Is Y Combinator the University of Internet Startups? | Hacker News Reader

Is Y Combinator the University of Internet Startups?

mashable.com

38 pointsthomasswift30 comments

It's certainly a better metaphor than boot camp. Though, interestingly, in this case the metaphor is accurate even to the misconceptions. This writer thinks that YC's most important role is as a gatekeeper. That's a common misconception about elite universities too. In fact the most valuable thing about going to one is what you learn there, and so it is with YC too, I think.

One way to understand where YC's "center of gravity" is would be to measure how I spend my time. If YC's role was simply as a gatekeeper, my job would be done as soon as we'd done interviews to select each new group of startups. Boy would life be easy if that were so! In fact about 98% of the work is still ahead at that point.

I tried to explain what we actually do here: http://www.paulgraham.com/ycombinator.html

Another useful part of universities and YC both is what economists call signaling. Going to a good college sends a message to those looking for people who can compete at that level. Some contend that it is the most valuable part of college for the average person.

The YC alums are a significantly under appreciated benefit of YC, though people understand why universities and their alumni networks are good.

I don't disagree with you that the education's the most important part, especially to the person. The gatekeeping isn't the hard part, but don't underestimate how important it is to the function of a society. Without the university system, we'd probably collapse, even if it is somewhat inefficient.

I was trying to bring more to light the trend in how VC is being conducted and get people thinking about how that will affect venture capitalists and especially entrepreneurs. The "gatekeepers" have a defining role in the direction societies take, especially in terms of development.

"They could sing campfire songs in all the classes, as long as admissions still worked the same." (or something to that effect)

I tend to believe that the most important part of a university is neither the learning, nor the signaling, but rather the networking.

The introductions made through university connections tend to prove invaluable over time, and some of the alumns I've spoken with have the same view of YC.

In-context version:

"It's no coincidence that startups start around universities, because that's where smart people meet. It's not what people learn in classes at MIT and Stanford that has made technology companies spring up around them. They could sing campfire songs in the classes so long as admissions worked the same."

(http://www.paulgraham.com/start.html)

Don't forget that when you're given some prestigious charge, the human reaction is to try to do something with it.

Much of the difference between hackers at MIT and elsewhere may be that at MIT, they're supposed to perform technical wizardry. Elsewhere, there are a lot of people who could do great things, but for some reason or another, don't.

To whom much is given, much is expected in return.

For a lot of people, it certainly adds extra motivation.

With Great Suit Comes Great Responsibility http://www.theonion.com/content/node/34279

Sorry. This was too funny to not post:

Meeting someone from Harvard, they will undoubtedly display faux modesty and say something along the lines of, "Oh, I went to school in Boston." I imagine they want me to guess MIT, just so they can say "Harvard" proudly in response, but instead, I say something like, "Hmm. You really look like a Northeastern kind of guy. BU, maybe?" This puts them in their place, and they will scramble away, crying crimson tears all over the floor.

More at http://www.radaronline.com/features/2008/07/leveraged_sellou...

I always wondered why Princeton does not produce more start ups...no shortage of riches or intelligence in that area. My guess is that the academic climate is centered around pure research...am I far off?

You're not too far off.

There are two very different populations at Princeton: undergrads and grad students barely recognize each other's existence, much less talk with one another, and therefore don't found with one another.

The undergrads are constantly, constantly marauded by recruiters from wall street and consulting firms. Many of the rest are trying to get into med or law school.

The populations in the hard sciences, engineering, math, and comp sci are quite small. Most of those are in fields which don't found companies out of undergrad much -- molecular biology, for example. And of the hard sciences, the dominant cultural influence is from foreign students holding international science Olympiad medals. It draws the fire from the notion of founding a startup.

The rest are grad students. Some of these do found startups, but they're fewer, and farther from capital, and often quite set on an academic path. Plus, unlike at MIT or Stanford or Berkeley, there is, for some reason, an urgency on the part of the administration to get you graduated by your fifth year. Don't ask me why.

Also, it's just not as strong of an engineering school in most of the disciplines or fields that are startuppy. Even within comp sci, there are mostly people studying large systems, and theory, and parallelization, and security, and computer science policy, and compilers, and so on. Not so many in HCI. And since most startups succeed where they can fill a before unarticulated need, that hurts them.

Finally, it's a little bit of the product of the admissions department. Many people are admitted based on extracurriculars which look good on paper, but neither stretch one's abilities, nor are particularly novel (if they were, they'd be hard for admissions to understand), nor are particularly technical. Bleeding heart stuff. A lot of the startups that I saw come out of Princeton (which friends got involved in) are either vaguely defined environmental/political websites, or connected to academia in some way. Not to say Princetonians don't have the chops: both Jeff Bezos and Eric Schmidt came from the EECS department.

All spot on, except for this: "Finally, it's a little bit of the product of the admissions department. Many people are admitted based on extracurriculars which look good on paper, but neither stretch one's abilities, nor are particularly novel (if they were, they'd be hard for admissions to understand), nor are particularly technical."

This is true of every high-end school, and is not especially pronounced at Princeton. The biggest issue is the first one you pointed out, wall st. and the consulting companies snap up all the talent.

I don't think it's fair to call it a misconception. It is a claim or maybe even an accusation. One that Universities are are commonly accused of. One that is being extended to YC.

Universities might similarly defend against this claim by pointing to the resources spent on educating as opposed to filtering. But I don't think that this is necessarily a good defence. If anything, it points to an institution's internal priorities, not its value to potential students.

