Dear Other Development Team: Just because the server is called "Test" doesn't mean that it shouldn't be treated like one of your production servers. You see, that's our test server. We sit around running code on it all day. Sure, most of the time nobody will be particularly inconvenienced because your service is rarely used in the grand scheme of things. But sometimes, you're going to deploy your breaking changes in the middle of the Fucking New Guy's test cycle. Said Fucking New Guy is consuming your service through two additional layers of black-box services and has a deadline in two fucking days. The guys who wrote those other services are buried under 30 solid feet of work and can't fix it. The FNG is not allowed to fix it himself.
And to make matters worse, the breaking changes you deployed to test aren't due to go live for MONTHS. My code goes live in four days and will cease to run at all before December is over. This has the excellent result of meaning my test environment is totally fucking incapable of mimicing what my live environment will be like. Ever. At all.
It is to your advantage that you work at another site.
TRWTF is the timescale for my project, of course. It would have worked if everything had gone smoothly.
So, you're blocked. It happens. It's not like somebody is going to die, is it?
So, you're blocked. It happens. It's not like somebody is going to die, is it?If I'm lucky, my project manager will have a stroke. As for the rest of human civilization, fortunately, this is one of the few projects I've touched that is purely commercial in nature - some other things I've worked on, though. People could TOTALLY die from those. It would take some really complicated and obscure circumstances, but it could happen. Hell, this project dying on the vine would be of net benefit to humanity.
But fuck humanity, I have a reputation of infalliability to uphold!
We missed the deadline by two days because of the actions of John Doe in operations. Based on the numbers sent around last week, that two days resulted in lost revenue of $XYZ,000. While you prepare the bill for that amount for John's boss, I'll be in the utilities room with John and a short piece of rubber hose. Please ignore the screaming and/or whimpering sounds.
So, TRWTF is having multiple projects operating in the same environment. This is really Configuration Mangling more than Change Mangling (to inject some much needed pedantic dickweedery).
OBBlakeyDisclaimer: The word Mangling has replaced Management in a play on words. Configuration and Change management are important concepts in software and other engineering disciplines.
The guys who wrote those other services are buried under 30 solid feet ofThe post line-wrapped here for me. I was expecting the next word to be something like "concrete".
And this has all been made irrelevant because even if I could meet my deadline, one of the systems my project depends upon is being taken offline tomorrow at noon - and staying that way until next Tuesday.
We still haven't officially been informed of that. It came through the grapevine.
Oh, and note that the reason my project depends on this system is because it's absolutely central to our business. The entire division can do nothing without it. Fifteen labor shifts are getting paid to hang out in the break room. Every piece of production equipment is getting a full service rebuild (which is usually done on the fly, but the vendor contractors who spend their lives fighting fires have heard and are going to leap upon the first opportunity in years to do any actual preventative work). We programmers, however, will accumulate inbound customer projects and be unable to get anything done on them.