A while ago I announced that I was starting my own internet startup. Since then, I’ve gotten nothing done. Well, not nothing, but nothing that resembles a startup. I did register a domain, setup WordPress, and write a few blog posts, but that was it. And then in a WordPress accident while backing up the database for this blog, I wiped out the blog posts I had written for the startup blog, so now it just links back here. So, since I didn’t have any code or users, and I now no longer even have a blog, I’m pronouncing SmallPunch dead (for now).

So if I didn’t lose much, why quit? A couple reasons.

First, determination and focus are two of the most important requirements to succeed at a startup, especially for one as time constrained as myself (part-time single founder). Right now because of my family situation (2nd baby due any day now), my focus is definitely not on a startup. All of my attention is focused on balancing home needs with work needs and there’s very little time for anything else.

Second, while a part-time founder might succeed with regular, predictable time periods to work, my available time is neither regular nor predictable. I take what I can get, but that’s hardly conducive to getting into the right mindset or the flow. Third, I still feel like I have some infrastructure and technology issues to work out. I’m sticking with Lisp because I think it will help me get the most done in the alloted time I have (I don’t see my free time increasing greatly anytime in the next 18 years), but that carries some costs. I’m still figuring out which CL implementation to go with. Since I lack a decent computer that I can run Linux on, I’m going with CLisp, but I’m a little weary of its lack of threading since I will be doing a web app.

Third, I’m planning on using Weblocks as my web framework, but since it’s at something like version 0.1, it’s only been tested and run on SBCL, so who knows what problems I’d run into using it on CLisp. These are not insurmountable issues, but I do need to take the time to hack them out, and it’s hard to do a good job at that with the psychic weight of a startup hanging over my head.

Fourth, I’d really like to have someone to work with. That’s part of my motive for starting the Chicago Lisp User Group and contributing to Hacker News. I think I might be able to succeed on my own, but for motivation and load-sharing purposes, a co-founder would be really nice. But I want it to be someone I know and am comfortable working with, so I can’t just whip that together.

Here’s what I have accomplished in that time frame:

• Got familiar with Emacs/SLIME as a development environment.
• Written blog posts that have been read by thousands of people.
• Made contacts/friends with other people in the Lisp and startup communities.
• Organized the Chicago Lisp User Group.
• Been an active beta tester for several new startup products.
• Continued working through the best Lisp books out there (Structure and Interpretation of Computer Programs, Practical Common Lisp, Paradigms of Artificial Intelligence Programming)

I’m not exactly lounging! So why do I call it a “Roger Clemens Retirement”? The man retired in 2003, un-retired in 2004 without missing a single game, threatened retirement in 2005 but accepted a generous raise, retired in early 2006 only to un-retire again in the middle of the season, and then retired for the last (?) time in 2007. Long story short, just because he was “out of the game” didn’t really mean he was out of the game. Just like me.

Fred Wilson at Union Square Ventures just wrote a post called “Can The Y Combinator Idea Turn Into A Movement?” that set of an interesting discussion at Hacker News. He says that because you can start a company for a small amount of money, investors should back many companies with a small investment instead of a few companies with a large investment ($25,000 * 10 instead of $250,000 * 1). If you have read Paul Graham, know anything about YCombinator, or even just know some basic economics (my current mental hobby horse), this isn’t news or even particularly insightful. It was the discussion around the post that raised some interesting points.

[Read more…]

[This is another post where I was writing a book-length comment somewhere and thought I’d just publish it in case anyone else cared. I should keep stats on things like this].

About a month ago, a new startup called Coordinatr announced their launch on Hacker News. I didn’t try it out at the time but I filed it in my mind for later use. Now that I’m the DFGL (de facto group leader) for the new Chicago Lisp User Group, I thought it would be a great way to cut down on the bajillion emails it takes to organize a group of people (combinatorial explosion is not an organizer’s friend). I gave Coordinatr a spin and was quite pleased with the results. I also reread the comments in the discussion on HN and found that the team had incorporated much of the feedback given. Then, when I offered some feedback of my own using their slick collapsible feedback sidebar, I got a response from the team within less than 10 minutes. There are a few rough edges left, but these guys truly care about doing a good job and who knows, half these things might be fixed by the time you read this!

