Hacker News new | comments | show | ask | jobs | submit login
Job Posting for Amazon.com Before it Launched (readwrite.com)
127 points by capdiz 1581 days ago | hide | past | web | 50 comments | favorite

'extremely talented c/c++/unix developers'... 'able to build complex systems in about one third the time most competent people think possible'

This language conveys so much more competence than the standard 'seeking unix ninja rockstar' stuff that seems to be de rigeur these days.

A big part of the problem is that everyone, of course, wants rockstars, they want the top 25, 10, or 1% of the talent pool. But not every company actually needs them. Not every company even knows what to do with them or can make effective use of them.

And not every company wants to pay for them. If you need rockstar results expect to pay rockstar prices.

As someone wrote recently, some companies say they're hiring when what they really want are shiny new people for their company. Rockstars are pretty shiny.

Isn't that a sure way of not getting anybody competent ? Isn't that the hallmark of incompetence to think that you are better than everybody else.

When I was younger, I was also producing software faster, doing in 1 week what the senior were expecting in 1 month. Well, in reality it took the rest of the month to make it really works, but in my mind at the time I made it faster.

Nowadays I would still beat the deadlines with working software most of the time. Most of the time is the kicker. I know that I could need 1 month even though I will likely do it faster.

The more I know the more I find stuff that I don't and the less I would present myself as "the best".

Unreliable, delusional, unexperimented people, ... will reply to that offer.

>>Isn't that a sure way of not getting anybody competent ? Isn't that the hallmark of incompetence to think that you are better than everybody else.

Not necessarily.

There are two extremes. The first one is the Dunning-Krugger effect, where the person is so incompetent that they are not able to even recognize their incompetence, and instead think they are awesome. The second one is the opposite, called the Imposter Syndrome, where the person is extremely competent but believes that they are actually a sham who just got lucky.

There are many extremely competent people who are aware of their own competence, though.

What you're describing is know as the Dunning-Kruger effect explained in Unskilled and Unaware of It: How Difficulties in Recognizing One’s Own Incompetence Lead to Inflated Self-Assessments (1999).

Correct. And it seems people have only figured that out recently. Companies that really do only hire the top 1% are starting to post jobs that look like a whole lot more people are qualified for. See Instagram, for example:


Tons and tons and tons of people meet those requirements, and they surely get a mountain of resumes, most of which never even get a phone screen. But they don't scare away people who are actually qualified but just don't see themselves as 'rockstars' or 'ninjas' or 'the best in the World' - even though they are.

But they don't scare away people who are actually qualified but just don't see themselves as 'rockstars' or 'ninjas' or 'the best in the World' - even though they are.

Yah, they select against self-awareness. I think the dirty truth is that thinking you're the best in the world is more important to these companies than actually being good. If you have an outsized self-perception, you'll put up with unreasonable deadlines and terrible conditions on account of pride. That's attractive to someone who's in the business of exploiting naivete.

I'm a mediocre (96-97th percentile) software engineer whose expertise pertains more to the software ecosystem and economy. Knowing it (and myself) as well as I do, I know I'm not the best in the world. I even have that rare trait of knowing what I'm missing, and there's quite a lot in that category. But I'm more than good enough for 99.8% of the tasks that people need done, given reasonable learning time.

I am curious, how did you assign yourself such a precise percentile (96-97th)?

The model is here: http://michaelochurch.wordpress.com/2012/11/25/programmer-au... . I don't claim that it's perfect, but it's the best simple model of devonomics I can come up with.

These percentiles are a bit shady, I'll admit, because "programmers" isn't a well-defined group. I have a pretty strong sense that I'm 1.7-1.8 on the software engineer scale but a weak sense of how to define the population. Whether I'm 96th percentile vs. 99th depends more on how tightly or loosely we define "programmer". I tend to think of myself as about 94th because I define it tightly, but I would usually say that I'm 96-97th in congruence with the looser definition that a lot of people use.

How is 96-97th percentile mediocre?

Compared to the average programmer, that's about 5-6x. I'd imagine the Google Fellows are 40x or higher. I'm not down on myself at all, but I recognize that I have a long way to go.

Although if I read that today, I would think the company is being unrealistic and overreaching.

Granted, they made it, but this kind of language would not build faith in a company if I were to read a job ad for them.

The added Alan Kay quote would further fuel my doubt that the submitter is just another worthless startup guy thinking he can get the best of the best with some shallow knowledge of what hackers think.

I would have been wrong this time, but I don't doubt for a second there are hundreds of ads like this for companies that crashed and burned shortly after.

