It was really an eye-opening experience to go through a big list of these videos. Thank you to everyone who has participated. It definitely gives me a new perspective. I once thought it inconsiderate that YC didn't even click the demo link in my application, but now it's obvious that it's a necessary evil in sifting through so much of this.

One thing that stood out to me was how late all the videos were in the application process. About 75% of the videos were submitted within 48 hours of the deadline. It seems that submitting just a week earlier than the deadline would gain a huge advantage in terms of how much time the YC folks are able to commit to your application.

If I had to sift through this many videos in a short amount of time, I wouldn't even finish most of them. Whereas, if I received a video for the first time in a few days, I'd probably watch the whole thing regardless, and even try out the demo.

Did you submit a video with your application? If so, where is it? :-)

I didn't apply this time around, for several reasons which aren't really important. Otherwise, I'd have definitely included it... you know, to show others how it's done.

This began in the node.js chatroom last night. But it just proves my own ycombinator application was a good idea (WikiTorrents): a website where you can list downloadable videos that you can get on other websites.

So why haven't you started building it?

Umm... the link IS WikiTorrents.org . I've been working on this project for a year. Check it out: http://wikitorrents.org/wiki/ycombinator_applicants

Oh, I didn't realize that. Good job!

Just a heads up, I'm getting a 403.

Hmm... we have a somewhat zealous firewall tool on our servers, but I modified it.

Try again and let me know. Thanks!

I'm in no way intending to be rude, but the standard set in those videos isn't at all as high as I would have expected.

Even without post-production editing, surely there should at least be adequate lighting conditions and camera alignment.

I think clearing breaking the pitch into parts would be a good idea also. Such as an intro, about the founders, about the problem, about revenue, etc.

I would make specific comments per video but that wouldn't be a nice thing to do.

Just my 2c guys, good luck to you all!

The worst video I was with really choppy webcam video and it also had a several second delay between the audio and the video. I mean, if you can't even operate the webcam, why bother labeling yourself a "hacker" and applying for start-up funding?

Second worst I saw was some stoned guy going on about how much he hated seeing "all the genitalia" on chatroulette. Seriously. This was a real video.

The best two I personally saw were a video of several founders for a question and answer site and a new-age guy with an idea for personal metaphysical space. Both videos were very engaging given their short length.

Hahahaha.

Stoned guy here. You just need to chill out and recognize the problem of the penis... man. It's giving the internet bad karma.

Seriously, I was a little tired, but I was not stoned. Could I have made an awesome polished video? Sure. But they said talk spontaneously as I would to a friend, and I took their instructions literally.

Any feedback on the concept?

oh man, you kick ass! thanks for being a good sport about it.

For those just joining us, here's the video: http://livejake.posterous.com/application-video

My recommendation would have been to seem more enthusiastic and professional. The idea seemed kind of, umm, unique. But if you had presented it more professionally and with serious facts and data, I think it would have made a better impression Also, maybe some ideas about who you will market the idea and how it could make money. You could've offered specifics about maybe marketing it to parents somehow. Or connecting your idea to some sort of succesful product that involves internet safety.

Yeah, you're right about being enthusiastic and professional. Thanks for the feedback.

As for stats and data, I tend to buy Chris Dixon's argument that narratives are more important than numbers: http://cdixon.org/2010/04/03/size-markets-using-narratives-n... Nevertheless, some statistics are useful.

I included additional details about potential users and what not in my application.

An example use-case, which I didn't get into in the video, is an entrepreneurial designers and developers event.

Here's how it works: 1) Designers and developers RSVP in advance for an online event that takes place at a set date and time.

2) Upon joining the event, each designer is presented with a list of developers in the event with basic bio info from their LinkedIn profile. Each designer can then check off who they want to chat with in the event. The same happens for developers.

3) Once the event starts, they are placed in a series of one-on-one chat with a designer (or developer) that they've previously indicated that they're interested in meeting. During the chats, each participant can view the other participants LinkedIn profile. These chats last for a set time, say three minutes, like speed dating. In the event, participants meet a lot of people that they're interested in meeting in short one-on-one chats. That's the general idea.

There are a lot of flaws with this concept. And I've actually started working on a different idea shortly after I submitted my Ycombinator app, but I still think this idea has some limited potential.

Do you have a working site or a demo or something to show off? If so, let me know.

What I submitted was this: rapidintro.com:8000 You need two people to really demo it. Oauth with LinkedIn and then you video chat. The video chat is still buggy. We didn't finish the event system. This is the first ChatRoulette configuration written in Erlang.

The reason the design looks like it was designed by Slovakian college student for $50 is, well, because it was designed by a Slovakian college student for $50.

As I said before, we are now working on a different application and we should be completing that shortly. I'm considering releasing this application, with some design tweaks, as is to Hacker News and other entrepreneurial communities and seeing what happens.

I'm surprised at how poorly planned and filmed most of these videos are.

Is that the point? Is it supposed to be so rough? (Note: I've never applied to YC or looked at the application)

Here's the official instructions for making a video: http://ycombinator.com/video.html

The instructions say: The video should contain nothing except the founders talking. No screenshots or postproduction wizardry please; we don't want this to turn into a video making contest. If you're going to spend time making something cool, put it into your demo/product.

However, I agree with you. I think people should strive to make a good first impression.

Beyond the bad quality of some of the videos, I was surprised how many people submitted videos saying words to the effect of "I have another co-founder but he couldn't be here for this video." If I were watching those and heard that, I'd automatically reject the applicant. If you can't even get your alleged co-founder together for a simple one minute video then things are obviously already falling apart.

"If you can't even get your alleged co-founder together for a simple one minute video then things are obviously already falling apart."

I've actually heard of people applying a few hours before the deadline. So if they are geographically distant from one another, they just use webcams.

I was on my honeymoon in Hawaii and we didn't put together the application until after I got married, so we used two webcams and spliced 'em together...

My comment wasn't about videos of co-founders spliced together.

My comment was about people that just basically said, in so many words, "I have a co-founder but he refused to show up even for a one-minute video."

I know there isn't any context without the application also, but there seems to be a lot of vague reasons why a new offering will be better than the competition.

Very cool! Nice idea - customizable lists, context but not content.

{ ouch - hard to watch myself }

The wiki has its benefits, the best being that I can revert to previous versions if somebody messes up a page edit or does graffiti on a page.

The wiki is not very granular, though. Nor is it easy to edit. The current setup is going an overhaul and being replaced with a "much better" version that I'm still testing. I hope to have it implemented publicly within a month.

Interesting that most people haven't put a nice shirt/ suit jacket on. I know it probably has little impact but it gives a good initial impression.