The basic pitch for Coordinatr is “Easy planning for spontaneous events”. These are the “What should we do this weekend?”, “Where should we go to dinner?” type discussions that everyone hates and are irritating enough that it’s almost not worth hanging out with your friends because no one can make a decision. It’s based on real life, so it includes SMS notification for a variety of events (lots of people aren’t always checking email and not everyone has an iPhone). I’ve just setup a few events for the Chicago Lisp User Group that I’m now organizing and setup was a snap. The event is a few weeks away so I haven’t seen it in action yet, but I’ll follow up when I do.

[Read more…]

Normally, when I read something on the Internet, I have one of three reactions:

1. “This guy is so wrong he’s not worth acknowledging.”
2. “This guy is stating the obvious so he’s not worth acknowledging.”
3. “This guy is so right there’s nothing to add.”

Sometimes, very rarely, there’s a fourth reaction: “This guy has a really good point, but something’s not quite right.” So, the Reaction #4 award of the day goes to this post on Entrepreneur2Be, called “Why There Will Never Be A Startup (Bubble) Bust“. I realized I was entering #4 territory when I started writing a really long, well-formatted comment that I had to use PageDown several times to read. If I have so much to say, I might as well put it here so others can read it to. So here goes (I think I’ll add more #4 posts in the future):

SUMMARY:

He says that people think there’s a startup bubble because so many crazy little companies are getting funded. There is no bubble because:

1. There are billions of people still waiting to get on the Internet
2. People can switch products quickly online
3. The Internet makes real life products and services (package tracking, listening to music, etc) better
4. The audience wasn’t there in the 2K bubble burst

These factors mean that Internet growth will eventually slow as they catch up to the real world, but since the cost of starting a company is going down, more and more startups will continue to be created.

MY ANALYSIS:

I agree with the main point (no bubble bursting) but not so much with your arguments. (well, I do like #3 and #4). If I had to give my arguments for it, I would say:

1. The internet lets people with common interests that would never have found each other find each other and create new markets that wouldn’t otherwise exist (like the zillion dollar internet knitting communities that keep getting mentioned, or Lisp programmers, etc). This lowers the cost to supply a new product, which will create more value (in a microeconomic sense).
2. Innovation in the physical world has been going on for thousands of years, and online for about 15. There’s lots of room for trial and error and improvement. I’m sure there are big ideas (like the equivalent of car financing in the real world) that will change the way people work on the internet.
3. There hasn’t been a generation of programmers/entrepreneurs yet that had the WWW around for their whole life. I think their perspective will create a lot of new ideas because they will lack older (possibly outdated) assumptions
4. The economics of starting a startup in college (something Paul Graham and others are promoting) are so favorable now that participation will continue to increase. A lot of really awful startups will get done by college students, and a few outstanding ones will also happen. A la “The Black Swan”, if 10,000 new startups are formed a year, and 100 of them become successful businesses or acquisitions, and 1 of them IPOs, that’s a lot more value created than the 99,899 that go nowhere lose (especially if they just go back to school or try again next year).
5. Some big problems still exist on the internet (trust, privacy, identity segmentation, etc) that will open huge new opportunities when they are solved.
6. As long as the only bubbly characteristic is “lots of little angel and VC investments”, then generally the only people that can get hurt are rich people. I’m sure it still stings, but the difference between and angel investor blowing $500K on some college students or a ski condo is a wash for the economy as a whole.
7. In the 90s, a successful company was one that IPO’ed for 9 figures and sucked cash out of investors from around the world. The much more common successful exit now is acquisition by a big company. That’s not a big deal because
   1. most acquisition prices are small compared to the acquirer’s resources (even a $1B purchase by Google won’t break the bank) so there’s little chance of completely wiping our investor value
   2. acquisitions are generally inline with the same company fundamentals that cause someone to buy stock in the parent company, whereas Bubble IPOs were based on a bunch of enthusiastic question marks
   3. Acquisition pairs a new innovation with a successful business model. IPO pairs a new innovation with a one-time financial windfall that doesn’t mean jack for the ongoing success of the business

Overall, it doesn’t matter what reasons anyone offers for why we are or aren’t in a startup bubble because history will decide that for us. But I think that E2B and I are onto something.

Buy Arnold Body Building Products!

Part of the joys of staying up all night is that I don’t have to fight with my wife over what movies to watch. My latest guilty pleasure was the bodybuilding documentary Pumping Iron, starring Arnold Schwarzenegger, Lou Ferrigno and others leading up to the Mr Olympia 1975 contest. I grew up an Arnold fan (let’s just say I could recite all of Conan the Destroyer as a 7 year old) so this was an interesting look into his life before being a movie star and politician. One new thing I picked up on is how seriously he took all of his competitions. At the time of the movie, he had won 5 consecutive Mr. Olympia titles, he had been the youngest Mr Olympia ever, and he hadn’t lost a competition in 5 years. He was the undisputed king of the body building world. He knew it, and everyone else knew it.

But he wasn’t content to rely on the strength of his reputation or his preparation. He went out of his way to get in the heads of his competitors, to psych them out before they even got up on stage. On the morning of the competition, he ate breakfast with Lou Ferrigno and his family, and while he seemed friendly and jovial, he was undermining Lou with his comments. He said things like “Oh yeah, Lou’s got so much potential, he’s got a great career ahead of him.” and “Lou, this competition today will be a great learning experience for you.” He planted the thought in Lou’s head that winning wasn’t even a possibility. Now there was little chance that anyone could unseat Arnold (he won his 6th straight title that day), but he didn’t take any chances with a potential competitor.

Another (slightly disturbing) thing Arnold said was how weight lifting and the pump you get, the rush of posing on stage, was like having sex, even better than sex. So when he worked out and competed, it was like having sex all the time. It certainly made me glad I wasn’t his workout partner :). But once I got over my creeps, I realized that if that’s truly how the reward system in his brain worked, no wonder he became world class so young and stayed so long. If what you do for your job also happens to be the thing that gives you the most pleasure, then you can’t help but excel. If you have found what you love, and it’s so satisfying that you can’t stop working at it, you will be on your way to success. Great success takes years of consistent work, and you have to love something a lot to put in that effort. Arnold is just another example of that.

Lou Ferrigno, on the other hand, seemed to lack that same internal motivation. Now don’t get me wrong, I love Lou and thought his height and size made me more impressive than Arnold on the stage. I must also confess that seeing his face strain while he was working out made me think a car was about to get flipped over or a hyperbaric chamber was going to get ripped apart! But whereas Arnold tended to do his own workouts or take time to help others with their technique, Lou always had his father, brother, and friends with him. They pushed him, they told him how hard he had to work, they told him he had to beat Arnold. He would try to quit halfway through his set, saying he didn’t want it anymore, but his dad would tell him he had to finish. Now don’t get me wrong, Lou had to work his guts out to get where he did, and he was still young at the time, but he seemed like he was doing what he was told instead of doing what he loved. He had the size and build to compete, which none of his crew did, but at least the way the movie showed him, he didn’t seem like he was intrinsically motivated, even though he worked hard.

It’s simple to draw a parallel between these two and types of startup founders. One loves what he’s doing and would do anything to succeed, one is great at what he’s doing but can’t stay motivated on his own. And if I had to honestly assess myself, I’d say that right now, I’m a Lou, not an Arnold. I haven’t gotten to the point where success breeds more success, where I’ve got enough of an application that build more is intoxicating. I’m still fleshing out ideas, improving my familiarity with the tools I’m going to use, and doing infrastructure setup. I know I can get to the Arnold level of motivation once I get to the sweet spot of the work, because I’ve been there before. I can get the same way with reading, and more recently with writing.

What I don’t know if I can do is to be cutthroat with my competition. I have that horrible gene that geeks have makes them think the best product/person/team/etc should win on its merits, and that using any methods besides excellence to win is cheating. I suppose I can remove some of this cognitive dissonance by telling myself that a clearly inferior product deserves to have its weaknesses exploited. But for now, I’m so early in the process that my offering isn’t even worse.

If you lack the passion for what you’re doing or the dedication to bend the rules in your favor, you’re going to have a tough time succeeding. But if you realize it, you can change it. Change takes time, so don’t wait. It’s time to go pump some code!

