notelab

collaborative notes taken during events

startupschool2006

Welcome to the canonical post-lunch shared notes!

ANNOUNCEMENTS

Please put all talks into this one document sequentially. Conversation is at the top, speaker notes go down below the conversation.

This document is now posted at the following URL (with periodic updates): http://notelab.infogami.com/startupschool2006

CONVERSATION

If you're interested in doing some Markdown-type markup on this, that might be cool. Markdown syntax reference

Or, see the Markdown cheatsheet

What is the "Marimba" company the PG just mentioned?

I think Marimba is a Java based technology circa 2000. It was either a Jini thing or a zero effort client install thing. Actually, I think Marimba morphed itself into a company that sold its SDK to TiVo to allow end-users to build TiVo apps. Battery is getting low. I'm going to sign-off for a bit. TTYL.

Sorry, I'm selecting all the text periodically to preview the markdown->html conversion. i'll try to remember to unselect.

Anyone have a sidekick charger?

SPEAKER NOTES

Speaker: Joe Kraus: "Startup Addict"

Summary: Joe (joe@jot.com) started Excite, got incredibly rich, and has 8 key lessons to share with prospective people

  • Persistence is key: "Negotiations begin at No"

    • Lost bit to become the netscape search button to MCI
    • Acted as if the bidding process wasn't done, stayed 21 days in Netscape's lobby until MCI reneged and they got the opportunity.

  • Employees are your #1 asset (not just lip service on this one)

    • Reject the qualified to get no bozos (Google does this, perhaps to their detriment at present)
    • DO NOT HIRE QUICKLY. You think you need to, you think you're behind, DO NOT DO IT.
    • A players hire A players, B hire C, C hire bozos. Graham talks about this in H&P

  • You make what you measure

    • Put it in your staff's faces. You won't actually even need to change things, just the awareness can double your intake (Good example is Cisco's executive compensation work)

  • Be a trend-spotter versus a trend-setter (Cisco as example, Juniper's doing optical, but Cisco runs laps selling IP)

    • No startup has enough $ to create a category the press will give a rat's ass about
    • Look at the recent Newsweek, Caterina Fake, Youtube Guys, etc. on the cover, doing technology that's been around since '95.

  • Opportunities creat opportunities NON LINEARLY. Synchronicity will be your guide. Joe tells a story about how he called Cringley based on a book he read, Cringely -> Infoworld, Infoworld -> IDG, IDG->Steve Coit->Coit to big funding w/ Vinod Kholsa

    • Put your business model into betal along with your beta product: you're not beta-ing your product until you beta your model
    • No milestone ends the race. You get 1st round $ it just means you're competing against equally competent / competitive orgs that were funded

  • How is the startup env different today, after the bust?

    • Easier to start a company
    • Harder to start a business that survives!
    • With lower funding there are more opportunities for a successful exit.
      • Get 250K, sell 1/3 the stakes: Valuation for successful exit sale is 3.75Mil. Not bad, Bucky
      • Get 5M, sell 1/3 the stakes: Valuation for successful exit sale is 15M. Not many people around that can buy you, unless you're monstrous

Speaker: Page Mailliard (Wilson Sonsini)

Steps to set up a company: 1) "Confirm the legal entity and the structure" * Delaware Corp (C) (why not S?) - no S because you'll end up needing your investors (if you actually need investors) to go through some crap.... check Brad Feld's blog (http://feld.com) * Simple is better * Why not an anonymous Nevada Corp? ;-) * Taxation in Delaware is mo' nicer * May still pay taxes in other states, but Deleware may offer benefits as far as lawsuits are concerned.

2) "Decide what the authorized stock is"
    * 3 to 5 million shares among the founders

3) "Establish board of directors and officers"
    * Board and stockholders have the most power
    * Setting up a board: know someone before putting them on
    * Make a small board: 1 or 2 people at first
    * Great people: give them stock and put them on board of advisors

4) "Consider stock vesting"
    * Create incentive for founders to stay over time:  See "Every rock band that got big".
    * Why? When VCs come, they want to see that people are bound to the company
    * Easier to flip (but less upside of doing so)

5) "Adopt standard documents"
    * Option plans usually have a life of 10 years

6) "Consider trademark and patent strategy"

7) "Figure out who owns the technology"
    * If it's not the company, you're sunk
    * Friends don't let friends work at micro$oft.
      - or talk to them while they are working there

8) "Know who owns the company"
    * Don't give people things on napkins.
    * Don't give God anything.

Types of Equity
    * Common stock (founders, employees)
    * Preferred stock (Investors)
    * Reasons:
        more priority in liquidity
        dividends, voting rights, anti-dilution, etc
First Term Sheet:
    * Be clear on value proposition (market, rampup value, etc)
    * Keep presentation short (10 slides, 15 min)
    * Be credible (do homework on fund/people/market)
    * Bring "A" players
    * Listen to others
    * Listen to yourself
    * Your best leverage is another term sheet
    * Get the best business partner you can -- not necessarily the highest valuation -- and the best board representative.

Speaker: Mark Fletcher (Bloglines)

"Lessons Learned Birthing and Building Web Start-ups"

Started two successful companies, Onelist and Bloglines. In contrast to Onelist, bloglines was totally self-funded.

