Coding Standards

frankus

so if it works in your team it might get taken up in other teams ... and break the silo?

@equalsql - in our place it's because someone decided that putting numbers in was maverick.. now they bitch every time their code is removed. (they're egotistical enough to run: find . -type f -name "*.p[ml]" -exec svn blame $USER \; | wc -l on the codebase to see their "stake" in the codebase - which gives you some idea on their incentive to write neat DRY code - even come over for a friendly chat if you're rummaging around on "their turf".

@Phil_aka_Pip - I don't tear strips in people's obfuscated code any more, these days I ask "is this production code?" and one way or another someone's favourite puzzle doesn't seem so trendy any more. The one time I critiqued it, I was talking to the author

Our place started going downhill when they hired someone who used a map statement in a void context...

Phil_aka_Pip

@frankus I too have to be, er, diplomatic sometimes when I'm talking about other people's code

Jalapeno

frankus said:

so if it works in your team it might get taken up in other teams ... and break the silo?

It might, probably will.  But in the mean time not very practical to re-baseline multi-national projects though.

My point (badly made) is that the external dependencies from these other projects/programmes need to be in the backlog, and a cosy little insular Agile project, however well run, will run into issues otherwise.

Axe_meister

I've recently dabbled in programming after a 25 year break (Used Fortran in the early 90s) this time VB.

One thing I have learnt is.

Create a standard and stick to it.

The problem as a beginner is, as you start learning more and more you adopt new standards as you go along.

When you go back and debug the first lot of coding you did, you've forgotten the standard you used.

I totally confused myself

Myranda

The company I work for paid a company to write it some software.

Then somewhere between purchase and completion someone decided to buy the company.

The company was dissolved and the coders were incorporated. 

The software was finished - full of bugs, slow and lacking all sorts of functionality. 

Then ... some years later we still had that software and the head dev was a board member.

Any talk of upgrading or replacing the software was seen as insubordination.

They finally left and emergency coders were parachuted in.

They were tasked with making the software modular and not-shite.

The new coders, apparently not much better than the last came up with a "modular" system which consisted of a single module - the whole thing with all extra functions within. 

The para-coders left on their off-white horses, saddle bags filled with shiny loots.

Years later and the IT department is tearing its hair out telling the sad tale to anyone who will listen and begging the current board to allow them to scrap the software altogether and start again.

Our IT department has become the town-drunk.

Note - the software is just a database interface. Huge database... but just a database. To code so badly the first time, they pretty much had to read the OP quoted link as some form of bible every night. 

Our sale team... they are not on the same site as our global IT team ... they tell the clients we use "unique custom made software" as a sales pitch. This has made the decision to scrap it somewhat harder.

Jalapeno

Axe_meister said:

I've recently dabbled in programming after a 25 year break (Used Fortran in the early 90s) this time VB.

One thing I have learnt is.

Create a standard and stick to it.

The problem as a beginner is, as you start learning more and more you adopt new standards as you go along.

When you go back and debug the first lot of coding you did, you've forgotten the standard you used.

I totally confused myself

Reminds me of the old gag - I can program Fortran77 in any language

mudslide73

Phil_aka_Pip;374870" said:

IMO there are 2 classes of people who should be banned from codIng:

Them that think they're so smart and that comments are for wimps that can't read the hex of the executableThem that have a small but very basic grasp of procedural logic, but no idea of how to use the language. Some of these people are actually quite good hardware engineers 

That's so true. I haven't done any code since college but I'll have a firk through stuff... A little understanding is a dangerous thing. Learned my lesson a long time ago.