[madaxe]

The best hires are fresh CS/maths grads with little commercial experience and a burning desire to learn.

All you have to do is provide an environment in which they can, provide a raise as and when they "level up", provide infinite caffeine, interesting problems, and the rest, as they say, is history.

[sramsay]

I assume you're basing that on your direct experience with twenty-five year veterans who have used their complete mastery of a popular toolchain to ship a couple of dozen production systems to customers who were in turn delighted by the quality of the work and the high level of professionalism.

[madaxe]

Show me a 25 year master in web technologies.

I'm basing it on personal experience of having grown a company from nothing without funding to an enterprise carrying over £1bn of client commerce in 6 years. But what would I know.

Oh, and 100% client retention and no marketing except word of mouth. I'd say that's satisfaction.

[somebehemoth]

This is called a fallacy of dramatic instance. Your problem domain and your company are satisfied with young programmers, but you cannot extrapolate that to all software enterprises.

[madaxe]

Sure, but you can extrapolate it to the entire web industry, which is my domain. I've built teams for clients (we started out taking equity in startups in exchange for prototyping and team assembly), and in every instance have followed that formula. In every instance they've had successful eight figure exits.

Sure, it's likely not universally applicable, however you shouldn't underestimate the power of an open mind.

[sramsay]

"you shouldn't underestimate the power of an open mind."

I'm not sure your ideas are quite as bold and radical as you think. In fact, I would say that you're merely repeating the common liturgy of the industry. Young tyros are more "creative" than seasoned veterans. Caffeine addled youths have more exciting ideas. They have fewer preconceptions. They're more nimble. They're in it for love of the game.

What I'm asking is why you're so sure all of this is true. "It's true because I am successful" doesn't seem to me very persuasive.

[madaxe]

I see it as true because it's a repeatable experiment. I haven't argued that my ideas are bold or radical, merely fact - and nor have I argued that this differs from common liturgy. Neophytes are indeed more creative than seasoned veterans, as they haven't condemned their minds to a single, myopic track of thought, as many, and in fact most, do. "PL/SQL is the only structured query language compliant with the PL/SQL specification and is therefore superior" "MUMPS can achieve anything" "Fortran '77 is good enough".

I've worked both in orgs with "greybeards" and in the org I've founded along with many clients. There's plenty to be said for veteran levels of experience, but this is something that's only really needed in a leader, as this experience can be communicated and demonstrated to others. A fool learns by his own mistakes, a wise man by those of others. I suppose this makes me a fool, but a fool who's decided to help others be wise, and to harness those several decades of experience in such a fashion that it benefits all involved.

So yes, you have a point in that years of experience are indeed valuable, but I would argue that this is not a prerequisite to be an exceptional engineer, and that in fact truly exceptional engineers arise through memetic experience, and a thirst for knowledge.

It's true that I speak from personal experience, but then again, so does every scientist who finds a repeatable experiment. I'd be happy to be disproven, but until then, I'm equally happy to treat what I find to be a valid theory as fact.

Young, passionate individuals make the world go round. You only have to look to the great leaps and bounds of science to witness this in visceral actuality.

[matthuggins]

"Best" in what sense?

[madaxe]

Productivity, enthusiasm, willingness to learn new tricks (get 'em in knowing PHP, watch 'em learn C, Java, and ASM as they go deeper into our stack), and willing to take the risk of sticking their neck out with outlandish but brilliant ideas.

[ryaf]

Frequent use of ASM in a web dev shop? Really?

[madaxe]

Not to a huge degree, admittedly, but enough that more than a few of us have ended up with a more in-depth knowledge of assembly than one might expect.

This has typically arisen from working with existing OSS projects which incorporate vector transforms and the ilk in ASM, often badly or subtly brokenly, and from building optimised functions for doing fun things like creating n-dimensional maps of customer/product relationships in order to determine purchase propensity based on cluster profiles.

So yeah, ASM in a web shop. Except we're only kinda a web shop these days - client sites are our marketing platform, but our business is business systems.