Put in 200K from start to finish. Had 1 salaried employee, 4 working for stock

Garage Philosophy: * Passion for the idea - Solve a problem that you have. - Had 100 feeds, wanted to save time, built Bloglines * Cheap technologies * Doesn't have to be perfect * Release Early/Release Often * Involve your users * If you have a problem you're hacking / solving for your self, there's probably a chance that others need that solution - Having a company becomes your life: you must accept this - If you're working on the problem already, you have passion for the topic, you're probably going to enjoy the work * You needn't get it 100% perfect - start a virtuous cycle that will move you to what really needs to be done. They are better than you at naming features. Respond quickly, respond often. * Moonlighting limits risk * Friends / Family funds - The longer you can go without venture capital, the more you'll be able to keep later * Free services = less pressure * Hire a laywer (See Will Shipley on this matter) * Outsource to eLance / Rent A coder - Did Bloglines notifier (windows app) for $300, internationalized to 6 languages for $3500 - Make your proposals as detailed as possible * PR is cheapest marketing you can do * Don't worry about biz dev * Focus on viral aspects * Focus on user growth

Software Choices: * Apache/MySQL/BerkeleyDB/Clearsilver/etc * improving speed/latency causes more pageviews * bloglines added more hardware, next day existing users increased pageviews by 30% * table level locking in MySQL doesn't scale

Hardware Choices: * dedicated servers vs buying/hosting * design for cheap hardware * ebay! * APC PDUs for remote power cycling - helps avoid 4am trips to datacenter to restart machine * HP ProCurve * Avoid Seagate Ultra-SCSI drives - 50% failure rate * Good phone for SSH - easier to fix remotely Architecture Choices: * copying files vs client/server - bloglines news rss feed * calculate on the fly vs cache - subscriber counts are cached * memory vs disk - notifications (how much we send to each user, so as not to overload anyone) Storage Choices: * relational DB vs flat files * RAID vs Redundant - Onelist/eGroups with hardware RAID - Bloglines without hardware RAID * Linux Software RAID 1 - bulletproof

Question: 
How do you partition work to be outsourced?
Answer:
outsource things that aren't core features, such as a desktop 
notification app that runs on windows

Speaker: Ann Winblad

TITLE: "How to write a fundable business plan" FROM: Hummer Winblad (Software-only VC)

I. Steady investment market for startup over last 3 years.

II> Software comprises a plurality of investment in SV

III. New era of software innovation - many uses, each provide a small incremental value

IV. Business Plans

* Idea
* Customers
* Secret Sauce
* Attact Excellence
* Numbers

V. Book: Good to great built to last, Jim Collins. This is used as a slide-by-slide basis in the following Phonecian-numbered points

  1. What is the business

    • Hook VC in 15 minutes (SUPER-important)
    • Describe solution not the implementation -- as a customer would describe it
    • Then describe the implemetation (technology)
    • (This is similar to what Steve Krug has to say in his excellent book: "Don't Make Me Think".)

  2. Do you understand the market?

    • Is there an unmet customer need today? Will that need grow?

  3. Is the need real?

    • Is it a vitamin or painkiller? A vitamin is a supplement, it doesn't kill pain.

    • (Do you have a product or a solution)

    • Can you define the market?

  4. Market Timing

    • Set the hurdles for the companies that will follow using PR.
    • This is absolutely crucial in the software segment

Use newspeak to define the market. ex(OLAP).

Ability to execute vs. Completeness of vision

Companies do not live in the products tab- need multiple tabs to define yourself

  1. Will Customers Beg

    • It's about aspirational design
      • (Ex: Newton vs. ipod)
      • Customers in CHINA SAVE FOR an iPod! Not go for a knock-off
      • jobs is a genius. Shipley sez: People tolerate Windows.

  2. What is your secret sauce?

    • Technology / IP
    • Business Process
    • Partners
    • Domain Knowledge

Engineering Roadmap * Can the product be built in less than a year. It will not be funded over a year.

Can you attract excellence?

Do not pack in your management team. You will not impress the VCs with a fully fleshed out structure.

Early stage VC- Two people will be funded. One won't. Ex. One company being funded has five people. Another has three.

Do you know what you don't know?

Assumptions are OK. If you know what they are, the VC respect that. See Michael Porter's "Competitive Advantage"

Be able to differentiate between assumptions and facts. Boil down to your key assumptions (top 10)

Do the numbers reach software economics? Sit down and build a spreadsheet at the beginning * YOU NEED: 80 gross margin, EBITDA 20 * Revenue - are services a major fraction? * Capitalization- Total < 15M? * COGS- Are COGS > 30%? * Mode - Enterprise v.s. Salesforce model. * Cash Flow - how long does the initial investment last? * You must be efficient with the capital

Learn how to distribute engineering across the globe.

How long will the initial investment last?
* Grind out the strategy.
* My dad sez: Avoid the "Ready, Fire, Hey how did that bullet get over there", Aim design.

Customers: * They believe in your product and team * They have a budget * There is a common problem * You know how to reach them

Customer feedback is part of investment process. * How do you reach your customer in a reasonable time while preserving the economics of your company? * Can you scale?

What VCs bring to the table

3P's People, Partners, Process

Money is a competitive advantage

Business Plan? (VC's frequently won't read it anyways- ultimately it's about interest in the pitch)

Why should I write a business plan? * How long should it be? * Who will read it? * Do I need an NDA? * Do I need an exec sumary? Max 5 or 6 pages- Will read it if the pitch is good, otherwise in the garbage

No VC will sign an NDA

You do need exec summary- No longer than 2 pages

10 Slides to cover the basics

* Executve Summary (Problem + Unfair advantage)
* Team
* Market Analysis (Size, Position, Competitors)
* Secret Sauce Overview / Background
* Key Assumptions- how is the market evolving?
* Customers (Potential / Referenceable)- Don't make up customers!  VC's will do due diligence!
* Product Plan
* Engineering Plan
* Sales Model and Marketing Outline

Business Model / Finanicals / Cash Flow- How long until the payment vests? Plan to be with invested companies for 6-7 years

Ann Winblad awinblad@humwin.com www.humwin.com She'll be happy to send the slides at request

Questions

A round funding- true early stage investors are interested in taking some risk B round funding wants to talk to real customers VCs are about- go big or go home- you must have the potential to be a large company "Could I describe this company, 6 years from now, in an S1 that I would want to put money in?" "Our job is to build companies" TOTAL GURUDOM: "Companies are bought, not sold- you want to be in a position where you can turn down acquisitions"

