What turds have you been asked to polish?

[BrilloPad]

Just for clarity, those people with the PhDs, etc, that I mentioned were so good to work with before: they were all men. It's wee boys with big egos I've no time for.

I find that, beyond a certain point, the more intelligent the developer the worse the code.

Some people like to prove how clever they are by making things as complex as possible.

[aussielong]

I find that, beyond a certain point, the more intelligent the developer the worse the code.

Some people like to prove how clever they are by making things as complex as possible.

Writing good code is not hard- there are rules/idioms/patterns you can teach "code monkeys".

Identifying the correct code to write is hard , and takes many many years of experience.

Having nice hand writing does not make you Shakespeare.

[alreadypacked]

Lucky for me, most of my roles have involved trowing turds out

[d000hg]

what I want to know
is why the games were so much better with 128 k to work with, than they are now

Possibly because you were younger?

[doomage]

When I worked in the hospital lab we'd occasionally have to do a 48 (or 72) hours faecal fat test, which involved boiling up someone's tulip that had been collected over 48 hours to work out the fat content. So I have boiled turds but not polished them.

[norrahe]

Lucky for me, most of my roles have involved trowing turds out

I seem to start on one thing and then get asked to take on "project poisoned chalice" it gets done but not without a lot of political interference from the big knobs, meddling from minions and a lot of stress.

[d000hg]

I find that, beyond a certain point, the more intelligent the developer the worse the code.

Some people like to prove how clever they are by making things as complex as possible.

It can go either way - they can do that (I think as much for fun as showing off), or they can go the other way and just churn out wonderful work.

[doodab]

It's a question of intelligence and experience. Even the most intelligent amongst us require practice to master a complex skill.

0 0 = usually doesn't work anyway

1 0 = usually works, often complex and impossible for an outsider to follow, not designed for maintainability or extendibility and eventually ends up a mess

0 1 = usually works, mostly well structured with occasional inclusions

1 1 = this is the stuff you learn from

Many problems stem from "polyglots" who use C idioms when coding in a modern object oriented language, leaving a trail of null pointers & invalid objects in their wake and completely ignoring the standard libraries, although thankfully this seems to be dying out now. Another good one is people who implement components that deploy into a threaded framework, but because they have no direct interaction with the threading mechanisms they fail to understand the implications for shared data.

[tractor]

...

oh deary deary me.
you seem to imply that just because a job is green fields , it cannot be a disaster waiting to happen. have you never seen changing requirements, bad spec, moving goal posts , project creep.
????

a sound sytem that has a new angle. cant possibly go wrong ???


well I may have been unlucky and had the same bad experiences as zippy.
but very few project run on rails in my experience

Yes, it has a methodology all of its' own called.....Agile

Back in the early days of DBASE, I was called in to review a project that after a lot of effort, was going nowhere. It was sad but the guy had really been thrown in at the deep end and bear in mind that anyone who could spell i386 was a computer expert then and usually was a prime candidate for development jobs. Each form in his app had an individual program associated with each field to capture user input. I think there were about 2500 programs for a few screen forms. I think he had done his best but simply did not get it. I was a bit sad putting the report in but oh well....

Another contract, I was asked to review a prototype one of the permies had put together for a planning system. He had mocked up a load of screens in Publisher and demonstrated them as a prototype when in fact they were nothing more than glossy wireframes. He had spent every waking hour for 6 months on this little pet project. Project board actually assumed they were partial developments and gave him a budget to finish it off - Then he was pretty stumped! Funny thing was, they were quite surprised when I told them it had never done anything but looked pretty! Even funnier, they could have done exactly what he was trying to do in Project or Primavera.

[Gentile]

Indeed.

Heaven forfend that you try to maintain code produced by a "guru".

I've produced some turds of my own <preen>

One of 'em ended up selling in Argos. <preen>

It really was a gigantic POS.

Haven't we all, zeity? There's a difference, though, between people that recognise and learn from their honest mistakes like you, and those that make the same ones over and over again because they mistakenly believe everything they've ever done was 100% correct. I'm equally dubious of anyone that either believes they've never made a mistake, or who messes up consistently in the same way over and over again and never seems to learn from their past experiences.