As a programmer that has occasionally worked in teams that delivered great software in 1/3rd the time that sensible people thought possible, that advert really appeals to me -- to me it sounds like someone who really wants to get stuff done, and that is exciting. Like the GP said, the language in this particular post is interesting -- it feels very different to the usual 'are you a ROCKSTAR NINJA PROGAMMER?!?!?!' stuff that tends to typify the worthless startup guys.

That kind of high-performance environment is all about the team, not the individual. Usually it's hard to know who you will gel with until you have some sort of history of working together, then you self select. However, like I say, this advert really appeals to me and could possibly help me get over that pre-selection process in my head.

I guess I have a more negative worldview than you do. And in this case, I would have been wrong, and you would have been right.

"The added Alan Kay quote would further"

Back in the day a large quantity of the emails that I received (or postings that I saw) contained some tag line or quote such as that. I'm not sure exactly over what years that practice disappeared, but I can't remember the last time I received an email with a tag line.

Bezos was just doing what was quite common at the time.

I don't agree. Sure, "ninja rockstar" has been over/misused to the point where it's beyond cliché, but at the core is a requirement for essentially supernatural skills and the implicit understanding that those are not realistic, so only the best need apply.

The language requirements gives the ad a certain solemnity in today's context, but I don't think it would have conveyed anything out of the ordinary in 1994.

It's not a "requirement" if it's not actually required. I can guarantee you that if someone were to go into, say, silicon valley and find every single developer who got a job in a role which was advertised at some point using the word "rockstar" or "ninja" and then compared the average talent/skill of that pool of developers vs. developers as a whole there would not be much difference, if any.

Edit: I think I slightly misread your post, so you have my apologies.

It is much in the same vein as the silly "requirements" of x years in y, z years in w, etc. And a stack of degrees. Usually it's just a way to keep the walls high to weed out the truly unqualified.

No, I understand it can't be formally understood as a requirement - but that's how it's phrased. Hence my comparison with todays unrealistic "requirements".

Nah to me "rockstar ninja" it is just a marketing term. Sure it sounds great and may fill arenas of recruiters but I still can't quite define what such a person is.

Is it a good programmer who knows all the bits and bolts of a given language or an adaptable developer who can master concepts from another programming language? Will the ideal candidate write good code AND good tests?

It's just too vague... Simply doesn't rock my confusion.

"than the standard 'seeking unix ninja rockstar' stuff"

People do that because they are mimicking seeing others do that and additionally haven't seen it enough (apparently) to realize how annoying and cliche it sounds. If the ads that the wannabees read said something different repeatedly (like "extremely talented c/c++/unix developers" and then listed 20 specifics for example) they might be mimicking that instead.

Look at how similar this copy looks to any other SV startups these days (minus some buzz words like "Cloud", "Social", "Disruptive", etc). It goes to show that hiring (well) is hard. Even harder to gauge whether or not a company is worth applying to from the job descriptions alone.

Phrases like: "You must have experience designing and building large and complex (yet maintainable) systems" are so vague and ambiguous that if I honestly saw this post from some guy named "Bezos" in '94, I would have written off as a jokester.

At least in '94 they haven't started using the word "disruptive" as if it's something you can do to a whole industry overnight. Thank goodness.

I agree, based on this ad, I would discard the company as an overreaching startup founder with a vague idea of how to reach out to developers and destined to fail within the next two years.

In this case though, I would have been wrong.

The difference that you're missing is where the ad happened.

It happened on Usenet, before most people had heard of Usenet, naming technologies that were brand spanking new. How new? The most popular browser was Mosaic. Netscape was barely founded and had not released a browser yet. http://oreilly.com/gnn/ was the first commercial website and was under 1 year old.

That shows a lot of knowledge of the tech world of the day. "Overreaching startup founders" weren't yet a significant concern. The target audience would realize quickly that this was risky, but unlikely to be a BS ad.

> It happened on Usenet, before most people had heard of Usenet,

Wait what? Usenet had been established for many years by 1994, and many people were using it. You could access it via several online providers and their Internet gateways. (Here's the first message from an AOL account: (https://groups.google.com/forum/?fromgroups=#!msg/comp.dcom....) )

That should say, "before most people were even aware of the internet, and who would go on to use the internet having never even known that Usenet existed." Back in '94, I think that most internet users used Usenet to some extent, but certainly "most people" in the world had never heard of Usenet, and never would, and still haven't.

I'd say you shouldn't just write that off as a freak incident though. You should be reconsidering how you read these kind of job adverts - obviously, a company can succeed even with this kind of approach - and it can obviously fail with it.

That means that the language or style used to convey a job opening is probably NOT a good indicator of the success of a company. Maybe it's time to rethink prejudices?

Sure, it's not a good indicator of success, but it generally is a good indicator for what kind of people you'll be dealing with and how you will be communicating with them.

