The reddit comments on this article were really good.

Finance pays extremely well. Why make $100,000 at Google when you can make $300,000 at Goldman Sachs? But here's the thing: revenue per employee at Goldman is $1.1 million. At Google it's $1.2 million. At Apple it's $1.3 million. Divide net income by revenue and you get a similar margin for all three businesses.

This all suggests that tech companies can, and perhaps should, be salary-competitive with finance for engineering-heavy jobs.

But big tech companies like Google aren't hurting for talent. Plenty of bright young computer scientists would far rather than work at Google than on Wall Street, even with a lower salary, simply because they consider it more interesting and fun. It's startups that are being drained of talent by finance, and that's what the author of the article is concerned about.

Shouldn't google at least have far more non-salary related costs than Goldman?

According to Wikipedia, Google's profit margin last year was 29 percent ($8.5B profit on $29.3B revenue) and Goldman's was 21 percent ($8.4B profit on $39.1B revenue). You're probably right that Google has more non-payroll costs, but at least last year, Google had a much bigger slice of its money left after all costs, payroll and otherwise.

They could be salary-competitive with Goldman and still have a >20 percent overall profit margin. The reason they just pocket the money instead is that even with their meager-compared-to-finance salaries, they still basically have their pick of the finest employees. I think they simply don't feel that finance is stealing all that many good candidates. But if Google et al start to have a measurable "Shit all the people we want are ditching us for Wall Street" problem, you can expect to see their engineering salaries go way up.

Like most, this article attacks investment banks in the wrong way. Investment banking is not quantitative finance. It's core business is M&A, underwriting and asset management.

Keeping engineers out of finance is good for both parties IMO. Engineers and mathematics majors became "must haves" for financial firms beginning in the 70's and 80's, but really seemed to hit full throttle in the 90's, with the emergence of hedge funds, such as LTCM. Many hedge funds and I-banks saw arbitrage opportunities and needed quant guys to come in and build models to exploit these quickly, especially due to the rapid technology shift and speed.

Engineers and mathematicians don't really understand some of the key issues with finance and their models never are able to properly measure risk or build the proper safety nets against things that happened in the Wall St. collapse of '08, the Asian Crisis of '97, Russian Crisis of '98, etc.

I agree with the author to the point that engineers are best served in other industries. But, I also feel it would benefit the banks, as well.

You can't blame someone for taking the sure bet, especially when they're so burdened with debt. Crippling student loans are issue number one to address if new graduates are to be persuaded to be entrepreneurs instead of I-bankers.

Besides, there are way more PhDs in physics and math than there are professorships; the fact that financial firms hire them is actually a gain in their case, because otherwise you'd have a bunch of really bright people doing data entry or teaching high school, and not using any of their trained skills.

I'm just quoting, don't give me karma, give it to the reddit posters.

Yep, you're missing the incredibly 1-sided contracts users sign to access any cell networks. Here are some relevant gems from the AT&T contract:

We may, at our discretion, suspend your account if we believe your data usage is excessive, unusual or is better suited to another rate plan.

Furthermore, plans (unless specifically designated for tethering usage) cannot be used for any applications that tether the device (through use of, including without limitation, connection kits, other phone/smartphone to computer accessories, BLUETOOTH\® or any other wireless technology) to Personal Computers (including without limitation, laptops), or other equipment for any purpose.

Accordingly, AT&T reserves the right to (i) deny, disconnect, modify and/or terminate Service, without notice, to anyone it believes is using the Service in any manner prohibited or whose usage adversely impacts its wireless network or service levels or hinders access to its wireless network...

Tethering without a tethering plan breaches your contract, so they can refuse to provide service, request you pay more for your plan, or do about anything.

You can. I know, the software that interfaces with iPods on linux are all something of a kludge, and Apple occasionally breaks compatibility and jerks you around. But you can run open software on your iPod and make it really easy. After screwing around with iPod loaders for years, I switched to rockbox and never looked back.

Hopefully they learn somewhat from digg and redesign in a faster less buggy & controlling way.

In the last few years, I've found search results have been dominated more and more by content mills like associated content, ehow, hubpages, about, and others; or some low quality Q&A page, like yahoo answers. The pages are hastily written and edited, and low content. The articles are also typically written by someone without any relevant knowledge or experience - so the information is common knowledge or wrong.

If google's metrics say quality is up, but their users think quality is down, then google's metrics need to be revised to match user experience more closely. I've started using duck duck go because they block content mills, and thus I think their results are as good or better than google, even without the complicated algorithms and all the data google has accumulated.

Thanks to the UDRP and ACPA, that isn't really a problem.

ICANN's UDRP simply requires a showing that the domain name is confusing similar (or identical) to a third party's mark, the registrant has no legitimate interests in the name, and the name is being used in bad faith. There is a lot that goes into showing & rebutting these steps, but that's the gist. UDRP isn't the law of a single country, so it governs disputes covering most TLD (it is mandatory by contract). The remedy is transferal of domain, no damages. It is low cost and quick.

Before the UDRP was released, congress got impatient and passed 15 U.S.C. 1125 (ACPA). It is a new subsection of the trademark statute (originally Lanham act) and imposes liability on a person who registers, traffics in, or uses a domain name that is identical or confusingly similar to a distinctive mark (or dilutes a famous mark), and has a bad-faith intent to profit from the domain name. Again there is a lot that goes into showing or disproving these elements, but that's the gist. Like UDRP, it only covers domain names, not usernames [e.g. twitter "united air" account] or sublevel domains. Unlike UDRP, ACPA requires taking someone to court, costs a fortune, takes forever to resolve, and damages are available.