Messing Around At The Fringe

Jonathan Libov has spent a couple years at USV in our two year analyst program and is about to move on to his next thing.

He took the time this week to write down some thoughts on what he has learned as a “junior VC” and it’s worth reading for both investors and entrepreneurs.

I particularly like where he ends the post, talking about “messing around on the fringes”:

But more importantly, talk to people who are building things. Don’t just talk to founders of investible companies, but also talk to people who are experts in their field or following modes of thought that contravene mainstream thinking. Play with their API’s and go deep in forums, even if there’s no investment in sight. Mess around in the corners of the internet.

Just as wading in the muck of cap tables and share purchase agreements is the best way for an analyst to learn the mechanics of venture investing, wading in code and API’s and forums is the best way to discover the things that others have not discovered yet. To frame it more practically: This will help you identify the opportunities that others don’t appreciate yet, and that will help you to invest at attractive prices.

Lastly, it is important to emphasize who this post is aimed at: Junior VC’s. The partners at your firm benefit from having established networks that will pass along interesting, valuable opportunities. Your job is to complement those investments with the ones they and their network would not see because they’re don’t fit the patterns they’re accustomed to. You have more freedom than they do to mess around on the fringe. Don’t waste it.

I try to hang out on the fringe as much as I can and also try to avoid being pulled into the mainstream as much as I can. I think Jonathan’s advice, while aimed at “junior VCs”, is great advice for entrepreneurs and investors alike. The next big thing is almost certainly a fringe thing today. That’s how it always is.

#VC & Technology

Comments (Archived):

  1. kirklove


  2. JimHirshfield

    Imma hang by the fridge.

    1. Twain Twain

      Haha, watermelon ice lollies are awaiting me in my fridge!

    2. Jess Bachman

      Ha! Yeah I thought this post was going to be about some IoT connected fridge. I’m glad it wasnt.

      1. JimHirshfield

        The fringe of the fridge is a scary place.

        1. Jess Bachman

          Indeed. I’m pretty sure I have a tub of hummus in there that violates the Geneva Protocal.

          1. JimHirshfield

            Oh dear, what a patè.

      2. awaldstein

        I tried to build one in the late 90s. before IoT was a dream.

  3. andyswan

    Twilio seems like a perfect example where the “fringe” led to investment conviction.Hanging around hack-a-thons or chatting with hobbyists it became clear that Twilio was not only starting to be used by every developer, but very enthusiastically so.

    1. Rick Mason

      Saw an interview with a VC who was asked why he passed on Twilio’s A round. He said they had a great team and passionate users but he wasn’t convinced it would ever be that big a market.Not only was my estimate way low he said but the market proved way larger than even the Twilio founders estimated. I’m convinced he never left his office to arrive at his market estimate, every developer I knew at the time was playing with Twilio.

  4. William Mougayar

    Shhh. I thought we were keeping it a secret.But hanging at the fringe and on the streets (as I like to call it) requires patience. You need to roll your sleeves and really care about who you’re hanging with.

  5. Chris Phenner

    ‘Mess around in the corners of the internet.’

      1. Chris Phenner

        John Candy, in ‘Planes, Trains…’ playing the keyboard on his dashboard as Neil sleeps…

        1. JimHirshfield

          Ah! John Candy. I miss that guy.

          1. LE

            Fat comedians die much to young. My favorite was Chris Farley.…They give up their health to ply their trade.

  6. jason wright…definitely worth hanging out with androids to get a feel for what’s coming next.

    1. sigmaalgebra

      Big things have small beginnings.The first time I heard that was in the movie Lawrence of Arabia.IIRC the response was, “Does the Arab Bureau want a big thing in Arabia?”

      1. jason wright

        That’s the android’s favourite film.”There is nothing in the desert and no man needs nothing” is David’s quote of Lawrence.

        1. sigmaalgebra

          Just from memory, maybe not quite correct:Faisal: I think you are another of these desert loving English. Gordon of Kartoum. No Arab loves the desert. We love grass and green trees. There’s nothing in the desert, and no man needs nothing.Dryden: Are you really going to give them artillery?Colonel: I was wondering that myself, Sir.Dryden: If you give them artillery, you’ve made them independent.Allenby: Then I can’t give them artillery.Dryden: For you to say, Sir.Allenby: No it’s not. I’ve got rules to follow. Not like that poor chap. He’s riding the whirlwind.Dryden: Let’s hope we’re not.I can never forget that exchange for its bureaucrats with their extreme, shameless, manipulative, unhesitantly duplicitous, cover their asses behavior!Have to suspect that such is a common situation for anyone trying to do something different!