Speaker: George A. Willman

Patents for the Startup Startup School Stanford, California

Intellectual Property Issues: Summary

Avoiding Liability Creating Value (focus of talk)

Contributions to Company Intellectual Property

Categories of Intellectual Property * Trade Secrets- Includes technology / business (EX: What's the formula to Coca-Cola) * Trademarks * Think of a brand * Serve a communication function * Source of goods and services * Patents * Copyrights * Cover expression * Original works of authorship * Specifically don't cover ideas or methodologies * Books literature movies music certain aspects of software

Intellectual Property: Cost / Benefit * Protection vs Cost (More cost -> More protection) * Copyright- protection might be less because doesn't protect idea or independent development * Trade Secret * Tradement * Patent- Does not require a copy to be infringed- applies regardless of violating party's awareness

Patents- Basically they give you the right to sue * Government-Granted Right * For Inventions -- New and USeful * Exclusive Right -- Right to exclude, not Right to Practice

Basic Requirements to Obtain a Patent * Utility * Novelty * Non-obviousness * Description * Also: Patent Prosecution Process Obviousness is looked at the time the invention is made. Thus prior art can undermine someone's patent (see next slide).

Patents U.S. Statutory Bars * No US Patent if More than one year prior to us or pastt filing date * Patented * Printed Publication * Public Use * Offer for Sale Even if confidential or not actually sold

Other Countries Absolute Novelty

Patents how obtained

Applicant -> File Application -> Office Actions <- Prosecution / Amendment -> Allowance <- Issuance <- <- US Patent Office Avoid abandonment- missing a deadline / fee (could be a 2 yr process)

Patent Strategy: What to Patent

Why patent when things change so quickly- make sure a patent is right for your product / software Core Technology may not change There may be a marketing advantage Applications in other markets You can patent something that is not your product Maintain an ongoing patent portfolio as you continually improve your product / software

Parents may apply to compatible products / parts There may be future opportunities (can patent future ideas)

Ex: Cupholder patents- Dispite existing Starbuck design, Peet's Coffee developed a similar but slightly different design to get around the existing patent

Business Method Patents

There was a time when software patents were controversial in the legal world they are no longer - you can patent software (many filed patents are ludicrous- Microsoft patented "internet messaging") Ex.

Secure method and system for communicating a list of credit card numbers over a non-secure network (amazon.com)

Existence of "Companies" that are essentially IP aggregators with an army of copyright lawyers

Patent Strategry what to patent

Patent Clustering Patent a number of sdioffernt aspects of the product ex. Razor Blade Alternative Handle Manufacturing Packaging - strategy to block competition from moving into space

Patent bracketing * You can have the right to exclude, but if someone else has patented all the other technologies necessarily used to practice your item, you're hosed. Example is someone patenting a phone, but if someone has patented dial tone, keypad, etc. This essentially forces you to license.

Questions

How can you avoid what happened between Blackberry and RIM? ... stay lucky... settle earlier...

What are the economics and feasibilities of patenting? When is it worth it (startups are poor)? "I recommend patenting your core assets"

Pretty difficult to get someone to license your patent, even if friendly (response is "how do we get out of this?")

Speaker: Tim O'Reilly

Desktop Application Stack Proprietary Software (controlled by API)

Meanwhile free and open source software is being developed cheap commodity PCs

Internet Application stack Proprietary software as a service (Ebay, Google, Amazon, Mapquest) Integration of Commodity Components (Apache, Linux) Subsystem-Level Lock In (might be data suppliers)

What do the killer internet apps have in common with Linux? (The way you succeed is leveraging network effects- users add value)

P2P- They thought it was about file sharing, but its really about leveraging network effects Don't be afraid to be ahead of the curve. Keep telling your story. "The difference between a man and a sheep is that a sheep just bleats, but a man keeps saying the same thing in a different way till he gets what he wants."

Marketing- Identify a big trend, find things that matter

"Find a parade and get in front of it" [ cf. with earlier speakers; don't create a trend, ride in front of it ] - Jim Barksdale, Netscape

"People don't care aboput books, they care about issues." -Brian Erwin * Let's go out and use the book to promote the internet

Marketing Lessons- * Engage people- the Amazon 1-click campaign- go talk directly to the source * It's about people-Bring people together to tell your story. Use stories to bring reality to the big idea

Craigslist is the 7th top site on the internet- 18 employees "Craig is increasing the workforce by about 5%" :)

Employees

Yahoo 9K TimeWarner 85K Microsoft 61K Google 5K ebay 11K News Corp 38K Craigslist 18 Disney 129K BBC 60K IAC 26K

Craig is taking business away from the newspapers, not intentionally, but simply because his service is better! O'Reily called this "asymmetric competition." "Find a buisness that is ripe for disruption"
* Take the cost out of the business * Deliver what the users want

Best way to shoot ahead of the curve is to be passionate about something Passion allows you to harness your creative potential- trust yourself, trust your vision

Speaker: Paul Graham (Y Combinator)

TOPIC: "The hardest lessons for startups to learn" (Expect this speech to be posted as a fleshed-out essay to paulgraham.org within the week.)

