Sergey Brin’s Resume (1996) | Hacker News Reader

Sergey Brin’s Resume (1996)

infolab.stanford.edu

134 pointseniz66 comments

Earlier in dotcoms, I dug up his old email address this way, for a reason that seemed important at the time...

Some Google megacorp faceless bureaucracy combat drones (that was their initial approach) were trying to force me to give up a domain name, and implying they would take legal action and "prevail". I couldn't afford a lawyer, and I had better things to do with my time, but...

I wasn't yet using the domain name, and probably would've let it expire if they'd never contacted me, but their heavy-handedness raised a moral concern. At the time, Google was considered good, and clearly it was very important to humanity at the time that Google be good and stay good.

The whole "don't be evil" seemed to come from recognizing that Google would likely be very powerful (this was pretty clear earlier, as soon as you used their prototype, and realized it was not only better than everything else, but that they were more competent than almost all other dotcommers). They were declaring upfront that they took the responsibility seriously, and wouldn't abuse their position. And there was some early evidence that they believed in that (such as in objectivity of rankings, and being very clear about what was sponsored messages and not).

The domain name in question had been intended for a social commentary parody, of some social media manipulation behavior that had just started to emerge on a lesser Google property. I told the threatening lawyer-types that. I also pointed out that the domain name obviously would never be mistaken for a Google brand, and that I'm pretty sure that the bit of intentional similarity to one of their brands would be considered protected use in the US.

When they still wouldn't back off -- and since I couldn't afford a lawyer to argue the points, but I was concerned -- I looked for the founders' old email addresses, and used Brin's (IIRC) to initiate a domain name transfer to him. Then I told the lawyer-types (and a PR contact there) something like, if they wanted the domain name, they'd have to talk to him about it, and maybe have a "don't be evil" discussion. IIRC, they said they'd wait for the domain to expire.

This is certainly ironic given how Google is positioning itself to become one of the most evil corporations in the history of the planet. Will it get there? I hope not but things are not looking positive.

Most evil corporations in history? Really? I mean they have nothing on the British East Indian Company, which took over and ruled over India as a private enterprises.

The most interesting part of this resume is hidden though. Click view source to see it..

  <!--<H4>Objective:</H4>
  A large office, good pay, and very little work.
  Frequent expense-account trips to exotic lands would be a plus.-->

Great find!

Another hidden comment on that page of historical interest:

EDIT: While looking for more information on the data mining link, stumbled onto a 2015 HN thread[0] in which the same two hidden comments were cited by two different users[1].

[0] https://news.ycombinator.com/item?id=9055516

[1] https://news.ycombinator.com/item?id=9055582

2016: https://news.ycombinator.com/item?id=12561966

2015: https://news.ycombinator.com/item?id=9055516

2014: https://news.ycombinator.com/item?id=7448568

2010: https://news.ycombinator.com/item?id=1158718

Hah, gotta love that animated photo: http://infolab.stanford.edu/~sergey/

Clicking around, the last class he taught was with another student... Larry Page.

http://infolab.stanford.edu/~sergey/349/

> World Wide Web

> Research on the Web seems to be fashionable these days and I guess I'm no exception. Recently I have been working on the Google search engine with Larry Page.

http://infolab.stanford.edu/~sergey/

Brin was on the Python train early. At the time, Guido van Rossum worked in Reston, VA, only about 30 miles away.

I feel like Google created the "Python train" to begin with. Maybe in an alternate universe Brin could've picked another random obscure language and it would become the most popular language today. It would be endorsed by one of the most successful company in the world who can pour resources into developing it, hire their BDFL, etc.

I see this claim repeated often. It seems logical enough, but is there any real evidence for it?

Not sure if this is entirely equivalent, but couldn’t the same be said for PHP with Yahoo/FB?

I had no idea Wolfram was older than Google

are you serious? wolfram research was founded in 1987 and google in 1998. what gave you the impression otherwise?

Oh yeah. I believe a version of Mathematica was included with the original NeXt workstations back in the early 90s.

Wolfram is like the sole survivor of the AI winter.

> It is unique in that it is in written mostly TeX

And they say, always proofread your resumes! :-)

He didn't proofread his resume and is rich beyond our wildest imagination. Maybe this actually counters your argument.

Walk the walk, but luck is a cruel mistress

Another one:

> to extrapolate how much the you will like some other movies

Number of typos and grammatical errors correlates well with receiving an offer in this small study: http://blog.alinelerner.com/lessons-from-a-years-worth-of-hi...

"Too academic... during the phone screening he kept going on about non-euclidean geometries and eigenvectors and stuff. His C++ skills look pretty good, but we need someone with strong skills in Oracle, Visual Basic and ClearCase ... and with at least 5-10 years of industry experience."

Here's his little bro's geocities

https://web.archive.org/web/19990129053912/http://www.geocit...

I'm bloo if I was green I would die

It seems that as a college freshman he was already a competent programmer.

That was not uncommon back in those days. Personal computers in particular were a relatively small niche. Given the limitations of application and difficulty to use, a much higher proportion of those who could use a computer could also program one.

Most of my computer friends and I were decent programmers by the time we entered college in the early 90's.

It must be fairly common even now for people doing CS degrees, right? I didn't do CS, but I'd been programming for 5 years by the time I started college (I partly didn't study CS because I already knew how to program to a reasonable standard...)

It’s not common in Australia at least. I wrote my first code in year 1 of CS. Most people in my class were in the same boat. Tutored the course after graduating and it was the same; in a class of 30 you’d usually see 1 or 2 who had any prior knowledge and breezed through it.

This was 10 years ago but I doubt it’s changed significantly. I think there’s also lots of people who don’t enroll because they think prior knowledge is required, which is a hard barrier to overcome.

Most of my peers had maybe a single class worth, if that, of programming experience by the time they started college to pursue CS. I would say the majority didn't. I wasn't at a university with heavy focus on computer science, so maybe it depends on the institution and the caliber of students.

Back in my day, almost exactly 10 years ago, relatively few college freshman had significant experience coding, beyond understanding basic syntax (probably <10%). Most of the most blatantly experienced coders were bored for the first couple months and dropped the course/school to make heaps of money, not realizing understanding syntax is like 2% of Computer Science. So the people most vocal about how easy the coursework was were at a significant disadvantage in the long run.

But it just goes to show that that 2% is all you needed to make heaps of money ;)

I mean heaps of money is relative to a college student fresh out of high school. Everyone I personally know who went that route seriously regrets it and hit a glass ceiling within 3 years of working. Their careers tanked before the rest of us even got started.

Maybe it’s different at other companies but in my company the people with fresh CS degrees barely know how to code.

Yeah, it's common to see people who already know how to code in college. Plenty of us are working on real cs projects, and with the internet, its easier than ever with the amount of resources we have to learn with.

I took my first programming class in 1983, after school while in the 2nd grade I guess (then reinforced by a BASIC/Pascal course in a Mississippi high school). Being just BASIC, I still had a leg up when I did C in college 10 or so years later.

Ironically, my second “better” Seattle-area high school had given up on programming as something to teach, I’m sure they’ve changed their tune now :).

Sure, in the 1980s the schools I went to in small-town Indiana had plenty of Apple IIs (and the Acer clones) that anyone could use. Kids magazines in the public library would usually have BASIC source code in the back that you could type in to get a simple game or logic puzzle.

Like most educational institutions they used June fiscal years, so they would buy technology near the end of the fiscal year with leftover money. Those computers would sit unopened in their boxes until September, unless you offered to take it home for the summer. I'd guess that I had a brand-new IBM PC (that cost 25-30% of our family income) at least half of the summers. You were kind of doing the school a favor - they didn't need to worry about storing it somewhere and you did the work to set it up and make sure it worked. And if a kid wanted to learn over summer vacation, they certainly wanted to help facilitate if they could.

Despite all that, I can remember going to the state programming contests in Indianapolis in 5th, 6th, 7th grades and being blown away by how much better than me all the other kids were.

When I was a freshman in 1995 'could program a computer' usually meant being able to write simple routines in BASIC or Pascal, not developing parallel algorithms in C for supercomputers.

At the time Brin was a grad student, as you said, it was not-uncommon, but also not the norm, even in top departments.

The students who had programming experience from other than just their classes had a big advantage for systems work, compared to other students.

Though there were also a lot of students who had only undergrad programming experience, and had a lot more learning to do. Which is perfectly fine, so long as they know they have something to learn, and want to keep improving.

(The only annoying ones are those who treat professional and collaborative work like they're only trying to pass a class -- including the analog of only trying to get a homework assignment past the TA -- and think this is what everyone does. I suspect the ones I've seen would pass a leetcode CS101 interview just fine, because it's pretty similar metrics as getting into and passing CS classes.)

Maybe if you had the money to grow up with a computer and had people around to help you. I didn’t program until 1991 in Turbo Pascal. It was a terrible experience and somehow it was something you were supposed to already know. On the Vax or Unix machines in the lab, literally nobody was helpful. If you asked a question people would bark “read the man page.”

Is that an unreasonable expectation? Programming is the easy bit, no? (Not including design and style)

I don't know, solving parallel algorithms homework assignments was one of the most mentally challenging things I've ever done. My brain was literally overheating sometimes :)

It does show that your parents involvement in your teaching plays a part, as does being in the right place at the right time and culture of the country you reside in. I'm not knocking his own effort either. When you look at the developments and achievements coming from the US, you have to wonder why does the rest of the world get it so wrong?

I think in the US, it’s either really great, or fairly sucky to live.

In the rest (of the western world), it’s just always meh. They’re optimizing for different things.

The US as a country is also just bigger, so it will have more successes in absolute terms than other smaller countries. The same thing is happening in China now and India will soon follow.

throughout history, depending on what country/civilization was at the top, a lot of scientific achievements came from there (egypt, china, india, Greece, rome, muslim cultures, portugal, england, france, germany, and finally usa etc...). I'm going to say a-lot of scientific innovation is sociological in nature. Meaning the confidence of a country is directly correlated to the amount it invests in scientific research and corresponding results. Once that confidence runs out, due to a host of reasons, the country falls into a decline and scientific pedigree moves to a different place.

Would be interesting to know which subset of math he thought was most useful. There is just too much to learn.

I'd be curious to see what his "Movie Ratings" project mentioned there was like at that time

Something like movielens, a project they eventually bought.

You do a simple k-means calculation of your friends or closest matches (if you have clustering or svm, I guess he didn't have) and then get predictions for free. Something like the Netflix recommender system, just very primitive.

Go Terps!

Leg up from dad. He even got work experience in the actual department in which his father worked...which surely wouldn't happen today.

...but this is kind of a Google thing. Page was the same (and somehow still didn't know how to code Backrub despite having a BSc and MSc in comp sci and the link idea came from Robin Li...made quite a career out of claiming the genius of other people as your own).

I am sure he is a business genius, there is no doubt about this and there is not doubt the Google search engine was the best solution at the time they launched.

Many millions of people have this level of privilege worldwide, but very few have founded Google-sized companies. Acknowledging the distorting effect of privilege is important, but this seems like you’re throwing out the baby with the bath water here.

You kind of prove the parent's point though. Given a large enough sample size of all the people with privilege, it is conceivable to imagine that Google-sized success would happen to someone in that group at random, and other explanations could be simply overfitting to the data.

(Not saying this matches my personal worldview, but just analyzing the commentary)

> Given a large enough sample size of all the people with privilege, it is conceivable to imagine that Google-sized success would happen to someone in that group at random, and other explanations could be simply overfitting to the data.

This is true of all groups, though, so doesn't give us any information.

Enough information to hero worship, apparently.

Not everyone wants to found a company though - you can just enjoy your privileged high quality of life.

> His father is a mathematics professor at the University of Maryland, and his mother a researcher at NASA's Goddard Space Flight Center.

I mean, he still needed to get a job

Not necessary I'd say, depending how good relationship one has with the parents :)

Can you elaborate on

(1). Larry not knowing how to code Backrub (sounds like you don't just mean that at first you don't know how to code something that has never been done before).

(2) Larry claiming other's ideas as his, particularly "the link idea".

Robin Li (who founded Baidu) created link-based search ranking (and patented it in 1996, two years before the famous papers).

Scott Hassan (a Stanford research assistant) coded Backrub. Page took a run at it but it didn't work. And it had actually been done several times before (again, it just sounds like you have no idea you are just assuming these guys were geniuses...several mining applications already existed, the largest came from DEC).

backrub/pagerank is a specific kind of link-based search ranking. Link-based ranking was well-know before e.g. citation count. Was Robin Li's the same kind?

What were the "several times" backrub had been implemented before?

A "mining application" is a very general concept, and does not imply a specific algorithm.

I was not assuming but requesting information; and not "again" but once.

FWIW I'm intrigued by the school of thought that Google's success was largely due to just giving people what they want. Instead of crowded portals, just search, fast search, and search that's relevant. When I tested it at the time, I found competing search pretty similar for relevance. But google was faster.

Too late to edit, but I'd forgotten paid-placement was a thing in search results before google. Can you believe it?

Of course, adwords has similarities (and copied from overture/goto), but don't masquerade as search results (though becoming less distinguishable over time...).

you are what is called in some circles "a hater".

Other than steve jobs, pretty much 95% founders these days come from upper middle class with professional parents (doctors, engineers, professors, lawyers etc...)

Reply on news.ycombinator.com