  7. Twain Twain

    I’m so deep in the fringes of AI that the tech gaps I’ve been highlighting for 5+ years have only just started appearing in specialist tech sites (so still not mainstream):* https://www.technologyrevie…Experimentation is one of the big pluses about hackathons.My track record is that, of 8 hackathons, in 7 my team has placed Top 3; I’ve been the sole common factor in the team.These are big hackathons like DevWeek 2016 in SF, which had 1,200 registered developers and took up all the floors of Galvanize. We placed #2 in that.It’s fun to go up to total strangers, ask them what they’re thinking of working on, change their minds about the product, learn new tools with them whilst “playing at the fringes” and “make s*** happen”.Hacks rarely become startups. People prefer to earn regular salaries at the big techcos in SF/SV.Still, fringe is where we can make really creative products and gain hands-on insights into the future and how different components of tech fit together.

  8. WA

    Bravo. Priceless.

  9. kidmercury

    few things are more at the intersection of truth and fringe than kookology. #kookpride

    1. JimHirshfield

      Do you have any evidence to back that up?

      1. kidmercury

        is 9/11 truth still at the fringe? i think it’s not so-fringe-y anymore. outside of the US it is widely recognized, though i’m not so sure about here in the land of the free. there’s enough evidence to get a trial going against dick cheney for treason and a bunch of CIA agents — richard blee at the very least — for obstruction of justice (interfering with the FBI investigation).concerns over vaccination also started at the kook fringe. now, they come up in prime time presidential debates.

        1. Lawrence Brass

          With all due respect, this is not a judgement, just something that I have observed on a very reduced ‘sample’ set. Please forgive me.Every friend of mine that like to smoke pot and that watch youtube videos at the same time, eventually end up getting very kooky. The problem is that after the high, all the ideas and concepts absorbed during the session seem to persist.

    2. jason wright

      What’s the origin of that word?

      1. JimHirshfield


        1. Girish Mehta

          Is out there.

          1. JimHirshfield

            “out there”, as in, on the fringe?

          2. Girish Mehta

            Yes.As Scully would say, but so are the lies.

          3. Girish Mehta


      2. kidmercury

        of what word? kookology? there is no origin of that word, i made it up. though i’d be delighted if it gained usage and became a recognized word! 🙂

        1. jason wright

          Ah, so there is an origin…you.But what’s a ‘kook’?

  10. Hu Man

    How much money does one need to be a VC?

    1. JimHirshfield

      $0. It’s all OPM.

      1. Hu Man

        Like a hedge fund?

        1. JimHirshfield

          Like ANY fund, it’s mostly made up of Other People’s Money

      2. Mario Cantin

        No, you need to put up some of the money, usually 1% of the fund. Skin in the game….

        1. JimHirshfield


          1. LE

            1% (if that is the number I don’t know if it is) is quite a bit of leverage.Let’s look at it this way. You have a chance to invest $990,000 of OPM if you only put up $10,000. That’s a pretty good deal especially if you get 20% of the returns. So defacto it’s OPM. Dejure (to others points) sure your money is on the line.To my other point if these numbers are true that is not much skin in the game assuming the 10k isn’t a significant amount of your assets.

        2. pointsnfigures

          I would never invest in any fund where the founders didn’t have their own money at risk side by side with mine.

          1. LE

            I agree. However I am reminded of a fictional person running a carnival ride telling you that his own kid is on the ride. As if that is supposed to make it safe enough for your child. All people have different risk profiles (let’s call it). [1]My point is doesn’t it matter how important that money (on the line) is to the particular person?If it’s a minor part of their assets and it’s not going to change them that much or impact if lost then how much does it really matter? I think loss of reputation plays a large enough factor that it might even be more important (for some investors).[1] My wife isn’t as safety conscious as I am. She takes the new kitten outside (which has been declawed) and lets it mill about with her back turned. I had to plead with her to even put a collar on the kitten in case it ran out of the house. My inlaws let the young kids play at their pool w/o really watching them. I could never do that. I’d be watching 100% of the time.

  11. kevando

    So like, not the default subreddits?

    1. JimHirshfield


  12. creative group

    CONTRIBUTORS:The source of your well water (information) will determine the quality of it. (Our quote)

  13. Vendita Auto

    Good on him, “The essence of the independent mind lies not in what it thinks, but in how it thinks.”

  14. sigmaalgebra

    Yes, it appears that the venture capital (VC) partners (1) understand with total clarity that they need to find exceptional startups of which there are fewer than 100, maybe fewer than 10, each decade but (2) look for those exceptional startups by matching with patterns from successful startups of the past.I don’t want to be too cruel here, but, really,Play with their API’s and go deep in forums, even if there’s no investment in sight. Mess around in the corners of the at least 90% very much the wrong stuff. Part of the reason is a big problem in both computer science and practical computing: Neither of those has the crucial seeds or prerequisites of the bright future of computing. E.g., the people building WWI airplanes didn’t have any of the crucial seeds of the future of airplanes, for an example, were hopelessly behind at understanding radar, jet engines, the Navier-Stokes equations and Reynolds number, thick wings, and much more.With their empirical patterns, venture partners are able to see some significant patterns that escape me, what is in common with Microsoft, Intel, Cisco, Google, and Facebook?Or, at one point, better to look at GaAlAs heterojunctions — Bell Labs knew this fact in crystal clear terms quite early on and were 100% rock solidly correct.By analogy, it seems that early in WWII venture partners would have looked at better ways to win the war by extrapolating from WWI artillery technology. So, they would have missed radar, the proximity fuse, aluminum airplane skins, single wings that were thicker, superchargers, sonar, oxygen for the crews, bomb sights, code breaking, jet engines, and the atomic bomb. All of those did have solid patterns, but they were mostly from applied math and physics and not from just empirical patterns from successful weapons of the past.For this VC emphasis on empirical patterns from successful startups of the past, in the words of an *Indiana Jones* movie, “They are digging in the wrong place.”. Thankfully for US national security, the US DoD, already in the 1930s, found that approach just laughably bad. Thankfully for the US economy, similarly for the US NSF. Thankfully for US health care, similarly for the US NIH.All but late stage venture capital needs to work effectively with work that is new. Very strong, wide parts of the US are really good at that, e.g., NSF, NIH, DARPA, ONR, NRL, Department of Energy, the leading research labs, the leading research universities, the leading peer reviewed journals of original research, and more.VCs? For 99 44/100%, they do not have academic qualifications regarded crucial for research, cannot read the recent, relevant, STEM field research literature, are not qualified to teach college or graduate school advanced courses in STEM fields, are not qualified for tenure track positions in STEM fields, are not qualified to serve as editors in chief, editors, or even reviewers of good peer-reviewed journals in STEM fields, have not published peer-reviewed original research in a STEM field, are not qualified for a problem sponsor position at NSF or NIH, and have done little to no advanced, technical work in a STEM field. So, such VCs can’t evaluate or appraise new STEM field technical work.

  15. DJL

    Great insight and wonderfully written! Technical progress is filtering useful ideas from the fringe and then applying capital and hard work to get them into the middle. As an owner, it is always refreshing when a VC team can go deep into your problem space.

  16. OurielOhayon

    Jonathan was already messing around at the fringe at appsfire (which was great) :)great post. really

  17. Adam Haney

    Very interesting, I feel like so many VCs follow the herd when it comes to what they’ll invest in. Obvious investments tend to be efficiently priced, whereas new things, things that don’t make sense on paper but make sense to you because you’re keeping track of things that haven’t made it into mainstream yet seem to be the most interesting.

  18. LE

    AVC is part of my fringe time.

  19. Richard

    The finge is only the fringe in hindsight.

  20. Brian Allman

    I think with insight like this so early in a career we can expect to see his name and impact in the years to come.

  21. bmathes

    This is very important, and necessary. But not sufficient.You also need partners willing to entertain ideas from the fringe. Human nature and incentives push back. Ideas that come from a new, weird source are more uncertain, and therefore less investable, to the partner. A deal that the partner did not bring in is not as good, politically w/in the firm, than a deal the partner did bring in.You, as a Junior VC, also need to both sell up and have partners willing to listen.

  22. Ole Jakob Thorsen

    I get it and agree, but we should also be aware of the fact that there is more than one fringe, and if you wander of fringe to fringe you will be very disconnected or far away from tomorrows trend or mainstay.

  23. pointsnfigures

    was thinking about this the other day. How do you know a deal is going to be something or not? If it’s in a large niche, or in mainstream, it’s an automatic pass. Otherwise, it’s a “me too” investment. I couldn’t find the right words to describe it. Messing around on the fringe is a good one.Some deals are so totally obvious. That’s neither good or bad-if it’s obvious to only you it probably means you have a unique insight. The evolution of the company will prove your insight right or wrong in many cases. If everyone else sees it, they you should pass because the odds are really good there are a bunch of companies pursuing the same brass ring.

    1. awaldstein

      And some just are gut.When you invest and I know you have in food businesses you are stepping into a different world where almost none of this matters unless you think you can get to warp distribution without third party distributors that will bleed your margin till you are gasping for solvency 🙂

      1. pointsnfigures

        At seed, gut plays a very big part in it. Food is tough. I have done 2 food deals. One is going very well, one kinda tough. I won’t ever do food again because I just don’t have any personal value add. I love to watch food for trends though.

  24. William Mougayar

    “when does he know to stop?”You move to another thing & you maintain a ratio you can live with, for these fringe activities.