Counterintutive Things that Startups Discover:

1.  Release Early (this echoes what was said by Bloglines guy)
* Get version 1.0 done fast (many have died because they were too slow to release)
* Improve it based on users reactions
* Since you plan on getting really popular, it's dangerous to assume you know your users- why not hear it from them first hand?
* Initial release acts as a shakedown (your product [problems] will be exposed)
* WuFoo Example: Releasing early led them to learn that Linux users didn't like Flash.  Duh?
* Most importantly- it makes you work harder- problems move from being intriguing to alarming
* If you're going to release something that doesn't do much, release it fast (then can +1 with feature adds)

  1. Pump out features

    • Improvements beget improvements
    • The more ideas you implement, the more you will have
    • Users love websites that get better all the time (see: Flickr)
    • Users are used to companies ignoring them and not taking them seriously
    • Keep thinking about how to improve your thing (SFW)

  2. Make users happy

    • A startup has to sing for its supper
    • Users are a "fickle wind", impersonal, casual, noncommital
    • Users will find what they want
    • When users arrive, they'll be poised, ready to fire the back button
    • <><> Don't assume users already know what you do upon arrival!
    • <><> Explain what you do in one or two sentences!
    • Give everything you've got to users right away- put it on the front page
    • Nothing will explain your site so well as using it
    • If you have decent growth, don't worry about if you're small- you'll be fine eventually

  3. Fear the right things

    • Startups are right to be paranoid but often fear the wrong things.
    • Disasters are normal
    • Fear not the established players, but other startups ("like you, they're cornered animals"
    • Most college graduates graduate thinking they need a job
    • You are your biggest threat (best ways to hose yourself): 1) Internal disputes 2) Inertia 3) Ignoring users
    • <><> Almost everyone's initial plan is broken! Be smart enough to listen!
    • Nothing is so interesting as having a beam of prototypes encounter a beam of users
    • "The imagination of nature is greater than the imagination of man." -- Feynmann

  4. Commitment is a self-fulfilling prophecy

    • Not intelligence. It's determination.
    • Biotech- professors commercialize existing technologies
    • In software- want to focus on grad students
    • Make up for lack of experience with determination
    • Do you have to be smart too? Yeah... but lacking commitment will kill you way quicker- imagine yourself as a corporate worker being asked to do the tasks of a startup founder
    • People will sense your ambivalence! Good things happen to your competitors but not to you... you will seem unlucky. Quoth T. Boone Pickens: "Everyone said I was lucky, but the harder I worked, the luckier I got."
    • Take money is to help it happen faster!
    • Successful running back- don't put your head down and run through people, spin moves, run backwards, whatever it takes- what he will never do is stand still

  5. There's always room for NEW STUFF

    • In 100 years, the only social software sites we'll have will be myspace and facebook? Nope.
    • "Why didn't anyone think of that before?" This will keep happening
    • PG doesn't think there is any limit to the number of startups- the only limit is the number of people that want to work that hard
    • Startups create wealth, which is addressing what people want, which for all intents and purposes is unlimited (cf. Hobbes Leviathan: The natural mode of man is avarice)

  6. Don't get your hopes up

    • Founders are naturally optimistic.
    • Treat optimism as you would the core of a nuclear reactor: as a source of power that can also be very dangerous. You have to shield it.
    • You have to draw the lines between what you expect of yourself and what you expect of other people
    • In a startup you're pushing the limits of... well basically everything
    • Shielding optimism is most important relating to deals- VC's won't invest in you, you won't get bought, that big company that says they're going to use you, they're not
    • Don't lean your company against something- If you think someone is going to invest money in you, you might stop looking. Resist that urge! To have kept looking, not only will you get better terms, but the backup is invaluable if the trapdoor opens
    • Deals are usually dynamic. Usually a lot of subsidiary questions to be cleared up post-handshake. If they sense weakness, they'll be tempted to screw you.
    • VCs are professional sharks- even if they're nice guys, they can't help but screw you (turtle and scorpion cross the river folktale)
    • [GURUDOM] The only leverage you can have in a deal is to NOT HAVE TO NEED IT
    • Continue to run your company as if the deal doesn't exist- it will behoove you in the long run NO MATTER THE OUTCOME

Conclusions * "I knew it would be hard, but I didn't realize it would be this hard" - Every startup founder ever * Economically- startups are best seen not as a way to get rich, but as a way to accelerate work * "If you love life, don't waste time, because time is what life is made of" - Ben Franklin * By getting the necessary task of making money out of the way quickly, you are in fact showing the ultimate respect for life!

Comments

I'd argue that "Google/Yahoo/Microsoft" (GYM) companies are not worth worrying about because they have other concerns that they need to handle that slow them down. Anything they release must be built with scale in mind (since it will automatically get users on day one), integration with existing sign-in mechanisms, etc. Whereas a startup usually is focusing on users before worrying about scaling up to millions of users. These approaches give very different times-to-market.

i think Google is doomed by its own arrogance. and the fact that so much of their revenue is bound to something that can easily be competed against : ads.

Mind if I play devil's advocate? How do you compete against ads? Yahoo/MSN are trying this exact approach. But you have to fight the same network effect that Ebay has. In order to get ads, you need to have content to display the ads on (AdSense, Yahoo Publisher Network, etc), and in order for publishers to display your ads, you have to have a big ad inventory.

All the ad servers are kind of dealing with this issue too- it's about behavioral targetting, and if google falls behind the times/ technology, then...

Yeah, if they fall behind, sure. But I'd argue that the data they've collected from users gives them a good longterm competitive advantage. But I suppose it's quite possible they become too focused on CPC or other ad models they're using, to be able to compete effectively with a different model. (The same way CPC sidelined everyone that was doing CPM banner ads.)

That beind said, I think google has a really good company culture that encourages keeping up- the 20% free time project is one of the best / coolest things ever

Yeah...but people keep working 19 hour days :-/

It's not work if you love what you do ^^ Aspire to startup lifestyle without the financial risk...or financial upside! Ergo, why do it? Risk averse smart people

Google has what's known as Founder's Awards, intended to reward people whose projects succeed and are important to Google. The goal is to discourage people from quitting, building a startup, and selling back to Google. (What people at Microsoft did). Instead you do it within Google, and get a Founder's Award. (Yeah, I'm sure they're very difficult to get, but so are most successful startups.)

Interesting! Is the remuneration on the par of had you been there at the IPO? There's no way it could be... I'm sure it's reasonable though

No idea on details like that. shrug But at the same time, I see a startup that takes on VC as giving the founders a small percentage of the company by the time they IPO or sell. Sure, MySpace sold for $500 million, but I heard the founders had to split $30 million among themselves.

