The curse of the north bay is the lack of trains. BART runs to the east bay, but not the north.
The ferry is basically worthless if you don't actually live in San Rafael. It's a nice ride, but very slow, and not cheap.
Personally, I quite like the bus system: if you don't mind taking the bus, Golden Gate Transit is great. The commute from somewhere like Novato is still long (1hr) but you can use the time to sleep or listen to audio books. Be sure to check the schedules of the express routes you care about: they end kind of early (~6, 7pm) and the non-express routes that run longer take easily twice as long.
Driving is a different story. There are three things to consider: 1) leave early, like 5ish, especially if you're north of marin 2) if you work in the financial district, the time to drive from the edge of SF to parking is substantial and 3) you can park in the financial district itself, but this will cost you around $25/day.
I suppose there's a fourth thing to consider re: driving which is that your drive home is almost guaranteed to be painful if you have to go beyond San Rafael and are not car pooling.
I expect location matters more. Even here, the salary difference between working in SF and only slightly further north can be $30-$50k for comparable jobs.
Then I expect the average developer could get a big bump in salary working for an SF company.
But it's not a free lunch: cost of living is high here and some things like commuting are terrible. I know developers who moved back to Minnesota so they could waste less of their lives in their cars.
Hence the appeal of "work from Ohio". 100% remote seems to be an ever more realistic option, but I expect it's still harder to find work, although work is so easy to find this may not matter much. A bigger issue may be finding a 100% remote job at a company with a remote culture. Working remotely with a company whose culture is centred around an office can be an exercise in a thousand little tediums.
The different social dynamics of working remotely can be a challenge as well.
I feel as though my post reads a bit like "don't try to work for an SF company", but that's not really my intent. I mean more to convey that the pitchbox salaries are perfectly real, but it's not too good to be true. Moving to SF to work, or working remotely are options, but salary isn't everything and these choices won't be right for everyone.
I can only speak for web application development, but $140k is definitely not for "extraordinarily experienced and talented programmers only" in San Francisco for someone doing js/rails.
Are the people you know working in startups where part of their compensation comes in equity, or the prestige of working at a well-known "great" place?
It is true that for people to buy ideas they must have value. But the transaction must also be feasible. The trouble with ideas is that I can't easily prove to you the idea is valuable without having already given it to you, thus destroying your incentive to complete your end of the transaction.
It may be that there isn't a market in ideas because they have no value. But it may also be that there isn't a market in ideas because it's too difficult to facilitate the transaction.
This is an excellent point, and perhaps an argument against at-will employment. Even if you had a candidate who was willing to commit to staying employed long enough to justify the time and expense of their training, in many US States they lack the legal power to agree to such a contract.
Extrapolating from personal experience here as a programmer but it looks tough for a couple of reasons, one of which is pay. Employers tend to give very good starting salaries for experienced people, and very poor raises. So employees sign up inexperienced at low pay, get trained and get raises that put them below market rates, so their incentives tell them to do exactly the job hopping you describe.
Manage in what sense? If you're trying to sync. between multiple machines, one approach is to check in the /etc files in your home dir and symlink. Or are you looking for something else?
I second dekz's encouragement. I went from windows/eclipse to windows/eclipse + cygwin to linux/eclipse to linux/vim(et. al).
It's worth it. It really is. At some point you will marvel that you ever thought it was a good idea to navigate filesystems with the mouse, or edit text in a non-modal editor.
On the other hand, be careful with this approach. It can be very difficult to get candid feedback in environments with a certain expectation of etiquette.
One advantage you may be winning for yourself with your caveats, without even realizing it, is a degree of credibility. By being careful that what you say is correct, and not merely "Basically right, more or less. Kind of.", you may find it much easier to convince people when you _do_ take stronger positions.
I think Aaron's counterexample is pretty terrible. Recalling one time you slept in once in 2003 is really not pertinent to the spirit of the question. The point is clearly not to ask you whether you have ever failed to show up for work on time, but whether you have reason to believe that it may be particularly difficult for you now, perhaps because you live far away or have a history of repeated difficulties with time management.
Personally I would recommend erring on the side of accuracy. I've witnessed plenty of cargo cult programming. "This technology will perform well enough for us" could mean from one person "I have run benchmarks on representative hardware and know that, under conditions I am about to explain, this technology will work for us", and from another "I have read about someone else using this technology in some way for something. I don't remember where I read it. Probably a blog post that made it to hacker news."
The difference being that the magician deceives you for your own benefit and at your request. Though the CEO and politician deceive you (or lie with the intent to deceive), you didn't ask them to.