The OFFICIAL programming thread

NSFW: Language warning

The Unhinged Nature of GTA V Source Code

Snippet of an email I just sent:

Having reviewed the changes on (redacted) in conjunction with the reference (redacted), I don’t believe it will cause a problem when migrated to production.

*I don’t believe". Urgh.

I believe there’s a fucking problem with our system landscape and testing methodology if it requires me to make statements of faith about what is going to happen when changes are migrated to production.

Gators1

As a person of faith, why not? Maybe end with “may the good lord bless you code”.

Yeah I should have at least ended the email with “Pray for us all”.

Anyway, I took one for the team. I consulted with two other devs before sending that email but I didn’t write “we believe”. I don’t know for sure but I’m guessing I can afford unemployment better than they can.

Body hell, the manager I wrote the above to approved the wrong change. That’s probably on its way to production and the change I was asked to investigate has probably missed the cutoff window.

This thing is giving me bad vibes now.

Pakoon

Can I have your stuff?

Kilemall

@Pakoon said in The OFFICIAL programming thread:

Can I have your stuff?

I want the tuk tuk.

No toys for either of you I’m afraid. The changes went to production ~~six~~ four hours ago and nothing broke that I know of.

Pakoon

@Kilemall said in The OFFICIAL programming thread:

@Pakoon said in The OFFICIAL programming thread:

Can I have your stuff?

I want the tuk tuk.

Dammit!

Pakoon

@Hog said in The OFFICIAL programming thread:

No toys for either of you I’m afraid. The changes went to production ~~six~~ four hours ago and nothing broke that I know of.

Dammit!

Kilemall

@Hog said in The OFFICIAL programming thread:

No toys for either of you I’m afraid. The changes went to production ~~six~~ four hours ago and nothing broke that I know of.

Aw fuck!

I mean congrats.

Jam

@Hog said in The OFFICIAL programming thread:

No toys for either of you I’m afraid. The changes went to production ~~six~~ four hours ago and nothing broke that I know of.

OK, so here is where you tell him that your son actually wrote the email suggesting that they don’t go forward with the changes. Not to worry, you’ve changed the password to “Sondontknowitnow1234.”

@Jam said in The OFFICIAL programming thread:

@Hog said in The OFFICIAL programming thread:

No toys for either of you I’m afraid. The changes went to production ~~six~~ four hours ago and nothing broke that I know of.

OK, so here is where you tell him that your son actually wrote the email suggesting that they don’t go forward with the changes. Not to worry, you’ve changed the password to “Sondontknowitnow1234.”

It’s in production now anyway with no adverse consequences so it doesn’t really matter what I had said any more. And we don’t have an elegant way to rollback changes even they wanted to. As it is, it looks like I made the right call so I wouldn’t want to undo that.

Speaking of the tools we wrestle with, the problem was caused by a change to a method (that was going to prod) where other people working on a different issue had been modifying other parts of the same class and their changes were staying in QA. The issue rested on whether the CD tool would only migrate the change for the single method or not. The code for the CD tool is over 30 years old in some parts and is pretty opaque in terms of what it does sometimes. It was state of the art before CD was even a thing but it’s not any more.

We’ve just got approval to move to a git based CD pipeline this year which should get rid of all these issues. (Most) source rollbacks will be a cinch and even in the one QA system we’ll be able to test different batches of changes in isolation just by changing branches.

Gators1

@Hog said in The OFFICIAL programming thread:

@Jam said in The OFFICIAL programming thread:

@Hog said in The OFFICIAL programming thread:

No toys for either of you I’m afraid. The changes went to production ~~six~~ four hours ago and nothing broke that I know of.

OK, so here is where you tell him that your son actually wrote the email suggesting that they don’t go forward with the changes. Not to worry, you’ve changed the password to “Sondontknowitnow1234.”

It’s in production now anyway with no adverse consequences so it doesn’t really matter what I had said any more. And we don’t have an elegant way to rollback changes even they wanted to. As it is, it looks like I made the right call so I wouldn’t want to undo that.

Speaking of the tools we wrestle with, the problem was caused by a change to a method (that was going to prod) where other people working on a different issue had been modifying other parts of the same class and their changes were staying in QA. The issue rested on whether the CD tool would only migrate the change for the single method or not. The code for the CD tool is over 30 years old in some parts and is pretty opaque in terms of what it does sometimes. It was state of the art before CD was even a thing but it’s not any more.

We’ve just got approval to move to a git based CD pipeline this year which should get rid of all these issues. (Most) source rollbacks will be a cinch and even in the one QA system we’ll be able to test different batches of changes in isolation just by changing branches.

Please post your git pipeline script here so I can do a find and replace your company ahit with my company shit and use it. I have learned enough tools this year and am not looking forward to diving into Gitlab too.

Gators1

I need this!

https://www.reddit.com/r/GPTStore/comments/1aqadis/i_made_a_gpt_that_turns_insults_into_emails/

I was wanting to doing some bitwise operations in my calculator app this morning and couldn’t get my function calls to work so I had to RTFM. For who knows what reason, they’ve decided that function arguments need to be separated by semicolons instead of commas. I hadn’t seen that before so I asked Gemini:

I love bucking convention when it delivers something new. Otherwise, stick to fucking conventions. You aren’t a special snowflake.

Gators1

Fucking hackerz!

The Malware So Tricky Even Programmers Fall For It

Kilemall

I handle no extension files at my job. Some programmers or vendors like those.

Leadership Is A Hell Of A Drug — Ludicity

The above guy has a pretty amusing, if sometimes a bit overly toxic, blog where he rants about his job and the industry. I’m frankly amazed he hasn’t been fired for it since it would be trivial to identify him using all the details he’s posted.

Anyway, this in particular amused me just now:

Consider this: the lawyer I mentioned in this post had opinions on whether statically typed languages are superior to dynamically typed languages, and had engaged in some complex meta-cognition. See, they reflected on how every time they felt that something in Go was ugly, it later turned out that Go was well-designed and they simply didn’t understand the issues being designed around. This turned into a rough rule that seeing a problem with Go typically reflected a flaw in their own understanding, thus flaw with Go is actually best processed as flaw in own brain until proven otherwise.

I don’t know much about Go, but that’s not important. What is important is that I’m on my knees, arms outstretched, gazing up at the sky, tears streaming down my face, and begging to have one question answered:

Why does a practicing lawyer have more thoughts on programming than every CTO I’ve had a 1:1 with at a big company?