Sounds like a good way to live on dreams. :-/

By the same token, I've heard Google likes to buy before companies get bloated with VC funding and are harder to "swallow". Buy at much smaller valuations, but before the founders get all/most of that smaller valuation. I imagine Chris Scacca will probably talk about that kind of stuff though, as I believe he used to do Google's M&A stuff.

Paul Graham's theory is that the little startup purchases (per the $3.75mn analysis done earlier) are really just signing bonuses for hiring smart coders. It's not about the technology, it's about the personnel. Hah - yes, the ultimate resume credential : I can make it big, don't you want me?

6800 employees, 1200 hired in the past 3 months. I'm sure many are international hires, but I can't walk down the street in san francisco without bumping into a google employee. They essentially double every year from their filings. But of course only a fraction of them are actually Software Engineers proper, as they hire a lot of marketing/sales/ops people as well.

What's their contractor / permanent ratio, i wonder? Not sure. Software engineers they have a tendancy to hire fulltime directly, but I believe some/many times the sales/marketing/recruiting/ops folk get hired as a contracter for a few months to "prove their worth" before Google decides whether they are good fulltime hires. There was a recent blog post about this from some guy who was interviewing in the Google Print division....good hourly wages, but not nearly stable enough for many people. So I think it depends on the position you're going for and such.

Speaker: Caterina Fake (Flickr)

("Fake" is pronounced like the synonym for "faux".)

tagline: "I love you sooo muuuuch!" [insert hamster crushing noise here]

alternate tagline: "On the internet, strangers are the source of everything good."

  • Nov 2005- Had to shut down their initial direction (game)- focus on their explosive user growth

Q: "If you were doing flickr again, what would you do differently?"

  • Team of 6 people- each person is incredibly important
  • Classic mistake- hired too fast... however that empowered them to build Flickr, so... mixed blessing
  • "Foo up fast development strategy." Strategy of if you're going to make mistakes, make them fast, the "

Q: "What did you do for usability testing?"

  • Almost none, got it out early as Paul also suggested
  • Flickr Alpha was the usability testing- "put it out with its zits and blemishes"
  • Heavy monitoring of/posting on their support forums
    • In response to a change in the product: "I had a really difficult childhood, and this new change you've just made has pushed me over the edge!"
  • "Users don't have a problem teling you that something really sucks..."

Q: "Why did Flickr not make the move into video sharing to take on YouTube?"

  • Initial concept- Flickr will be all things to all people- video, games, rich text, files, garageband, online collaboration tool, the list goes on
  • Must have for product development- you have something that people can point to and say "This is what we are"
  • Crucial that we didn't start the company to build a photo sharing site- just happened to evolve in that direction. If we had researched photo sharing sites, we would have realized it was a crowded market. We would have been scared.
  • "I have nothing to announce on that front, but I can't exclude the possibility."

Q: "How have things changed fo Flickr now that you've been acquired?"

  • No longer have the freedom to create brand from scratch and be in total control
  • Caterina shows her sadistic sense of humor with a story about her friend crushing his hamster while proclaiming his great love for him (a hug of death, as it were).

Q: "How did you first start getting some good press? PR firm? How did you start getting users?"

  • Necessity (Being underfunded) is the mother of invention- blogging / viral
  • 80% of initial users came due to blogs
  • Alpha-geek reporters started writing articles without talking to them
  • Eventually hired a PR firm- not to generate publicity, but to handle it
  • The UI evolved organically.
  • Strangers are the source of fear... however on the internet... strangers are the source of everything good
  • [GURUDOM] Turn users from interacting with the monitor, to interacting with each other
  • Greet people coming into forums (unheard of customer service! don't be corporate!)
  • [ Über-GURUDOM ]: Make a site where people want to come back, by letting them know that people who think they're great are expecting to see them there! Hey Goat-BLood I see you like Black Death metal, mr. satan from indiana would love to talk with you, etc. Let users build the sticky for you
  • Joining a community is like arriving at a party without knowing anybody- someone should take your coat, serve you a drunk, and rub your feet ( y-combinator take notice :P )

Comments

I wish flickr would eradicate myspace. myspace is so teenybopper

I feel like if you don't at least appreciate the appeal of myspace, you might have problems making things people want to use.

I understand it, but it's so... juvenile. Figure out how to make people with a lot of money want that same experience is something else....that is to say, adults. (i.e. I don't want to change my software, I want my users to change?)

It's just the feeling that you don't want to be using the same software to talk about yourself as the thing that the under 16 set use. Maybe a bit nicer, not so prone to " <3 pwn r0x0r" ,etc.

I mean, I totally agree with you, myspace is super crap as far as I'm concerned, but like it or not, there you have it, the american populace.

Speaker: Om Malik (Business 2.0)

SUMMARY: "What makes a startup work from a journalist's perspective."

  • Why did you start the company? It's not about the money...? "Everything in life is about the money" Nobody who's poor is ever going to get their idea across

  • The startups that succeed are the ones that cause behavior to change (this is contrary to the "ride a wave" we've been hearing about all day- it only applies on a grand scheme- the journalistic bias)

  • Users should not have to be locked-in at the get-go to discover your site This is why Ophoto, et. al. sucked, but Flickr (and Del.ici.ous) is great.

