lot of interesting stuff.

Most of their code was utter shit to read and maintain... but then so is that from most recent comp sc graduates...

Want nostalgia? Run this happy command...

  enscript  --landscape   --line-numbers   --media=A4   --font=Courier7   --highlight-bars   --verbose   --highlight=c myFileName.c