It seems to me that if YC is consistently successful, it can't avoid becoming a filter in the eyes of investors. This would eventually come to affect applicants' perception. If YC increases a company's chance of attracting investors, this might become the motivation for participation. Investors listen to pitches or even invest because of YC's track record. Startups apply because of investors' perception.

Similarly, University applicants often think along the lines of creating an impressive resume. This is partly because it is true (an Oxford degree is better at getting jobs then an unknown one) and partly because people tend to undervalue educational potential of things beforehand.

Throw in the fact that top Universities get the power to choose the best applicants & it becomes difficult to tell if they are picking winners, creating them or just in control of some bottleneck.

In any case, I think it's a point of interest.

Some bait: http://www.techcrunch.com/2008/09/04/secretive-storage-compa...

Well, it's better than American Idol at least.

There's an additional component other than filtering that YC serves: it provides a pool of talent and experience that helps you out of jams, but without something like 'formal instruction'. The best features of a finishing school and guild (apprentice/journeyman/master) are represented. A community -- not so specialized as a technical or vocational school, but people who are likely to have common drives and purposes, and whose cross-disciplinary skills enhance the whole environment.

Maybe a question that needs answering, either by pg or by YC veterans (graduates?) is what additional purpose doesn't YC serve that it needs to?

YC is less like a university (where they teach building blocks) and more like a highly selective vocational school (where your skills are directly applied)

I considered going to grad school or taking that money, time and effort and use it to start a company. It was a very easy choice given my current situation.

What made it easier was some friends just finished grad school and now have an average job with slightly higher pay - that doesn't sound appealing.

YC does not look much like a University. Apart from the application process, there is not much to YC that you can find in traditional academia.

YC looks like what a University should be. YC is a model that is constantly changing and ameliorating itself using what it learned in previous semester.

At YC there are no professors. There are mentors. OMG. Imagine the amount of knowledge students would actually accumulate if they were taught from a young age to find solutions to problems and not take someone else's answer for it.

By the time you get to university you have already developed a curious mind and if you were to use that for 6 years to challenge your own thoughts on any subject and seek your own answers, you will come to the work place with a hell of a ability. Mentors should be available when we need them or to frequently stop and take a look at what students work on and only only suggest how they can do better or how to look at problems in a better fashion

YC tells you "Ok, so you want to be the solution to that problem, go ahead wow me". There is no A or B. There is only a continuous effort to fix the problem and after you fix it, keep improving the solution.

YC is not like a University, but Universities (Heck the entire Schooling system) can learn from this model.

If anything, from what I have read, it seems more like a university at the graduate level. You know of a problem, you have the tools to tackle the problem, but the solution is a big question mark. Your goal is to create a solution. In essence, you are doing original work.

true but the diff is that with startups, if you tackle the problem well, you will be heavily rewarded

I don't deny this at all. That's the glory of capitalism! The problem I have is that I can't find a problem that I could do on the cheap legally. I'm heavily involved with finance and market data. I can think of numerous ways something like Bloomberg could be improved, but the costs are prohibitive when you consider all the data feeds. The only other idea I have managed to come up with is a Digg-like site with a social network aspect attached to it. The idea is people submit just like on Digg but they also make recommendations as to what would happen in the market. They then earn a credibility rating based on whether or not their prediction was accurate. The problem with that is this: I don't think there are enough people who would use it to make developing the site worth my time.

Make it private and charge a hefty monthly fee to participate. Apparently in Wall Street scarcity and limited access is what makes people rich.

That's what I am trying to avoid. One of the problems I see with Bloomberg is that it costs $1500+/month/terminal. My idea involved the concept: data for the masses.

This would only work if your product gives the subscriber an edge over their competitors. They wouldn't pay for a product like Google Finance because it doesn't give them an edge.

Your plan would work for Monitor110 if the product actually worked.

Sounds like a pretty good idea. You should go for it.

Tell that to the boys over at Dealbreaker. They've been waiting for a Bloomberg killer for eons now.

But i don't think it'll be easy to build a Bloomberg killer because the main reason why Bloomberg is so popular in the financial circles is that it provides breaking news and lots of historical financial data that you won't find elsewhere.

So it is not much of a technology edge than a content edge.

As for the Digg-like site, you are right in saying there would not be enough people who would use it. I would think most online investors are those who thrash talk on Yahoo Finance boards. A site like yours would require input from intelligent investors, which unfortunately form a small subset of the online investing community. Hence the lack of scale may render this idea ineffective.

I used to work for Bloomberg's biggest rival, and I know exactly why they are so popular: they understand their end users. RivalCorp doesn't really get it at all. R (as we'll call them) tries to sell to a bank's IT department, and Bloomberg tries to end-run IT and sell to the traders (guess who has the most power in an investment bank). R spent years and hundreds of millions retrofitting TCP/IP onto everything (which traders simply don't care about), Bloomberg spent that time and money setting up deals with tailors and wine merchants (I'm not kidding) for users of its terminals. They made their users feel like insiders, a privileged elite, which is something all bankers crave. Plus, Bloomberg terminals are sexy. They look like something out of Star Trek. R terminals looked like beige-box PCs.

Sure there are technological reasons too, but Bloomberg's success is cultural first and foremost.

That's very true. I'm guessing you used to work for ThomsonReuters/Telerate.

Monitor110 tried to be the Bloomberg killer and they had the street cred to pull it off. But they lack the content that traders want. So at the end of the day, content is one of the most key ingredients of any Bloomberg killer.

I think its more of an apprenticeship

Reply on news.ycombinator.com