  • It's not just getting people to adopt technology, but getting them to use it time and time again

  • The "slacker culture" that's permeating corporate america is not a problem--it's a huge opportunity. Web traffic spikes between 1 and 2:30 across many timezones

  • "If I don't participate, I don't get the story." You have to make the tech-writers participate if you're going to get the press. Tech writers are a proxy for the end customer

Quotes

  • "When you think about starting a company, don't forget the money. Do not forget that, ever. EVER! It's about the MONEY!"
  • "Ghod dhammit thees is amherica, it's about tha money" (accent)
  • "I still don't care what tags are. I don't know what they mean"
  • "Web 2.0 companies are just doing cool stuff for geeks"
  • "What's cool is never profitable--unless you're Steve Jobs."
  • After lunch, you're sitting in your cube, not doing any work- "Has anybody noticed that between 1 and 2:30, YouTube performance declines?"
  • "Someone noticed people like to get drunk and buy things at night, thus QVC"
  • "YouTube's success shows that there's a lot of bored people in big companies"
  • "I just like to talk and pontificate and bring other people down... I do a good job of that"
  • "If I have to spend 2 hours to understand what you do, it's sooooo painful!"
  • "If you wait too long to get your point across, it may be too late..."
  • "When I'm writing for Business 2.0, it's like I'm writing to my mother. When I'm blogging, it's like I'm talking to you"
  • "Why's myspace successful? It's messed up like a teenagers room..."

Comments

I think believing the stuff he's saying is dangerous... he's kind of fish out of water a little bit...

Right, but most of the paying customers are like him. -- Hey i agree, firefox only isn't the way. to go....although with the GOOG advert'ing FF on their main page, perhaps that number will be changing. -- I don't think it's about firefox per se, but rather about "architectural purity," vs "what user wants." -- Yes, of course.

Because he's a journalist, most of the things we deal with slip under the radar for him. He only notices very large issues, which I think is maybe not the most fecund grounds for entrepreneurs.

It's cool and all to be passionate and to want to change the world, but it's suicide from a business perspective.

50-50 shot. Dispassionate people (MBAs) "all about the money" == pets.com. Passionate alphageeks with LISP-based solutions (may be over-intrigued be the problems to deliver). We call it "Lisp" these days. :)

Hah, did you ever hear the witticism that Harvard MBAs are a leading negative indicator? Now I have :) Hm, don't we have a president who is a Harvard MBA? Eep. He's batting 30% approval these days. Yet another company he's foo'd up, in the words of Fake.

Yeah, blogging is not cool. Ironically, Steve Jobs does not have a blog ;)

Zimbra is very sexy.

What was that product he liked? Netwipes? Net Vice Netvibes? I think netvibes rings a bell for me. So many geeks, so little internet around here :) TCP/IP over carrier pigeon lol

Speaker: Chris Sacca (Google). Head of Special Initiatives

SUMMARY: "Organize the world's information and make it universally accessible and useful."

sacca@google.com

He's currently engaged with the expanding the fiber optics

"It's never too early to start"

There's no point in overbuilding- stay as cheap as possible

  • How to successfully run a company

    • Food is where people let their guard down- pretty much all the good ideas are discovered during lunch / dinner
    • Try to be open with your users and employees (increase transparency)
      • Use open source software
      • Starting to blog, finally
      • TGIF- everyone gathers in the cafeteria and all the going ons of the company are discussed weekly (Noogler)
    • Provide laundry service (!)
    • You hit 1 to 2 productive hours a day, everything else is screwing around
    • Prioritize geeks over everything else
    • Recruit at startup school (heh)

  • The state of Web 2.0

    • Too many of the same pitches
      • Brilliant: We're at a crisis point on valley production: all the Web 2.0 stuff looks the exact same! WHY!!??
      • Flickr figured out interestingness. Which is missing from all the pitches I see. And most of the people.

  • Next Steps

    • Encourage people to build products that they themselves would use... but too specific! 24-27 yr olds. Now, think about building things for other people
    • Build things for OTHER PEOPLE:
      • Fortune 500 companies
      • People in the developing world
      • Demographics that are not you
    • Get out of your comfort zone. Travel, get a difference perspective
    • If you want to know what it's going to be like a year from now, visit Asia
    • Life is happening all around us- The Internet is supplemental! (Log off now and then)
    • Care! He's discouraged by the lack of passion that he sees- we're in a time where we can't afford to be apathetic
      • Who they are is expressed by what they do.
    • Best way to be interesting is to listen more than you talk.

"Feel free to email me, I read each one" sacca@google.com

Quotes

  • "We have a couple people working on c = 299,000,000 m/s, we think if we can push the formula even a little we can eke out a little performance"
  • "Geeks rule!"
  • "Interesting people make interesting companies"
  • "Forget about coding for yourself."
  • "We live in the past here [in the US]. If you want to see the future, go to Asia, go to Europe-- just leave this country."
  • I know there's some of you who are real estate developers in second life and you're like "no man, you don't need to log off..."
  • "We thought it was total bullshit that the government came and tried to take all your search information from us."
  • Dummies pitching to Google- "If you don't respond in 7 days I'm taking it to Microsoft"

Questions

Q: What's the worst thing about Google?

How do you stay small when you're getting big? Hard to collaborate over dinner

Q: Any suggestions to make women pitches more effective (frequently less forceful)?

Blogs and email are the great equalizer Organizing your thoughts- Has it been done before? How are you situated to address it?

Comments

This is like an ad for Google. Too much rah rah.

yep

It was the same thing at the previous startupschool.

Where's the beef? Bush bash, guaranteed crowd pleaser in NorCal!

yeah seriously, know your audience!

the money shot folks, the recruiting brochure, onscreen. no shit. Here's my question: Why such a long grueling interview process? 3 return interview? Bullshit. It's the cult of the geek. It's like the gauntlet in gang initiations. Destructive cults' techniques: first make the newbs suffer and then embrace them warmly. There are some more $ companies that don't disrespect you like that.

haha, "send me your resume". BRILLIANT!

go to ENTREPRENURIAL pow wow, imply send resumes

here it comes- "it's easier than ever to start a company... but why bother? work at google instead!"

Ah wow - something interesting!

Google saves children. By giving them ethernet. Impoverished children with brand new internet connections are a great business opportunity for your web 2.0 startup. nike represent!

Should we leave the comments in this doc or move them over to chat?

I like having them here when they're relevant to the talk. And for stuff I guess that you want posted to the web version.

