Paul Graham on Mixergy, today 2pm EST/11am PST | Hacker News Reader

Paul Graham on Mixergy, today 2pm EST/11am PST

mixergy.com

192 pointsvaksel77 comments

Open articleView on HN

My questions:

- What advice would pg (2010 instance) give to pg (2004 instance)?

- What property of the 4 yc partners, if removed, would do the most damage?

- What has surprised you the most?

- What is the biggest mistake people have made?

- What is the biggest thing that will have to be different (about yc) in the future?

- What kind of yc startup is most likely to hit a grand slam home run?

Wow, these are good questions. But it's 10:58 so I'll try to answer them afterward.

Edit: Answers:

1. Not to be influenced by credentials. Where people went to school in particular.

2. Our own determination, probably. We need it too, not just the founders.

3. How many people just give up.

4. That founders have made? Building stuff users don't really care about.

5. Doing everything on a larger scale.

6. The kind with the right founders.

People are hard-wired to give up. There has to be a carrot, stick or both to overcome it. Ideally, just wanting to make a change, or being really excited by something is what keeps people going against the odds: like Steve Jobs. Developers are basically problem solvers, I suspect giving up occurs when a certain set of problems seem or have proven to be insurmountable.

My questions:

- Is he always calm? Does he ever get angry and throw things around?

- Has he ever 'cut off' a YC company from access to YC?

- What is the current relationship between YC and the early successful companies like reddit and loopt. Is there any? Do they call ever now and then?

- What proportion of YC applicants come from name-brand universities?

- If a YC company becomes a lifestyle company, does YC still profit? Or does YC only profit on a big sale?

- If a YC company folds and significantly changes, like DraftMix, is YC still involved in the new company, or do they lose their 6% when the company changes?

- What mobile phone does he use?

- Would he accept a patch to get rid of the tables on news.ycombinator and change to css?

Additionally:

- What is his opinion of the risk/reward trade-off in attempting to do lots of co-marketing between early stage companies?

- Is any publicity good publicity when you're small?

Only my eyebrows; we once sold our stock in one; we talk quite frequently to the Reddit and Loopt founders; I'm not sure, maybe 1/3; we only make money if there's an exit; it depends whether the new idea is the same company-- bluefrog is; some crappy old Motorola flip phone; you would have to reach deep into the Arc libraries to do that.

What's the reason behind using the motorola phone? to avoid distractions?

You mean not getting an iPhone? Yes, because I already spend too much time online. The last thing I need is for it to follow me out into the world.

- What qualities make a strong leader when taking the business from the garage to the next level.

Andrew Warner (founder of Mixergy) has a great interviewing style. Should be good.

I think Andrew is doing an amazing job. My respect for him skyrocketed when I watched his interview with the bribe kid. He handled that extremely professionally.

He's really great at interviewing these guys. I always have an interview ready on my iPod. I hope he keep doing this.

Absolutely - he always seems to ask what I want him to ask next.

And a podcast a day - astonishing output and great interviewees. The advice in these interviews, along with advice on HN, constitutes my main source of not-on-the-job entrepreneurial learning, a sort of 'startup MBA'.

If you have any questions, add them here. I'll check these comments about 10 minutes before we go live.

The focus is the biography of Y Combinator.

Along the way I'd like to learn how pg gives the kind of insight that past Mixergy guests said turned their businesses around.

How much time should be invested in an idea before approaching YC for seed funding (or more importantly the mentorship) and if rejected once does it hurt your chance at being accepted a second, third of fourth try with different ideas?

You should talk to us when you're ready to go fulltime on the startup; it doesn't matter how much time you've spent thinking about the idea.

We often fund people on the second or third try.

As a programmer with a decent design ability... what talents should be sought after in a co-founder when building a web application. My gut feeling is a UEX/Design expert would provide the most benefit.

Determination.

What's usually the biggest mental hurdle for hackers to transition into entrepreneurs? An understanding of their market? An empathy for users? How to tell a compelling story about their product?

Were past YC alumni who were more hacker-ish molded into entrepreneurs through YC, or did they already come in with such a mindset?

Coming to terms with the effort required.

Is it possible to pitch to the YC team any time during the year, not just during the application seasons?

What's the reasoning behind having the twice a year rounds, as opposed to open for investing at any time?

EDIT to clarify: I understand why they have startups move to California for three months. I've simply never heard why that has to be done seasonally.

It's for economy of scale for presentations, demo day, etc. Also to make it easier for YC to keep track of what stage people are at - i.e. all 3 weeks in, etc. Having people start any time creates more administrative work.

You've had quite a few guests on Mixergy that started their own businesses solo, without partners. PG has repeated that most of the companies he's looked at have at least 2 founders.

Has he changed his views on this any or is this still the main advice?

Either way, what would he look for in a single-founder startup (besides being Relentlessly Resourceful).

Not much different than we'd look for in founders generally.

From the interview: Jessica Livingston is working on a second edition of Founders At Work.

can i buy just a diff between the two editions?

it's a full diff, I guess?

It's not a new edition; it's a second volume.

I stand corrected. This makes the news even more exciting!

What would be very interesting would be a book based on one YC group, from profiling all the companies and people, to following the experience and how ideas and people change over the time.

The video is edited and posted along with an MP3 download & transcript.

http://mixergy.com/y-combinator-paul-graham/

Thanks Andrew. I didn't get to see it live, but that is my favorite interview you've done (and that is in a group of quite interesting ones). Also thanks to pg.

I just hope it's a lengthy interview. I don't think I've ever seen an in-depth interview with PG.

I'm aiming for about an hour. I haven't seen lengthy interviews with pg either, which made my research harder.

Here is a one-hour interview (audio) Paul Graham did with EconTalk on Startups: http://www.econtalk.org/archives/2009/08/graham_on_start.htm...

"...I don't think I've ever seen an in-depth interview with PG. ..."

I've seen one. The interview was by David Weinberger and you can see it here (16:52) ~ http://media-cyber.law.harvard.edu/VideoBerkman/paul_graham_... Interesting & talks about YC, Taste for makers.

Andrew Warner: "how does a talk start a revolution?" (re: pg's talk at harvard that led to YC)

pg: "hmmmmmm (laughs)"

That would have been an awesome talk to have attended...

Yeah, it was a (I believe related) talk of his at MIT that really got me thinking seriously about rejecting the big cubicle farms and doing a startup... before that, it was just an unrealistic-seeming dream.

I'm still amazed by the effect he had with one speech, but it was just what I needed to hear.

I know YC recruits founders from all over. However as someone in Asia seeing more and more great Chinese web services, I'd love to know if any YC companies primarily serve a non-English speaking market.

I think iJigg does.

I'd like to know who the first company that YC wanted to invest in was - whether or not they were able to and if it worked out (in case it's a company we don't know the name of).

I'd also like to know if there are parallels in running an venture firm to running a startup. Does it take the same kind of guts? Do you get discouraged from a lot of failures? etc

There were 8 simultaneously in the first cycle. Some definitely worked out: Reddit (which was a merger of 2 from that batch), Loopt, Textpayme (the first YC funded startup to be acquired), and Kiko (which was not itself a big success, but afterward the founders started Justin.TV).

My question: Why does Paul Graham think you must be in silicon valley when doing an Internet startup? Living expenses there are expensive, broadband is available everywhere, and many people are outside the valley and successful.

Community, getting away from "home" helps with intense concentration, more access to mentorship, investors, partners, etc.

http://www.paulgraham.com/revolution.html

To rewatch, or see for the first time, here is the link:

http://www.justin.tv//mixergy#from=11.00%252CFebruary-9-2010...

That link doesn't seem to work.

When I paste the following in Chrome it works:

http://www.justin.tv//mixergy#from-11.04,February-9-2010

The link changes after the pages loads

Link doesn't work. "Sorry, we couldn't understand the date and time '11.00%2CFebruary-9-2010'."

Click the link under the video pane to get to the archives... archive for Feb 9 is 10 minutes of some guy with super-short hair looking into the camera and drinking from a cup... with wacky indian-ish background music. The picture almost completely obscures some kind of moving text... with an occasional glimpse of the phrase "Paul Graham" ... WTF?

Searching for "mixergy" in the justintv search box comes back with no results and a suggestion to watch a video about Bioshock 2

"archive for Feb 9 is 10 minutes of some guy with super-short hair looking into the camera and drinking from a cup... with wacky indian-ish background music"

Yeah, that's exactly what was on the channel post-interview. "Some guy" is actually Andrew Warner, though.

Here is a link that links directly to the video:

http://www.justin.tv/clip/256d3b4ba66875ae

The interview starts at 8:40

My question: Consulting treadmill - how do I get off it without a killer app idea?

Apply to YC.

I never understood why people consider consulting to be a "treadmill". For me, a consulting gig is an instantaneous injection of capital and another 6-9 months runway for the entrepreneurial stuff.

I mean, a 1-month gig will pay your rent for the year. Another one will pay for your servers, hosting, dev machine and food. The rest of the time is yours to spend as you like.

Is it possible that people are doing "consulting" work for <$50/hr and mistaking it for the real thing? That, in my mind, is not consulting but rather "being taken advantage of." Especially if you do it in an office surrounded by other employees in a similar position.

Consulting is a pie-eating contest where the reward is more pie. If you're good, you get just enough work to keep you from doing something else. I know I'm not alone here.

Reduce your personal expenses and you can acquire a surplus. If you let your personal expenses expand to meet your cash flow, you'll never have excess cash flow to do with as you please.

Indeed, that's essentially another way to rewrite what I said above, though perhaps a bit of a pessimistic one.

If you're just squeaking by, getting enough consulting work to pay the bills, then your schedule is likely pretty open most of the time. Unless you're spending 28 hours a week chasing work, you should have plenty of time for entrepreneurial stuff.

And of course, if you're working 20+ hours a week on billable consulting work and just squeaking by, you're simply not charging enough.

> getting enough consulting work to pay the bills

"Consulting" does not just pay the bills. You're thinking of freelancing for dipshit entrepreneurs who think $10k will buy them the next facebook.

$10k might very well buy the next facebook, in the startup stages. It wasn't always this big. All it takes is the right idea and some money to pay the bills.

> $10k might very well buy the next facebook

Get the fuck out of here. The amount of engineering the founders put into facebook far outpaces $10k. You're living in a fantasy land.

My point is that those founders didn't draw a salary until well after they had a value proposition for investors.

Or to put it another way, Stephen and I have put > $400,000 of hours into Gridspy, but our costs are limited to board prototyping, outsourced layout and web hosting. For most internet startups, costs are perhaps $10,000 per founder for 3 months + hosting (perhaps $100/mo until there is enough traffic to drive ads).

It only gets expensive when you need to bring in contracting help. That is why it pays to be a switched on hacker in this game.

Remember, the giants weren't always giants. As they transitioned to giants, there were doing it with VC money - not founder's cash.

Don't look at the destination and get put off. Take it one step at a time. You can probably get to traction with $10k.

One further point. I wouldn't be surprised to learn that facebook is burning through $1m per day now.

http://www.google.co.nz/#hl=en&q=facebook+burn+rate would seem to agree with me ($200m per year perhaps).

> Is it possible that people are doing "consulting" work for <$50/hr

Ding Ding Ding. Real consulting starts at $100/hr and has no cap on effective rate. One month of this should pay one person's bare-bones living expenses for at least one year.

"Freelancing" is the term most people think of when talking about bullshit web programming jobs that pay very little. They only take these jobs because they've failed to save up enough money to say "no" and they lack the experience to know that real money can be made consulting.

WherE do you get the consulting gigs typically? Almost all leads I get(good and bad, pay wise) come in form of a project with fixed cost attached.

If the budget is bad, push for reduction of scope. If they won't budge, run.

Glad Y-combinator think about graphic designer. it reminds my post a few years ago http://news.ycombinator.com/item?id=50375

This was a great interview; I can't wait until it's available online to watch again. I had to take a few phone calls during it so I missed a few things.

From a Y Combinator's "customer's" standpoint, (the new companies applying) with all the iterations and increased class sizes, what's to keep the magic sauce from disappearing? Can larger class sizes get the same attention from you guys as founders did during the first session?

We can measure that. Founders book office hours online. We know the founders are getting enough attention (by their standards) if not all office hour slots get taken.

Just out of curiosity, how many hours do you guys keep open each week for office hours? Or is that something you'd rather not say?

Depends on the time of year. People want to talk a lot more the week before Demo Day than in April. On average I'd guess 10-15 hours a week. But I also talk to startups outside office hours; e.g. at the dinners, which span about 6 hours each.

Is it all at the table or do you have some kind of presentation too from past founders?

There are all kind of presentations and special events. I'm just talking here about individual conversations. (I'm not sure what you mean by "all at the table.")

How do you gauge how well founders gel/get along/chemistry/rapport?

Is there hope for non-technical founders?

How do you redefine failure?

RE non-technical founders: the first part of building a company is building a successful product/service, and technical founders can do that with sweat equity. Non-technical founders can still add value, but they need technical people to make the product. Spencer Fry of CarbonMade has written some great stuff about duties for non-technical founders: http://spencerfry.com/whats-a-non-programmer-to-do

can a mod edit out the title since the interview is over?

I was able to get (at least the beginning) to play at the archive: http://www.justin.tv/mixergy/archive

Reply on news.ycombinator.com