Two possible reasons:
1. Even with severability, there is the question of how much of the contract remains in force. This clause tries to define the portion of the contract that will be severed.
2. Even with a severance clause, you may not have a severable contract. The judge may just decide it's not severable: a contract is between two people, it has no
power to tell the courts what legal reasoning they must apply in a dispute.

Truer words were never spoken.

If you can actually design a solution, throw together a suite of unit tests (that ideally show the basic API,) and deploy it to production, you are already ahead of 95% of the "software engineers."

I wish I had mod points. I lived in the south and was saddened by what I was seeing:

Many of the smart and ambitious leave. The culture, though, remains in place: an ever more pointless divide between the rich and poor (or lucky and unlucky.) Low taxes mean low social services for the poor and insular privately provided schools and social services for the rich, and pretty soon you have an out-of-touch and uneducated rich class and an out-of-touch and uneducated poor class.

If you don't get out of the south at age 21, you are screwed.

There are a ton of jobs out there for people without "3-5 years experience," but who do understand how software works. We basically hire anyone who is not a lunatic who is able to answer all questions along the lines of:

1. What's a hashtable? Time to find/insert/delete? How would you implement one? Issues to consider?
2. Explain the execution time of strcmp("apples", "apple")
3. Describe how a couple of sorting algorithms work.
4. You have X on your resume, please explain it clearly.

It's scary how many people (even PhDs) can't answer these questions. It's like interviewing a lawyer who is a bit unclear on procedure, but figures his raw talent will win you your case.

No, xkcd thinks you are an idiot.

Start tracking what you are doing (helping users, babysitting machines, provisioning, upgrading, debugging, etc.) Then track time wasted by your co-workers, users, etc. Then track hardware costs, licensing costs, etc.

Estimate dollars/year for the stuff you can see/could improve. Focus on the high cost stuff: ask the users of that stuff if they would prefer you to take ownership and get the cost down, or if they would prefer to just manage the stuff themselves.

Then start fixing the stuff they would prefer you to own.

Exactly why this type of survey is so absurd.

Comparing a work from home sysadmin's salary with an enterprise architect's salary is like trying to get an average salary for lawyers (add one public defender to a credit default swap lawyer, divide by two.)

IT jobs cover a lot of space. $78K/year is just a silly summary statistic.

Exactly. Last month I watched $30K in salaries, etc, being wasted because engineers couldn't whiteboard a concept well enough to get buy-in or good feedback from the people in the room.

You want responsibility for a year-long multi-person project? You better be able to explain it well.

Wow. I don't care if a candidate says "I need a paycheck," but I'd never hire someone if I didn't feel they had the potential to double their salary within a few years.

I only want to hire people who will grow into bigger roles (and yes, I expect them to hire the person who will do their old job.) The bigger role doesn't need to be management, writing code that is more strategic works just as well.

If I want someone to be in the same role for 5 years, I'd hire a consultant. But, ideally, I'd hire someone to fix the underlying problem that we need a body in a pointless role in the first place.