Remember when he said I remember last year when people were collaboratively taking notes and ripping on the speaker... ummmmm... awkward :)

whoa... he's encouraging us to go riot... "organizing riots everywhere" (and making them universally accessible and useful?) (And socially acceptable? Isn't that co-opting the revolution?)

I remember one day in 1989 i saw the Guns n roses video "Paradise city" and i thought, had Axl said, go riot, they'de have done it.

V for Vendetta :)

That's great that they told the DOJ "no," but what about China.

shhhh, don't mention that (someone should ask him that in the Q & A)

A good pitch is nice, real code is nicer.

Speaker: Joshua Schachter (del.icio.us)

SUMMARY: ""

  • Backstory

    • Started out with huge individual tagged bookmark file, turned into muxway, got lots of readers
    • Made multi-user version: del.icio.us

  • The Problem

    • Feature proliferation- too many links, need a way to find them
    • Wrote code in small increments when he could
    • Make sure it's a problem you have to deal with- will solidify / encourage your passion- must be GENUINE
    • "You can't be the smartest but you can be the most passionate by finding a problem that is appropriate for you."

  • Ideas

    • Keep an idea log, keep even the bad ones
    • Make sure you revisit your idea list periodically
    • Try to figure out which ideas are fertile ground that are interesting to you and easy to implement.
    • (Before delicious took off) Tried to do a project or two a year besides the day job
    • Make lots of ideas and lots of implementations -- you can't get really good until you do it a whole lot.

  • Secrets- Not worth it; Don't keep them

    • VCs won't bother signing NDAs- ideas are cheap, passion and dedication are the real asset
    • Bad Example: Websites that say "coming soon" and collect emails- just release it
    • Talking about your ideas creates traction... if you don't talk no one will hear you.

  • Market

    • The best way to encourage people not to blog is to force them to
    • Any delicious competitor allows you to import your links from delicious, but only delicious allows you to export.

  • Listen

    • Fact: Delicious used to be called "social bookmarks" -- no one got it because IE calls bookmarks favorites, and that never would have been picked up on without usability testing.
    • Go to Starbucks and offer people lattes to sit down and talk about your startup
      • Don't give them tasks--that won't represent how they'd really use your product
      • Try to replicate the "sitting in your underwear at home" experience

  • Design

    • Design is kind of marketing and kind of speaking to people.
    • Throw out everything you don't really need to solve the problem
      • del doesn't allow users to send messages to each other. URLs yes, because that's in the context of the design. There's already a global instant messaging/email infrastructure that works fine, and people can use that.
    • URL is prime real estate, url as command-line

  • Scale

    • Don't spend a lot of time building an awesome system that can scale
      • You're doing it wrong anyway so you'll redo it later
      • Social scale -- A lot of systems do not grow up appropriately
    • Tiny features of your design will be huge cues to people's behavior later
    • Large flat spaces -- crinkle up space so people come together, not just by affinity, but figure out how to bring people together. Examples: Tags, familiar stranger

  • Abuse

    • Abuse, spam, attention theft, etc. eats up a lot of time
    • If this happens, it's at least a good sign that you have a system worth abusing
    • Include some level of logging (24 hrs) so you can go back and check
    • People will be surprisingly clever about finding ways to be annoying
    • Users will use your system in unexpected ways -- watch what people do, not what they claim.

  • Technology

    • Budget extra time to deal with technical issues
    • Get good, or find someone who's good

  • Lazy

    • Be lazy -- don't engage in low value transactions (Yak Shaving)
    • Hire an office assistant to deal with scheduling, hire somebody to talk to VC's, hand things off and hire people that are better than you

Quotes

  • "The idiots who want to harrass you are much smarter than you would expect."

Questions

  • Did you make money when it was just you doing del.icio.us?

No. I never worked at a software company, I still had a full-time job in finance, it probably would have been a conflict of interest if I had made money.

  • Enterprise vs. consumer apps advice

Most companies don't have someone in charge of knowledge infrastructure- Wikis are great, but who do you sell them to?

For something to make sense as an enterprise app, there has to be someone at the company who it will save a lot of time for.

The value of data.

  • What was the turning point when you decided to quit your job and do del.icio.us full-time?

My group at work blew up, my coworkers were very supportive. they told me to go try my thing and if it didn't work they'd hire me back. it was very low risk.

Make sure your co-founder is somebody you're comfortable working with!

Having a working product in hand is much more effective, when talking to VCs, than any pitch.

  • VCs in NYC vs West Coast?

Hiring in NYC is tough. Hard to compete with the big financials.

Hired people all over the place instead.

Fred Wilson is the guy. Beyond that it's tough.

Comments

I can tell from Joshua's first slide that he is a presentation/powerpoint master.

"Presentation Zen". :-?

http://presentationzen.blogs.com/presentationzen/atom.xml

That is such a brilliant site. Indeed. If you're stuck using MS PowerPoint, also check out Beyond Bullet Points. It's funny his programming style matches his presentation style ;)

Supposedly Einstein kept little baskets all around his house and puts notes in them and then once a week gathered them and reviewed for good ideas to follow up on

"Everyone else has an import-from-delicious feature, we have a delete-all-my-bookmarks-and-take-my-ball-home feature." i think that's important, people do pick up on that. Everyone realizes that import-only, no export is a statement.

Haha. go to starbucks in your underwear and talk about your startup haha

What does he mean that URL is prime real estate? I understand not having session id, .php, etc, but what do you put there? --rob

del's url scheme is very simple: del.iciou.us/lemonodor, del.icio.us/popular.

Speaker: 8 Founders

From right to left (order of speaking usually):

  1. Beau? - Pixoh (at podium)
  2. Kevin Hale - Wufoo
  3. Justin - Kiko Calendar
  4. Steve Huffman- Reddit
  5. Adam- Inkling Markets: Prediction Software for enterprise
  6. Sam Altman- Flipt- GPS
  7. Mat - Flagr
  8. Srini - YouOS

Quotes

"If we could do it all over again, we'd still be using Lisp. Probably." "The number one story on reddit is "Fix this bug" jerk." - Steve, Reddit

Questions

Q: Where do you find cofounders?