Depending on what you want, of course, you should look for different things. I'm not going to make the judgment call on whether a company will be wildly successful - I don't have the background for that. I'll see if they have any chance of surviving and if they do, if it would be a pleasant environment to work in.

If you are looking for a list of significant USENET posts, including this one by Bezos, go to: http://www.google.com/googlegroups/archive_announce_20.html The list was created in 2001, when Google Groups reconstructed a huge archive of USENET postings.

I don't know why exactly, but it's really grating me that people have been replying to them. It may go against the spirit of Usenet but it would have been nice if Google disabled replies in their user interface, or at least filtered the display somehow.

It's like going to a museum and seeing chewing gum stuck to the sides of the exhibits. "OMG 30 Years..." - yes mate, thanks for that.


It's like going to a museum and seeing chewing gum stuck to the sides of the exhibits.

I agree with you, but on a fun side-note, I remember climbing the bell tower of a centuries-old church in the middle of nowhere, and being staggered by the amount of Victorian-era graffiti carved in to the walls.

I guess that the need to write "TOM WOZ 'ERE" goes back a long way.

It's funny, new graffiti is vandalism, but old graffiti becomes history in and of itself. At the Metropolitan Museum of Art in New York, they have an entire Egyptian temple that at some point was shipped across the ocean, and to me the most fascinating part of it was the graffiti left by occupying British soldiers long after it was built. Go figure.

It's time someone setup an alternative usenet archive to Google Groups.

The only people to do this were DejaNews.com, later acquired by Google.

Then there was the 'look, we recovered all this old stuff off tapes' which is locked behind captchas and against the open spirit of usenet. [edit: the tape archive is available here http://archive.org/details/utzoo-wiseman-usenet-archive]

Perhaps the Internet Archive should archive usenet?

Having an archive that is properly searchable, and which isn't broken when re-ordering the results, would be lovely. (EG: Google groups often falls over when trying to sort by date.)

I'd happily pay a small amount for that.

Jun 1982 - First mention of Star Wars Episode 6

"I wish Lucas & Co. would get the thing going a little faster. I can't really imagine waiting until 1997 to see all nine parts of the Star Wars series."

- Randal L. Schwartz, 6/9/82

You poor bastard...

From Linus' post announcing Linux:

> Hurd will be out in a year (or two, or next month, who knows) [..]

How prescient!

Aw the memories of reading the Morris worm thread with barely a faint clue of what a Unix system was like.

But who got the job?

Possibly Paul (Barton-)Davis [1]?

[1] http://en.wikipedia.org/wiki/Paul_Davis_(programmer)

That job posting has taught me so much in just a few paragraphs.

Even if I was the founder of a well-capitalized StartUp, I'd have great difficulty in promising people that work for me that we were going to be big. I just don't have that kind of confidence, yet it shows Bezos knew very clearly and intuitively that he was going to hit it big.

Sense and definiteness of purpose i think contribute way more to success than anything else.

I also think it's interesting to see that Bezos shares similar traits to Henry Ford in that he knew intuitively that it was possible to do certain things that were seemingly against the grain.

So much learning. So impressed with that job post, even if i can see the flipside for the candidates assessing that vacancy, since it does seem a little "out there".

I do wonder if that technique would attract someone like that today. In my own experience i've found hiring slowly and using non conventional methods for bringing people into a business seem to work quite well, so perhaps the economy has changed on this.

Either way, great posting.

Edit: spelling/grammar

> it shows Bezos knew very clearly and intuitively that he was going to hit it big.

This is ultra-common with startup founders. They believe unwaveringly in their eventual success, which is why they're founding a company in the first place. Who would choose to be a founder if they thought they'd be mediocre and eating ramen forever?

Some are right, some are deluded. Sometimes with the really good ideas, you can only decide which after the fact. For every Bezos there are 10,000 or more guys with the same intuition that failed.

Ultimately, that unwavering belief is table-stakes as a founder. It's what gets engineers and lawyers and marketing people to come on board and fight the good fight with you.

Why is the link to this job opening posted so often? I see it every 1 - 2 months here. It's not so interesting to see it posted as often as it is.

regular HN reader here, but i have not seen this posted before.

Probably because they don't always make the front page.

See this: http://news.ycombinator.com/item?id=4845645

Posted again, today (as of writing, of course).

Alan Kay quoted for truth.

That's a really interesting snapshot of 1990s nettiquette. On Usenet or Email you generally selected a pithy quote for your email footer. The trend went out of fashion I'd say around the turn of the century.

It does when the mailing public stopped using Unix and mail programs that were smart enough to shell out to external programs to compute the signature. Now we have webmail where only one company decides what programs you cannise to compose mail.

I find it odd that this article from two years ago is now No. 3 on Hacker News. :P

Are you concerned the subject matter will be out of date ;)

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact