Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

In domains in which the underlying domain is complex, with an irreducible informational complexity, as well as rich and frequently highly significant interactions, there's a frequent emergence of faddish, or ritual-driven, behaviours. Both seek to reduce risks and accountability, as well as to increase credible signalling of traits.

Both sides of the technical recruiting transaction (job-seeking and recruiting) exhibit these behaviours and tendencies (what to wear, how to format resumes, presentation, side projects, for seekers, interviewing, tests and screens, and other filtering practices for hiring teams).

The consequence is a tremendous amount of friction, inefficiency, and fear-driven lore. Some years ago a senior Google staffer commented that they'd found a guaranteed hiring heuristic: "No". That is, reject all candidates.

My response was that if this was serious (and it was at least partially), that this was a profound sign of weakness within Google: an inability to seek out and onboard talent successfully.

There are other possibilities.

It could be that the notion of private firms hiring highly-skilled talent is inherently flawed.

I've speculated that one of the justifications for the ancient Egyptians to build pyramids was as a combination of a skills-development, skills-retention, skills-demonstration, and brain-drain-mitigation programme. From what I've read there's at least some independent informed speculation along similar lines.

One of the functions of writing a book, a notoriously unremunerative practice, is as a credible signalling of skill and ability. (And of book-writing capabilities, for what that's worth.) Books are very fat sales brochures.

In a tech world in which typical tenures are measured in months or single-digit years (2--3 years being typical from what I understand), and correlations between any hiring practices and actual performance ... at best weak, there's an inherent issue.

There's also the question of equitability of a process in which employers have vastly greater access to information on individual prospects than prospects do on companies or hiring managers / management teams. George Akerlof's "Market for Lemons" suggests that more information makes markets more efficient, though my fear is that highly asymmetric information access further tilts the employment market in the hiring firms' favour.

https://www.jstor.org/stable/1879431

http://libgen.rs/scimag/10.2307%2F1879431 Some of my earlier fad/information theoretic musing here:

https://old.reddit.com/r/dredmorbius/comments/62uroa/clothin...



> I've speculated that one of the justifications for the ancient Egyptians to build pyramids was as a combination of a skills-development, skills-retention, skills-demonstration, and brain-drain-mitigation programme. From what I've read there's at least some independent informed speculation along similar lines.

Brain drain mitigation? Brain drain just wasn't a possibility 4400 years ago.


If you're living in a domain in which your core competency is in constructing complex stone structures, then leaking that capacity to another kingdom or empire would be a risk.

Pharonic Egypt circa 2580 was not without neighbours and there was both trade and warfare in North Africa and across the Levant and Mediterranian, notably with Syria, Canaan, Lebanon ("cedars of Lebanon" are a significant reference, as Egypt had virtually no timber), Ethiopia, and Nubia, amongst others. There's also the prospect of defection to internal factions. The Old Kingdom seems to have been generally peaceful with little internal or foreign warfare, at least until the First Intermediate Period, which was largely an internal rivalry. But it wasn't entirely without defence concerns.

https://en.wikipedia.org/wiki/Ancient_Egyptian_trade

https://www.worldhistory.org/Egyptian_Warfare/

I'm very happy to admit that my hypothesis is just that, and that there's no evidence and only a very little external support, though there is some. But if you're going to develop an advanced skill that would be of use throughout the region, it might be advisable to find ways to hold on to it, and as a pragmatic explanation for what was an absolutely immense effort ... there's some sense in the concept.


> If you're living in a domain in which your core competency is in constructing complex stone structures, then leaking that capacity to another kingdom or empire would be a risk.

How? The only military use of large stone structures is as fortifications; their primary feature is that they can never move.

And Egypt isn't even known for its city walls. That's the other civilization of the time, ancient Mesopotamia.

But on top of all of that, none of that is a brain drain issue. You're talking about a hypothetical issue of loss of state secrets. Brain drain is the concern that the local population of skilled workers will all emigrate, leaving the country unable to do skilled work. It's not the concern that other countries may develop the technology to do the same things that you can do.


Building pyramids isn't simply piling rocks on top of one another.

There's mathematics, surveying, architectural design, measuring, logistics, labour organisation, planning, transport, engineering, and a whole mess of related skills. If not kept in practice, they are lost (you're focusing strongly on the "brain-drain" element at the expense of "skills retention" and "skills development bits).

Too: once you've got those capabilities, there are numerous other abilities which derive from them. Large structures means civil engineering, construction, grain storage facilities, and quite probably some degree of metalworking and related crafts, again, which can prove useful in either foreign or civil war.

Take some time to think through possiblities, consequences, options, risks, and opportunities here.


> you're focusing strongly on the "brain-drain" element at the expense of "skills retention" and "skills development bits

Well, yeah. Look at my comment, in its entirety:

>>>> Brain drain mitigation? Brain drain just wasn't a possibility 4400 years ago.

If you can't defend that, then... don't? Make the argument that isn't obvious nonsense; you don't get more credible by throwing in a laundry list of "concepts that sound bad".

> There's mathematics, surveying, architectural design, measuring, logistics, labour organisation, planning, transport, engineering, and a whole mess of related skills. If not kept in practice, they are lost (you're focusing strongly on the "brain-drain" element at the expense of "skills retention" and "skills development bits).

There would have been no lost opportunities to exercise these skills in the absence of pyramidal efforts. They built temples, palaces, and cities on a continuous basis. Surveying is a constant need of anyone who collects taxes. (And it's particularly important in Egypt, where everyone's property lines move every year to match the extent of the flooding of the Nile.)

As far as I've read, the Old Kingdom pyramids stopped being built when the colossal economic strain they involved nearly collapsed the state. That doesn't suggest that they were useful in employing otherwise idle technicians. It also doesn't suggest that the system governing them was especially capable at logistics and planning. Planning would have involved noticing "this pyramid will cost X amount to build, which is more than we can afford".

> Large structures means civil engineering, construction, grain storage facilities, and quite probably some degree of metalworking and related crafts

I think this is backwards to a certain extent; I'd run causation from grain storage -> large structures, not the other way around.


I'm not interested in litigating minutia. I've addressed the point. I've admitted, repeatedly, that this is a very weakly-supported hypothesis, though not entirely without merits. Substantive proof is unlikely to emerge from a tendentious HN debate.

You're the one constructing a far more magnificant pyramid of this than I'd ever intended.


This is the most hacker news comment I've ever read on hacker news




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: