HN Theater @HNTheaterMonth

Damien Katz - CouchDB and Me: https://www.infoq.com/presentations/katz-couchdb-and-me

The talk is about how Damien quit his job to hack on open source software. It shows his struggle and doubt while embarking on the project and then finally invented CouchDB. It's a passionate and human account of the process of creating something significant. I recommend every hacker watch this.

Getting someone to pay you to work on what you want and how you want is likely to be rarer than a unicorn.

Starting your own business is a much easier path to that and even then you will be pulled in all kinds of directions.

The only case of a unicorn job that did actually spring to mind was the CouchDB founder. This is a very motivational talk by him - http://www.infoq.com/presentations/katz-couchdb-and-me

⬐ matznerd

You can have a unicorn job if you just do consulting and/or project development. While I personally prefer entrepreneurship, I understand that it is not for everyone (as the author mentions). If you do consulting though instead, no one owns your time aside from what you decide to sell, and you can choose which projects you work on.

⬐ None

None

⬐ marquis

Starting a business begins as a unicorn job, and if you're lucky you find out that you also enjoy running a business. And get to keep your unicorn part-time job. Also, having other people help build your unicorn. I LOVE what I do every day - used to complain I wasn't coding enough, but it was because I was controlling the output too much and not delegating. It wasn't that I wanted to code more, it was just that I wasn't getting enough done. So I hired more staff and found I really like the balance of entrepreneurial life, helping my staff code their best (and towards our goals) and coding in the quiet hours.

⬐ graeme

Completely agree here. Anything specific that helped you delegate more? I'm getting started on that, but still feel I'm not getting nearly as much done as I could, and I'm a bottleneck for my outsourcers.

⬐ marquis

Delegation means knowing exactly what results you want, and if your staff aren't skilled enough being able to lay it out down to the line with direct tasks with clear finish lines for each task. We do a lot of training, so when working with junior developers you want to break it down, break it down, and be able to think laterally in case they can't achieve exactly what you want - there are always other ways to get things done. Then trust them, and make sure you get daily reports. At first it feels like you are not getting anything done and then they start flying and it feels great.

⬐ xenonite

May I ask how many people work for you?

The kind of delegation you use sounds a lot like micro managing, which scales to teams of 20 at most.

⬐ marquis

We only use this method for new and junior developers, for maybe 2-3 months depending on how quick they are to get up to speed and we trust that every day is spent on something constructive.

⬐ graeme

That's a great way of putting it, and I have been working on precise templates for certain tasks, which are now bearing fruit.

Why daily reports, as opposed to any other interval?

⬐ marquis

I think all knowledge workers should note what they did at the end of the day - it helps with accountability and gives a sense of achievement. For example a tool like idonethis.com helps you and everyone stay up-to-date and see quickly where problems are. Or ask everyone to just email you at the end of the day or a fixed period. Always read them and respond if needed.

You note that you feel you are slowing your team down - if it's because you are relied on to be building core tasks, try building mock-ups first so they can work. Another good tool is apiary.io if you work with APIs for example: it combines fleshing out your API with actually providing a functional framework so the front-end guys can get to work.

If it's documentation slowing you down, give everyone some days off or get them on another task and catch up. Use paper and pen or whiteboarding or a design tool like Flairbuilder if you need to really understand your ideas before trying to explain them to someone else.

Congrats to Damien Katz, J Chris, and the other founders.

Damien's all-in story to build CouchDB is inspirational:

"This is a talk I gave last year at RubyFringe, about the whole process of me and my wife selling our house and living off savings to create CouchDB."

http://www.infoq.com/presentations/katz-couchdb-and-me

http://news.ycombinator.com/item?id=465653

Side note. I think this speech by Damien Katz is one of the most inspiring I've ever heard.

http://www.infoq.com/presentations/katz-couchdb-and-me

⬐ mahmud

I am half way through it and couldn't wait to confirm this.

Damian Katz is a fine human being, and an artist. Never knew much about him before, but as of now, I consider him a role model.

I found Damien Katz's talk about quitting and risking it on CouchDB to be really great (from resources): http://www.infoq.com/presentations/katz-couchdb-and-me

Demien Katz has a similar story, you can watch the whole presentation here: http://www.infoq.com/presentations/katz-couchdb-and-me it's really inspiring.

⬐ trevorturk

This is a great video - 100% worth watching.

⬐ drawkbox

It is great fun to use CouchDB and knowing more about the creator is great and makes it better. That was a great one. The honesty in the story was amazing and it was one of the most inspiring things I have seen. Damien Katz you are victorious.

Its not too late, you just can't afford to be in Ramen mode unless you save money ahead of time or your wife has enough income to float you on a startup salary.

My advice: get in a startup as early as you can afford to. If that means being employee 10, so be it. You'll still have a good time. If you can afford to be a founder, do it. But with a wife and kids, joining a funded entity is probably more palatable.

As a more mature adult... you're actually really valuable to a startup: you bring wisdom and perspective. And you're still willing, you haven't 'grown out of it.' Use this to your advantage.

I'd suggest this talk from Damien Katz of CouchDB, about taking the leap to work on a project full time, with a wife and kids - http://www.infoq.com/presentations/katz-couchdb-and-me

⬐ MartinCron

This advice is similar to what I've done, and it has worked well for me. The experience you get in a small growing company is (to me, anyway) more valuable than what you get at a larger, more established company.

Advice addendum: Be patient. There aren't many organizations that will work well for what you need, small and fresh enough to remain interesting yet stable enough to support a real family. The good news is that you only need one.

⬐ dantheman

Agreed, here's a previous submission for past comments: http://news.ycombinator.com/item?id=465653

⬐ ricree

He recently was interviewed on the StackOverflow podcast ( http://blog.stackoverflow.com/2009/06/podcast-59/ ). A bit of repeated material, but overall it was a fairly interesting program.

⬐ nir

Awesome. I really hope this gets to front page.

⬐ wallflower

In his personal, excellent talk (as the 1st commenter on Damien Katz's blog post wisely notes), Damien Katz shares and talks about quitting to do a startup when you are not young and single but older and responsible but still have that startup urge on simmer/low.

Laura (his wife) and Damien's share their differing responses/perspectives on what the personal/family risks were in the comments on his blog post below:

http://damienkatz.net/2009/02/couchdb_and_me.html#comments

In the beginning...

"Sell our house, live off the savings and build cool stuff... There were a lot of reasons not just money...

The first was just educational - I had interests in a lot of areas. The first was, distributed systems. I wanted to know more about them. I wanted to know how to build them... It seems crazy but if you're going to go off and pursue a Master's degree - you're going to quit your job, you're going to have to pay your tuition - and there's no income - and the end of it you have a degree. An Education...

It was about more family time. I had a beautiful one-year old and I didn't want to spend all my time at the office, working on somebody else's crap...

I wanted to see what I could do...

And this is the reason why my wife was interested. It would be an interesting story...

No wants to be wrapped up in this consumerist lifestyle, where if you have money you spend it...It was a good thing it forced us to downsize...

I wish I had sort of a vision... It wasn't like that - it was just day after day and letting it bubble into my consciousness...

The development process...It was really stressful once I decided I wanted to build it because I didn't know how to build it...I couldn't see past all the complexity...It just took a long time grinding at the problem, thinking about it...

I really just went into a panic mode. I was thinking I was a total fraud, I don't know what I'm building, I've got my family wrapped up into this - I don't know how to build it - so I got on Amazon and ordered Code Complete"

What a great essay! It brings to light the day to day emotional struggles of focusing 110% on something, especially on a solitary level. Check out this video of Couchdb founder talking about a similar struggle

http://www.infoq.com/presentations/katz-couchdb-and-me

⬐ zupatol

I found no transcript so I'll try a short summary, although I'm already forgetting the beginning.

Damien Katz worked on lotus notes, later he had a startup, from which he was fired. He started looking jobs but regular jobs didn't seem to interest him anymore. He believed he was a competent programmer, but felt that he could contribute more. He decided to live on his savings and build something before knowing what he was going to build. His wife liked the idea of living differently. He soon dropped his first idea which turned out to be horribly complex to realize. He then tried to somehow extract the essence of what was useful in lotus notes, leaving aside the unnecessary complexity, switched from c++ to erlang along the way, and ended up with couchdb. He was running out of money, so he took up a job at mysql, (although he had never used mysql, and barely knew sql). The development stopped until he took a break for the birth of his second (or third?) child, and brought couchdb to a point where suddenly ibm was interested in it. Negociating with ibm was so frustrating that he ended up calling them douchebags in a mail. This led ibm to agree that couchdb should be open source and to pay Damien Katz for working on it.

He says many more interesting and funny things.

I listened to it only once, without taking notes. Please correct me if I'm wrong. (I posted this on reddit also)

⬐ dantheman

This is a great presentation on how to build cool stuff, and the sacrifices he made to do it.

⬐ fortes

A good reminder that success doesn't always come overnight, or without sacrifice.

⬐ inovica

Good presentation. To be honest I didn't know of CouchDB. Looks quite cool. Anyone using it on here and what are your experiences of it?

⬐ dantheman

I'm using it and like it a lot. I'd say it's still relatively new and there isn't mountains of documentation or examples so sometimes it takes a little banging your head against the wall. I think it's still at the alpha stage, though I think 0.9 will be considered beta.

For more information on couchDB: http://couchdb.apache.org/

⬐ ahpeeyem

I didn't expect to find inspiration to follow your dreams and do something that really matters to you in this video, but it's a great insight into the self-doubt Katz experienced while he took his family off to another state and lived off savings to create something awesome.

Really interesting presentation and I love the hand-drawn slides!

⬐ dantheman

Agreed, you could really see the emotion in face/voice when he was describing what he went through.