Akin to- where can I find a wife that I can have a kid with right away Smart and hard working

Q: What would you do differently if you could do it all over again?

I wish I started earlier. I was just stopping myself.

Wufoo?: LLC bad- pain in the a$$

Reddit: Stick with Lisp. If you want it done right, do it yourself- hosting- Avoid A+.net

More conversations and agreements needed about e.g., vision.

Flipt: Control your own destiny- I wouldn't have started a company that's so dependent on 3rd parties (i.e., the mobile carriers)

Have the tough conversations up front.

YouOS: Don't keep it secret. It doesn't help out so much.

Q: Stock option plan- timing, logistics?

Legal pain- but just make your lawyers do it on Day 1

Give way above market stock grants

Q: The most influential books in starting your company?

Justin: Read anything by Seth Godin. (Seth has an awesome talk @ Google on video.google.com)

Wufoo: The 10 Day MBA

Q: How do you find investors?

Mat- People you know from the past. People you've impressed.

Q: What was the hardest thing about starting a company?

Adam: Just starting it. Regular jobs weigh you down with materialism and a bunch of crap you don't need.

Q: What will an investor ask you?

Sam: Trying to raise money- why can't someone else do this? How will you make money? (The good ones won't care if you don't know) I think he said that they took a series A from Sequoia (verify??) and NEA.

Board meeting- Do users love this? Are you making this as good as you possibly can? How are you hiring?

Q: What happends when you get lots of users?

Users start to expose a lot of holes in your system. Users generally want to help you.

"It kind of sucks when you wake up in the morning and the number one story on Redit is fix this bug"

Srini: (Web operating system, YouOS)- Big gorillas are also trying to do it. 50% users are from China because they don't have a computer.

Q: What keeps you motivated when things aren't looking good?

Pictures and other reminders of their dream of sailing around the world.

Reddit: Motivation driven by traffic and feedback. When things are going well, they think they'll save the world, but when things are done, well, they think maybe they can make some money.

Q: What things change when your company grows larger?

Sam: You have to pay people real money- which is good because you dev lots of cool stuff, but... you lose the culture of a strapped startup, the urgency

Q: From audience: How do you deal with your personal relationships (with girlfriends, family, etc.)?

"Work life balance gets skewed when you're working 16 hours / day... Well, I'm single, let's put it this way"

"You can't just work til you die"

Make the time to do or you're just going to break down and die.

Q: (From audience) Could you be more specific about the conversations you wish you would've had with cofounders at the beginning.

Kiko: Referring to his own situation... if you have a cofounder that's not doing anything... you guys need to be on the same page. Kick might be the only way.

Q: (From audience) Everyone seems to say put off taking money as long as possible. How is Y Combinator helping you?

Wufoo: We knew we were going to start a startup a year-and-a-half before we met Y combinator.

Inkling: Y-Combinator doesn't give that much money, so it's not really an issue at all.Helping get a clue on all of the questions that need to be answered.

Q: (From audience) Worklife balance- What do you think about the idea of not having a worklife dichotomy? Make it all one thing?

Flipt: If you can make your workplace your home and place to have fun, that's the ideal.

Pixoh: Think about a big idea, and that's the way.

Q: How do you vet other cofounders? If you're an engineer?

References and word of mouth are the only way they hire now. Even a VERY strong recommendation is iffy. Someone we know.

You can do a lot more than you give yourself credit for- The only way that's going to hurt you is if you believe it

Q: What has surprised you most? What was not told you at the last startup school that surprised you the most?

Pixoh: I thought that there was something magical about people who started companies. Brilliant, experienced, etc. The one constant in the people we met through Y Combinator was that they were driven. Determined.

Wufoo: There's tons of more paperwork than you think. Getting a good lawyer is worth a lot to save you from taking big lumps down the road.

Reddit: Note- "Not all lawyers are competent"

Q: Do you have any product management experience? Do you think that has any value?

Inkling: Not an engineer. Political science major. Had some product management experience in previous job.

Flipt: It'll be a dark day when we hire somebody with the title "Product Manager," we try to hire engineers that could be product managers if they wanted to be

Q: ? [Couldn't hear it.]

Q: How many of your cofounders do NOT know how to code?

You want somebody who knows about coding?

Q: How important is good (graphic?) design?

Looking for a design counselor more than a design guru. [Flash sucks! Linux folks freaked out. Check out Laszlo, they're pretty sweet] Acquiesing your own ego.

Q: Do you need to live with/near your cofounders? How possible is it to do a remote founder thing?

Pixoh: "You can tell from the commit log when I moved in." Real-time bouncing ideas off of each other is very valuable.

Q: Have you outsourced anything?

Flipt: We've outsourced customer service. We do the rest inhouse.

Q: American citizenship for foreign co-founders?

I'm from Canada. It's much easier if you have a university degree. You can get an H-Visa.Is it worth it to startup in the vallley or in Cambrige?

Proximity is important to be able to have good conversations.

Yeah that was cool

Comments

Everyone sounds sleepy

Subethaedit is so awesome for rapidfire note taking!!

Indeed, it was great the way the names got filled in

Extreme coding, meet extreme note taking.

Well that was fun kids. I live in SF, so drop me a line if you want to grab lunch sometime - jeremy (see below)

Dinner tonight? sorry i got plans :( have fun though! I will probably go to that mountain view party though- check the wiki

Great work. That was great to watch. (and participate a little.)

"Nomad Hackers" Google group

Notetakers

Hey all I gotta go home now! This has been a blast and i'll be checking all your sites once i get back to the civilized wworld.

last updated 2 years ago #