Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Super last minute advice for startups applying for Y Combinator (garry.posterous.com)
74 points by rantfoil on March 17, 2009 | hide | past | favorite | 32 comments


That's a thoughtful post.

But why do I literally need to get out onto a life raft in the ocean with my co-founders? Are there customers out there?


It's a metaphor for the fact that all you have is your cofounders, so choose wisely.


I assume he's playing off of the usage of 'literally' regarding a metaphor.


This is a very interesting article about how "literally" is used to mean "figuratively." http://www.slate.com/id/2129105/


Perhaps I should stop calling my cofounder my 'wife', and call him my 'raft-mate'?


I don't think it is so much a matter of all you have is your cofounders, but more that if this goes down you all go down together. That is why you must use each other to get to shore. But still use all other avaiable resources as needed.


Great post Gary! My favorite is #5 "A little help from friends." It took me a long time to realize the value of letting everyone know what you're doing and what you need. I use to try to keep everything secret until launch. Now I realize that if you tell people what you're doing and what you, they're much more likely to HELP you.


Even if they can't help you, they get excited about you and tell other people who can help you. I'm amazed at the amount of support, help, advice and ideas I've been getting from people even in entirely unrelated fields. Of course, dealing with it all and not following every last piece of fishwives' wisdom is another story... ;)


I should probably listen to this advice myself. And thank Gary and other people who're trying to open my eyes. All in all great advices for someone like me who's only just getting started.


There's always something awesome about reading a guy's blog that's hosted on the engine he himself built.

Excellent post, as always.


There's always something awesome about reading a guy's blog that's hosted on the engine he himself built.

There is? He didn't write nginx, or the C libraries it uses, or the linux kernel it's running on, or the Xen virtualization layer slicehost is using, or the C compiler which was used to compile all of the above -- but each of those is far more sophisticated than the blog engine. What's so awesome about a very thin layer at the top of a very deep stack?


One memorable Thursday I got to eat dinner with a writer and a teacher of mine. He cooked an excellent meal. Part of me was thrilled that he cooked for me and my peers.

He didn't plant the seeds that grew into the vegetables he used. He didn't grind out the flour he used to cook. He didn't raise the cattle, or clean the meat. He just took some finished products and he made a meal.

Every step of any process requires an awesome effort. I'm just as impressed with the people who wrote those many, many laters. But in the end, design is that layer you're talking about, and Posterous is very well-designed, and when I use a web site that's all that I care about. That's the product of the Posterous team, and it's a pretty awesome one.

Why bother doing anything if all you're ever doing is building a thin layer? Could it be that any product is a result of lots of people working on something, and that each contribution is integral? Nah, couldn't be. We're all insignificant and nothing is awesome.

What happened to the ability of hackers to look at something cool and say "Gee golly wow"? I still do that all the time, and every time I do I get these little snide remarks. Why? What harm is there in thinking that lots of things are awesome? Lots of things are awesome! Life is so awesomely super-duper cool! Let's not ruin it by sneering.


Allow me to clarify somewhat. This isn't the first time I've seen someone make a comment like yours; but I have never seen anyone say "wow, it's really cool that a FreeBSD developer is running FreeBSD on his server" (substitute "nginx", "Apache", "linux", "Xen", etc. for "FreeBSD" as required). It's more the opposite -- it seems that those of us who work on the lower layers only ever get attention if we don't eat our own dogfood.

To turn my question around: Ok, so the blog engine and all the underlying layers are all awesome; but why don't all the underlying layers get as much attention?


I think it's cool that Linus Torvald's is using Linux to teach his daughters computing (with handy parental controls he wrote himself).

I think it's cool that Guido van Rossum uses Python in most of his daily programming.

I think it's cool that thet Lighttpd website runs on Lighttpd.

I think it's cool that I use my own project for my daily websearching needs (okay, bad example, since I'm in Search UI and most of my job consists of putting a pretty web interface on the stuff the backend engineers do. I think it's cool that I use lacker's stuff that I slapped a frontend onto for my daily websearching needs. ;-))


Where you use "cool" I'd use "expectable"...


That was cperciva's point, I guess. I still think it's cool though.


Okay, that makes more sense.

I think it's because a lot of that stuff doesn't get as much publicity. I'm fairly clueless about Unix: I know that Linus guy did something, but I am not well-versed in Unix's history.

That might actually be a fascinating thing: a publication or web site or blog devoted to exploring stuff like that, explaining why they're so huge and so important, and getting information like that out there. I feel bad* that I don't know more about my computer and about coding systems beyond that top layer, but at the same time I don't know where I'd start to learn about things like that. (The closest I got was with _why's explanation of why Ruby is neat, and even that was enough to inspire a real fascination with the language. I'm sure that sort of attitude could be more commonplace.)

*Specifically, something to do with Linux.


Its all about the percieved value. Its amazing how hackers (me included) don't get this.

Normal people rarely care about the technology, they do care about little things that improve their live


While good design is important, I'd think twice about having a designer/cofounder. Look at this site for instance; design easily done by a developer, yet the service itself fills a niche and built a community.

As a side note, I've seen early design hires hold back the company in later stages as they tend to wield too much power over the product (often playing a pseudo product management role).

It's easy enough to outsource a logo and some paint. Think about your use cases and make sure the service caters to them but good product is often "designed" at the functional level.


A good designer is a good designer. If someone holds back the company later, that's not good design.

Design is way way more than logo and paint. Interaction design is just plain being able to make stuff that works well for people -- that's the kind of design I'm talking about. Should have made that more clear.


It looks like you are the "designer" on your team. It also looks like you have an amazing background in software development, program management and entrepreneurship. That's the kind of designer you want as a cofounder ;)

As you say, design is a lot more than some paint and a logo. Interaction design, information architecture... are all VERY important at the early stages. I do think the early devs should have this as a skillset though (much as you do) opposed to cofounding with a straight up interaction designer (no matter how good they are).

Given the amount of equity a cofounder gets it I would tend to look for people who could be 100% utilized (working all the time to get the company up and running). It can be hard to keep a full time designer busy unless you have a team of devs pumping stuff out for them to craft.


Thanks to garry for posting this, always nice to hear a bit of advice distilled down.

Even nicer when you're able to check off most of the points :D


This useful article comes from Garry Tan, cofounder @ Posterous.


Paul, can we have a feature that when we save an application there's also a 'feedback' button, which sends the application to select YC alumni and others who have time to give feedback?


Nobody is going to hold your hand when you are starting up.

It would be unwise to have a feature that gives you that false impression.


What are you talking about? The whole YC program is explicitly about getting your hand held. Anyone can get a credit card or family/friend loan for 5,000 for a three month project. YC is not about the money; it's about the advice. YC is about nurturing seeds to see what comes up, rather than testing them for the Winter. There's a reason seeds sprout in Spring -- to give them as good as possible running start before Winter comes. They'll get tested soon enough.


No. I like how uncluttered the application is. It made me feel like we were applying for something professional, unlike some of the other applications I've seen, which either look decades-old or really sloppy and amateur.


Just save the webpage as html, and send the html file. (I find safari does the best job here)


Is posting your application here for feedback in the form of an ASK YC post frowned upon?

Surprised I haven't seen anyone do that.


In my opinion that's a little bit tacky. I don't know if I'd like people seeing exactly how I answered every application question a week before applications were due, and I don't think that an application post would be HN-worthy. (Especially not if a bunch of different people all posted theirs at once.)

I'd advise emailing a few people that you think could give you feedback. I sent Garry a rough draft of my application, and his feedback helped a lot. Either write to a few HN people (and don't get upset if some people you write to are really busy right now), or join the IRC channel and ask people there.


Didnt know there was an IRC channel! Does anyone have the channel and the server?


freenode #startups




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: