lwhalen

This guy has also obviously never met the strange and rare creature known as a 'bass player'.

The ideals you've posited sound pleasant initially, but they bear little resemblance to being able to do real work in the messy, impatient, human-infested world others live in. Would it be nice to eschew the linux kernel, or the openbsd community, because a large fraction of those folks meet the ESR 'busy hacker with attitude' definition? Sure. Would it be career-killing to say "Linus/Theo were curt with me, _how wude_, let me agonize over it on my Tumblr and waste dev cycles finding another kernel/crypto library/whatever"? Absolutely.

ESR's guide is pragmatic, and far more effective at solving the real problem in front of the questioner, than the "pout and shun" methodology you've espoused.

ESR did it better, imho :-)

- Account is registered for a /56, their router only responds to requests for a /59

- Once my CPE is configured for that /59, routing from anything smaller than that /59 is just dropped on the ground (no delegation to/from /64s, etc, is allowed)

- significant investigation later, the root cause is because I also have a /29 static v4 block on my router. Removing the v4 block from the Comcast end, everything works great (as a /59). As soon as the Comcast equipment receives the static v4 config, v6 delegation grinds to a halt

- Comcast Business Support is WOEFULLY under-trained on v6. Talking about v6 netblocks gets me strange stares and "But I see your /29 on the account, is that what you're talking about?"

- Comcast line-managers have told me on multiple occasions, "Oh, if static-v4/v6 doesn't work, then it's just not supported. Sorry, just take your business elsewhere if static is important to you"

www.stelligent.com

www.reactiveops.com

www.fuzzy-logic.org

none of these businesses are hurting - in fact, they're growing. $300k, in the scope of an individual, is a staggering amount of money to most. To a (healthy) business, it's more along the lines of a small/medium capital investment. A lot of the pitch is "You can home-grow this over the course of X years, making occasional mis-steps and paying your expensive engineering salary along the way, with every hour they spend developing this automation thing being an hour they're not spending developing your core product. Alternatively, you can pay Y and have it done in weeks/months while your FTE staff focuses on your core product the entire time. You get to reap the return (on time, on money, on agony, whatever) much faster, you have experts setting it up right the first time, and it's delivered turn-key, with training, to your FTE staff so they can start using it immediately"

There's a reason most engineering talent is under-paid, and it has much to do with the individuals not understanding the business mindset, the actual value of the services they provide to the business, and frankly being snowed by 'big personality, business-oriented' hiring managers. $200k for a talented mid-level engineer with a history of producing should be the 'good start' number, not the 'pie in the sky' one, not just in SV but from sea to shining sea.

"We have zero CI/CD expertise in-house, build us a pipeline that does all the things vis a vis testing, deploy, etc". OK, that'll be $90k, we'll have it done in 2 weeks, plus another week for training your team on its use and upkeep.

"Our CTO is actually the tech person, and he's tired of provisioning VPCs and VMs by hand. His time is better spent doing more business-ey things" Poof, here's your CM deploy that lets your CTO either update a single YAML file and press 'go' to spin up all the things, or allows him to hire a junior/mid-level to twist all the knobs and pull all the levers. $300k please, it'll be ready to go in six weeks.

so on, and so forth.