About 80% of the developers work that I test is broken or a lie and the other 20% is coded against specs so comically badly written you would think the analysis who created them got their degrees in marketing.

Of course most of the time the higher-ups decide that something is an 'emergency' so it gets to skip my already cursory testing groups regression suites and go directly into production. Normally it's things like "I need to change the color of the following screens" or "We are going to increase the field length for claim codes" which should get tested, but are fairly low risk so I don't stress.
Of course sometimes they come up with real fun stuff like "We're going to drop the following tables out of the database right now because we need to import some make/model information that we have had since last month and we are a little tight on space.....but no one thinks that table is even used anymore"

Those are the days that I just shut my door and yell "Fuck" anytime the knob moves or someone knocks..

Lord help me!! Dev managers just don't seem to understand that each 'quick-fix' that get's pushed into production is not just a 'testing issue' but a sign that the developers wrote bad code or maybe, just maybe the testers would have found the issue if the specs had been ready BEFORE the code was delivered vs. 2 days into testing.

Grrrr....

It's been a long morning already

Why won't people write specs? I just don't get it. People claim that it's because they're saving time by skipping the spec-writing phase. They act as if spec-writing was a luxury reserved for NASA space shuttle engineers, or people who work for giant, established insurance companies.

The software is ALWAYS going to have a specification...the choice is if it's going to be written or just sitting around in some developers head.