The OFFICIAL programming thread

Bit of an anti-rant but, this morning I refactored the behemoth method that led to the earlier rant. I was particularly impressed with how easy and clever the refactoring tools I used were so I sent out a meeting invite to demo the whole process to the 7 or 8 devs I regularly work with.

I titled the invite “Refactoring for fun and profit” and made everyone an optional attendee so I didn’t know how many (if any) would turn up. They all did and they were really impressed with the demo and a couple made comments like “the code is beautiful now”. So I’m pleased they were so receptive to it.

Tazz

This is all I can add

1 adel.gif

Gators1

The gorilla is bad at seeeeequel!

Had a similar situation many years ago in the early days of business computing. We connected MS Access, like one of the first versions, to our accounting system via ODBC because the system reports sucked ass. IT created credentials and didn’t worry about permissions so we had read/write access to the system database. A coworker was making a report and didn’t know Access very well, so she started deleting rows in the query output like it was Excel. The deletes passed straight through to the source system and after a while people from accounts payable started raging that vendor records had “disappeared”. Poor girl thought she was going to get fired that day.

Tazz

@Gators1 Did she work at CRS by chance?

I was just installing an Eclipse theme and saw this which I liked:

oyaji

the image above made me think of this one:

aww crap, i thought surely i could invert it here. doing so makes it look like a rocket ejection frm a P-38. i posted it thus years back in teH Anger in the OOBF, where it was good for a laugh.

Gators1

We need this too! My morale is at rock bottom.

Gators1

This is kind of interesting, especially thinking about how half-assed some of the thinking is about systems in my company. Basically Cox decides to put a bunch of functionality for a customer to control their own hardware in their account website and by doing so exposes those same capabilities to hackers via api calls.

researcher accidentally finds 0-day affecting his entire internet service provider

Jun 30, 2024

A Git story: Not so fun this time | Brachiosoft Blog

Before TeamWare was completed, Larry continued developing NSElite, which would make the TeamWare team look bad: one guy with Perl was outpacing eight people with C++. The VP then told Larry, “This has been reported to Scooter (Scott McNealy, Sun’s CEO). If you release it again, you’re fired.”

Urgh.

Love this. 25 year old bug that is fixed in the most recent Firefox release:

33654 - TEXTAREA incorrectly applying ROWS= and COLS= (horizontal / vertical scrollbar extra space, with overlay scrollbars disabled)

There’s been activity on that bug report nearly every year since it was created in August 2000. I assume they had to wait for the chosen one to be born, grow up and graduate college before joining the Firefox project to fix it.

tigger

I have a Firefox scaling bug somewhere that hasn’t been fixed for decades

Gators1

Maybe that chosen one is you? Fixing some old ass code seems easier that pulling a sword out of a stone and ruling filthy England for the rest of your life.

Gators1

@Poo, let’s write some OSS government software that will secretly transfer all of Tigger’s Nazi gold to us!!! We will be rich!!!

Gators1

This is the tech job market today.

Has AI hype jumped the shark yet? It must be close with this:

As mentioned, LLMs have emerged as a game-changing solution for mitigating the risk of technical debt. AI has made enormous strides in understanding and generating text, and with its ability to process and produce human-like responses, it’s evident that LLMs can be integrated with existing codebases and ticketing platforms to create self-healing code.

Self-healing code! Stupid humans can’t write self healing code!

Maybe the good stuff is still coming but I’ll be less cynical when I experience an LLM generate a non trivial code example that wasn’t broken in some way or other.

Blanks

Code is just like the human body, if it gets a boo-boo the LLMs will send blood and begin scabbing it over for repair.

Gators1

Kind of a thought provoking piece @Poo. Being in data I am always focused on value because that’s mostly what I am trying to deliver. My management is just trying to deliver bits of software though, checking off some feature they committed to. My team specifically has spent the last 8 months developing their CICD/Terraform process so their development process is solid and not delivered a fucking thing. Not sure I agree with his minimalization theories without giving a thought to the future. I have seen projects get in trouble because they work, but the result is very hard to untangle and scale or whatever later when the company may hit some acceleration point and needs that scale right now. The idea ring true for a startup though that is just trying to put something out on the internet to get funding. Does that really work with complex systems?

Working Software Is Not The Primary Measure of Progress

Kilemall

The stuff I built lasts for decades, but that’s mainframes and dropping out of everything but finance and banking.

Gators1

@Kilemall said in The OFFICIAL programming thread:

The stuff I built lasts for decades, but that’s mainframes and dropping out of everything but finance and banking.

His argument though is that programmers should be building for short-term customer value, not worrying about 10 years from now. Agile displaced waterfall because waterfall had long ass cycles where people were assuming what the customer wanted and there was nothing they could do if they got it wrong. When the internet came and deployment cycles were shortened, it didn’t make sense to do long ass cycles if you could just deploy and pivot if it wasn’t what they wanted. He’s taking agile in a different direction and saying the objective isn’t to deliver software every cycle, but deliver value. So the priorities are driven by things that improve the customer experience or create demand, not stuff like automation, refactoring, etc. I kinda agree, but also ignoring some of that back end stuff can bite you in the ass later.

@Gators1 said in The OFFICIAL programming thread:

Not sure I agree with his minimalization theories without giving a thought to the future. I have seen projects get in trouble because they work, but the result is very hard to untangle and scale or whatever later when the company may hit some acceleration point and needs that scale right now.

Yeah I was thinking something similar when he said this @10:15:

“If the performance problem doesn’t matter to the customer, then it is fundamentally not a problem”.

I agree with a lot of other stuff he said but the above is just irresponsible.