You are where you live. That’s the message that Prizm, a market research tool from the company Claritas says. In the United States, groups of similar people tend to live near each other, and the size of these groups roughly corresponds to ZIP code boundaries. Claritas has identified 66 lifestyle types, and they geographically aggregated demographic information to companies looking to refine targeted marketing. This demographic information has proven to be very valuable. Try your zip code and I think you’ll agree they’ve done their homework.

The web magnifies this effect. On the web, companies know more about you as an individual as opposed to a member of a group. They track your individual activity in a way that physical companies couldn’t dream of. Search engines have managed to turn one activity, answering simple questions (search queries), into one of the most profitable businesses in the history of the world. The combination of wide appeal, high effectiveness, ease of use, low transaction cost, and an effective method of monetization (paid ads alongside normal search results) has produced hundreds of billions of dollars of wealth and many billions in revenue and profit over the last 10 years, Google being the leader.

Google has over 50 products (not counting Labs products), including dozens that aren’t variations on search. Google is the undisputed king of search, but none of its non-search products are #1 in their category (except YouTube, but that was an acquisition). Orkut is way, way behind MySpace and Facebook in popularity. Finance is behind Yahoo. Maps, despite redefining how online maps should work a few years ago, still isn’t as popular as others (it boggles my mind that people still use MapQuest). Talk is way behind AIM and MSN. Gmail is a fraction the size of Hotmail or Yahoo Mail. I’ve never heard anyone say anything positive about Checkout. The list goes on. How can Google, which completely owns search, be so dysfunctional in every other space?

Because it’s not trying to win those categories!

Google made $10.6 billion in 2006, a 73% increase from the previous year. That’s almost $1B/month (and at the rate they’re growing, it probably is by now). Nearly all of that comes from two products: AdWords (paid search ads) and AdSense (contextual ads on other websites). Simply put, for any of their other products to even be a blip in their revenue, it would have to be one of the most successful software projects in history! Google is smart enough to know this, and so what they’re really trying to do with these other products is to gather usage data and refine their picture of customers.

There are really only two ways Google can bump its revenue: increase Internet usage in general, and improve monetization of search. Google is indeed trying to increase internet usage (through things like the citywide wifi initiatives, bidding on the 700 MHz spectrum, and the Android smart phone platform) because they’re the biggest provider of one of the most common and most profitable activities on the web. But their ability to speed this up relative to general societal trends is small. On the other hand, if they can improve their search monetization even slightly, they make swimming pools worth of cash. If data mining users’ search queries against email messages against blog entries and comments could improve monetization by just 1%, that’s $100,000,000 a year. A couple percent improvement and pretty soon you’re talking about real money!

How could usage data from other products improve search? Here are a couple ideas off the top of my head. Right now (as far as I know), Google returns the same search results to everyone, and just counts on covering all interpretations of a query in the first few terms. If they ever do (do they already?) offer personalized search results, they could use other data to get a better idea of how to answer a query. For instance, geographical ambiguities would disappear (Aurora, IL vs Aurora, CO, Portland, ME vs Portland, ME, Springfield wherever), and terms with multiple meanings could be handled better (if you’ve never asked or talked about programming, then a Lisp is a speech impediment). Heck, location specific results could be tailored to information in your calendar that identifies where you are (ditto with mobile queries). Searches for products could return results from Base. I don’t know how advanced their Natural Language Processing is, but if they could understand your email and documents, then they could get to know you pretty well. They could go beyond what you ask to who you are and what you feel. Even if they don’t give personalized results, if they could create hundreds (or thousands or more) of “lifestyle types” (as opposed to Claritas’ 66), they could target general search results more effectively. I’m going to stop this now because it’s starting to scare me.

Plus, by being second or third (or eighth or whatever), Google can still get a useful amount of data without worrying about running into antitrust regulations. They (especially Eric Schmidt) have seen what happened to Microsoft when they bought or crushed competitors in different markets, and how much of Microsoft’s decision making in the late 1990’s was cramped by fear of the Justice Department. So it doesn’t matter if Knol is smaller than Squidoo, if Base is less popular than Craigslist, or if Docs never kills Office. If the data these apps collect on users helps improve their searches even the tiniest smidge, Google makes much more money than the company they’re losing to (well, maybe not with Docs). Interestingly, for the area where they’re most likely to run into antitrust issues (online advertising), they find it worth the risk to make a valuable acquisition (DoubleClick).

What does this mean for a company that Google copies? People used to be terrified that Microsoft would move into their space, because they would make a similar product and then just crush you, and all you accomplished was “doing market research for Microsoft”. With the internet, though, so much of the value is in the quality and size of the audience and the data they contribute. Many popular sites have technology that wouldn’t be that hard to duplicate, but their community is irreplaceable (take Google Video vs. YouTube for example). This means that if Google wants what you have, they have to buy you. Second, if they copy you, all it means is that you’ve found a useful online activity that they can mine behavioral data from. They will build something to get a representative data sample, but no matter how valuable your business is to you, it’s less than a drop in the bucket for them. Just consider your idea validated and keep your nose to the grindstone.

I’ve expressed my admiration for Paul Graham a number of times already in this young blog. He’s a successful entrepreneur, investor, innovator, writer, philosopher, language designer, and hacker. Most people would be happy to be as good at one thing as he is at any of those things. Today, let me add to his mystique.

I’m a regular on Paul Graham’s Hacker News site, but I had a hard time following the conversations and getting the full value of the site until I would hystry.com by Kartik Agaram. Kartik scraped all of the comments from Hacker News and displayed them with some extra formatting and features. There were some neat things like the ability to filter out conversations you were unintereested in and see all the parents any comment. But the thing I liked most was the way each comment had the article it related to next to it so it was easy to visually filter conversations.

Well, one day it just stopped loading new comments. I figured it was temporary and waited a day or two for it to come back, but when it didn’t, I emailed Kartik to see what was going on. He said his crawler’s email address was banned because it was putting too much of a load on the Hacker News server. That was certainly reasonable of pg to do, but I wanted to make sure he knew that I found hystry’s features to be valuable. I thought the best way to let Paul know would be to post a request on Hacker News, with a slightly inflammatory title (Reinstate hystry’s Hacker News) and a much more polite request in the text field.

What happened after that is pretty amazing. It’s worth reading through the entire thread, noting the time stamps for when things happened. To make a long story short, Paul made three changes to his running site in real time response to a dialog he was having on that discussion thread. Total time for feature request, notification, communication, two design iterations and a bug fix: 7 hours. Here’s a timeline of what happened: (note: I can’t say exactly how long this took because posts over an hour old don’t show minutes)

• Start – I make the request to let hystry crawl again because I like the way they displayed the content
• After 3 hours – After some other people commented on the issue, Paul explains that it was hard on his server and asks if I’d like anything changed on Hacker News itself
• After 4 hours – I make my recommendation and Paul makes a change to the live site within the hour
• After 5 hours – Someone recommends an alternative design and Paul implements that suggestion, again withing the hour
• After 6 hours – I notice a bug and point it out
• After 7 hours – Paul fixes the bug

Here’s what the new site looks like with the root link shown:

How did Paul make these changes so fast? Hacker News is written in Arc, his new language built on top of MzScheme. I don’t know whether the web server is his own or if it comes with MzScheme, but he has REPL access to the running server. So he made changes to the site’s code while it was running and changes were instant. Hacker News doesn’t have customers per se, but choice of technology and architecture let him provide an unforgettable user experience. This should certainly raise an eyebrow for anyone who has had deployment or downtime issues while making changes to their site.

If one of your users had a reasonable feature request, how long would it take you to publish it?

Comments at Hacker News

Comments at programming.reddit.com

First spewing-Coke-out-my-nose funny comment of the day: “Nice. Here I thought this was an article about Graham rewriting some history. I was pleasantly surprised.” -now look back at the title of the article. Thanks boredzo!

UPDATE: Several people have pointed out that this isn’t some technological miracle and you could do the same thing with X (for many values of X). I know. That’s why I spent about 5% of the article on it. I was impressed with his response to unsolicited feedback that was sent to him indirectly, and that he didn’t just slap out the change, he made two design iterations and a bug fix while communicating almost live with users. Much more of a social and business feat than technological.