[pg]

"If I were Paul Graham, these types of companies simply would not satisfy me (despite what look to be very lucrative payoffs)."

That's hilarious, because I'm the one who wrote all the summaries of the startups that she's judging them by. I deliberately express startup ideas as x for y whenever possible because that's the most efficient way to describe a startup in a few words. Airbnb, for example, was in its day "eBay for space."

On Demo Day we accompany each startup's name with a brief summary of what they're doing. The goal of these summaries is not to convey the full breadth of a startup's eventual ambitions, or even what they're currently doing, but just to help investors watching 65 presentations remember which company is doing what. TechCrunch reused these descriptions, as reporters often do. And now this woman has read the summaries printed in TechCrunch and is treating them as if that were all there was to know about the startup. It's like a game of Telephone (http://en.wikipedia.org/wiki/Chinese_whispers).

My standard advice to startups presenting to investors is that it's better to start with an overly narrow description of what you do, so that investors at least know what you're talking about, and then expand from there.

See http://paulgraham.com/investors.html, particularly items 1 and 14.

It's surprising to me that someone who wasn't even at Demo Day would feel confident enough to judge the startups' originality based on the 3-5 word summaries we use on the schedule to help investors keep them straight in their heads.

I was about to say that it's also surprising that people would upvote such a post. But on reflection it's not that surprising.

[tlogan]

I will explain what I think from the point of view of an engineer: none of the companies inspired me to apply for a job to work for them. Probably the problem is that I'm not smart enough to see the grand vision: the vision of these companies is just not clear for mortals like me.

For example, I was hoping to see something like Word Lens on iPhone: I can see how they can rule the world. That is something frighteningly ambitious.

[pg]

Like I keep saying, the descriptions she's writing about were not meant to sell people on the company's vision, but simply to help investors remember which was which.

[justin]

How has Word Lens taken over the world? It is an awesome technology, but there are many awesome technologies, and not many of them become amazing businesses. (However, that said, there are awesome technologies like the Word Lens in this class of YC).

When Facebook started off people just said it was MySpace for college students. Those with imagination that joined early now reap the benefits of vision.

[mattmanser]

Vision? For a third generation social network?

I certainly wouldn't use that word. You wouldn't say a dating site had vision. Facebook falls in that category. Awesome achievement, but no real vision and it's going into no other fields, like google did for example.

[RuggeroAltair]

Although I don't mean that the other ones aren't worth working for, I'd say I'm surprised that you don't think Flutter is an interesting company to work for.

It seems like it could be a very exciting playground for an engineer.

[wisty]

Here's a few X for Y things which are likely to succeed:

X for the masses.

X for people who don't like computeres.

X for people who don't speak much English.

X in the cloud (way overplayed).

X for free (but good luck getting a low enough cost of customer acquisition!).

X for people who are locked into Y.

X for large businesses (requires connections).

X for businesses which aren't great at tech (X better be something they really really want).

[jrkelly]

I wasn't at the Demo Day, but it's hard to consider making car repair more efficient to be a problem that really matters. Not saying it can't be a decent business, but your essay about ambitious ideas was pushing people to do better.

[tptacek]

I am just baffled by the idea that people here think there isn't a huge dent to be made in the universe in auto repair. There are 723,400 auto mechanics in the US alone, making a median of $35k in an incredibly inefficient marketplace that fleeces and inflicts misery on millions of working class families (who get to work and their kids to school in cars, not fixies) every year.

[moocow01]

Lets not go overboard - I'm sure some customers have their complaints but your making the car repair industry sound equivalent to today's Wall Street. I genuinely rarely hear people complain about car repair - now if we were back in 1970 or 1980 Id have a different opinion.

[gruseom]

He's not going overboard. When you don't know anything about cars, it's really hard to find a mechanic or car repair shop you can trust. One reason I always try to help people who ask me any question about their computer is that I just think of how ignorant and defenseless I've felt when my car breaks. I do what they do: find the person in my life who knows the most about it and cling to them like a drowning man.

If the Your Mechanic guys can solve this problem, they'd be rewarded with fantastically loyal customers (including me) in a big market. Actually I'm a little bewildered why anyone would see this as an unambitious idea. It sounds hard to me.

[kerryfalk]

Agreed, it's not an overreaction. As an auto enthusiast and someone who has a large group of friends that are also auto enthusiasts we are acutely aware of the issues our non-enthusiast friends have with the auto industry.

We're aware because they come to us with their problems and seek advice. Everything from helping to buy a new car to trying to understand what the mechanic was telling them is wrong with their car.

When I was younger I worked at the front desk of a reputable auto shop. It was very common for customers to mistrust us, and they had good reason to feel that way, too. Not because our particular shop was bad - it wasn't - but because many others are.

The auto industry is huge and has many huge segments that can be targeted. There is room for many companies to grow into significant players.

[yabbadabbadoo]

He also says this in that same essay -

If you want to take on a problem as big as the ones I've discussed, don't make a direct frontal attack on it. Don't say, for example, that you're going to replace email. If you do that you raise too many expectations. Empirically, the way to do really big things seems to be to start with deceptively small things. Want to dominate microcomputer software? Start by writing a Basic interpreter for a machine with a few thousand users.

So for all we know, there are a bunch of YC companies aiming to take over the world, but we just don't know about them since they start off by tackling a very specific, narrow niche. Take greplin for example. I could envision the founders aiming to eat Google's lunch one day, but they certainly didn't launch by announcing that to the public.

[ajju]

Every car owner wastes time and money at repair shops that tend to be so opaque and corrupt that there are sitcoms about it. Most mechanics employed by these shops get paid much lower than they can charge as an independent mechanic.

Yet, the middlemen who run the shops make many times more than the sum of the cost of labor and parts.

Are you saying that trying to fix this for the 250M+ car owners in America is a problem that doesn't really matter?

[jrkelly]

Given the caliber of people that YC can now recruit - yes, that problem seems like a waste of their talent.

[pg]

Do you feel confident that if you had seen Altair Basic in 1975 you would not have said the same thing about it?

Like humans, startup ideas all look similarly small at first.

[wglb]

Perhaps off topic, but I bought an altair in 1977 and did a bunch of consulting based on it. Accounts receivable, inventory for a local furniture store. I was insanely excited.

Edit: (In fact when the intel 4004 came out, I was a little frightened that it was going to disrupt our business built on the Sigma 5, but then I realized that it would need software like the stuff we were writing.)

[vectorpush]

I have to disagree. For many Americans, car repair is one of the most expensive and opaque processes out there. Most people who drive have a critical need for that transportation, so when their car won't go, they pretty much have to suck it up and pay whatever fees the mechanic pulls out of his ass.

My point being, this type of disruption would be a very big deal for your average joe.

[waterlesscloud]

I'm curious what it is about Your Mechanic that will make the process any less opaque for the end user?

Ok, so they arrange for the "best" mechanics to come to you to fix your car. The coming to you is cool, and certainly improves convenience. But lessens opaqueness? Not seeing it.

What guarantee is there that the whole thing won't just turn into something like the big car repair chains are now? "Looks like you got a burned out frabulator, that's gonna run you about 14 hunnert." And the customer doesn't know if they do or not, they just have to trust this "best" mechanic. What about this system prevents that, which is the part that most people hate about dealing with mechanics. The big chains already have standardized pricing, that's not the problem here. The problem is trusting that the work being done is what needs to be done.

What's the criteria for "best"? How will they keep the criteria from slipping as they go through explosive growth?

It's not that I don't think the system can't be disrupted, it's that I think it's going to be a lot of work for marginal improvements. I don't see how it's going to be transformative.

[pbreit]

1) They are probably going to try and address most or all of your questions.

2) If they are successful in one industry, they might be able to apply the model to many/all industries.

[evgen]

If they are even mildly successful in this one niche they are going to see four me-too startups in three months (one of which will be bought by a national auto parts chain before the year is out) and see the core idea applied to five other niches before they close an A round.

[simonw]

Are you saying that they shouldn't bother because if they are succesful someone else might copy them?

[pbreit]

Possibly. But not a given that the copycats will be successful. Can you point to any examples of your thesis?

[derrida]

...and eBay sold trinket-junk at first. I wasn't at demo-day but it's easy to see a path of iterations that goes from mechanics repairing cars today, to plumbing, to other trades to people designing and building real-world houses from their computers tomorrow.

I'm not saying this is where it is going, and as an investor you can not value a business based on these far-out ideas. But as a hacker/entrepreneur I imagine the founders are having dreams every bit as lofty as the one I just presented.

[waterlesscloud]

I don't know if it's fair to say it's like Telephone. From what you're describing, she's reading exactly what the lead investor in these companies had to say about them.

I mean this in the best way possible, but if the descriptions don't sell the vision of the company, maybe that's something that needs some attention?

[pg]

I meant it's a sort of semantic equivalent of Telephone, where the message stays the same but its meaning gets changed.

As I said, those descriptions aren't meant to sell the vision of the company. They're just so one can remember which is which.

[bertm]

It seems difficult to analyze the degree of ambition in any given startup from PR. Although ambition makes for a good story, it doesn't necessarily sell product/get users. So it seems, early stage startups focus on simple messages which the audience will understand. Look at the evolution of PR coming from Air BnB over time.