  25. Jonathan Libov

    Well to be honest, this fringe idea is something that took me much of my two years to fully realize. I wrote this up so that others might learn that lesson faster than I did.

  26. LE

    how much of your time is spent wandering around different playgrounds, and when does he know to stop?15% is the correct amount of time.

  27. Twain Twain

    These parts were informative, thanks:”Would-be A.I. assistants that need permissions for all the data that Apple and Google largely get without asking — this is a challenge for venture capitalists.”Especially now that Google has released Crowdsource which means its AI assistant will be even more accurate than smaller competitors:*…”Which is not to say that there won’t be a few valuable new bot or virtual reality companies, but rather that they’re really additive interfaces to current internet paradigms (I failed to appreciate this for a while), and so opportunities will be fewer and further between.”There’s a huge difference in product engineering between shipping additive or incremental tech and shipping paradigm-shifting tech.For founders and investors, alike, it’s the difference between merely collecting low-hanging fruit (e.g. building on top of an existing platform as the bots do) or planting and fertilizing the seeds that may grow into their own platforms and become the Amazon forest, an ecosystem that supports many species so they too can flourish.

  28. Wendy

    For me – playing in the fringe is chasing an interesting idea without specifically having something that I’m looking to find. Although, it helps me find connections between other things that are on my mind.

  29. iggyfanlo

    JonathanI’d love to connect. What’s your next thing? We’re working on something pretty intriguing (and definitely fringe) at I’d love your feedback and perhaps more

  30. pointsnfigures

    How much bourbon did it take?

  31. fredwilson

    I think it’s great that you are passing it on. But in many ways it is the lessons you learn the hard way that stick the most

  32. Lawrence Brass

    ” … but also talk to people who are experts in their field or following modes of thought that contravene mainstream thinking.”Mainstream is usually contaminated, dull and over marketed . One can’t distinguish leaders from followers. It is refreshing that rebels are still valued by people as you and companies like USV.Good flight Jonathan!

  33. Richard

    imagine if bankers of the industrial revolution took the kind of credit that VCs take today.

  34. PhilipSugar

    I agree. I took 15 minutes today to make this helmet for a person that fell and hit their head in the tub (wife’s conditioner). He said he was fine but we made him get checked out and he was indeed fine. (it would not be funny otherwise)

  35. LE

    That’s funny!The fit and finish is pretty poor. But I guess to do otherwise will exceed the time allotted and make people question where your head is at. It’s almost like you have to make it crappy or you will feel guilty for spending the time for quality. By the way those lunchroom chairs yell “don’t spend a lot of time in this room”. (Prison surplus?)You are right to have made him get checked out as an undiagnosed hematoma in the head can be a big issue (from what I know) you always need to do a scan to see if there are any leaks.I am planning on packing the cat in a box closed up to look like an amazon delivery for when my stepdaughter gets home from a vacation with her father. I hope it doesn’t die from lack of oxygen before she opens it. [1][1] If he does we will just visit the animal shelter again … no big deal. The shots are a pain but having a kitten makes up for all of that which the wife takes care of anyway….Below: Kitten vs. Lobster…

  36. Lawrence Brass

    It is like a minecraft helmet. I see potential here. 😉

  37. PhilipSugar

    Those chairs came from corporate and are more than $500 each. That is not the lunch room that is the meeting space outside my office.

  38. LE

    Impossible for me to believe that those chairs are $500 each or even close. I went chair shopping (a manufacturer/importer is local) within the last year for office chairs and saw various chairs. I also bought waiting room chairs. Even at MSRP and given 40% discount hard to believe MSRP on those chairs is $500 or even $400. You sure about that? I am talking about the blue molded chairs with metal bases.

  39. panterosa,

    The things you teach yourself or learn by yourself ( I’m never sure how to best describe this) are the most interesting. They are hard because you have no teacher, you lead yourself in the dark, half the time backwards. Once you finish it is all very obvious how everything fit together and why.Not sure if @libovness:disqus agrees with me on this process, but he certainly agree it all seems obvious in retrospect. The most difficult part for me is to cohesively write a single question which was so subconscious I couldn’t voice it. This verges into @jerrycolonna “live the question” mantra – and would still like a name for this process other than just “explore”.

  40. Drew Meyers

    So consumers give you all their data, with the hope they make some money out of it? Is the ultimate goal a “socially conscious” consumer credit bureau?How are you thinking about educating consumers about what is being done with their data? Seems there are some strides that can be made in terms of which sites are selling to what other sites (and the $ at play). Bring clarity to what data is available about consumers on the black/grey market of data sales..(I’m guessing virtually everything).

  41. PhilipSugar


  42. iggyfanlo

    We aren’t collecting ANY DATA…. we are just the union reps for the consumers… The blueprint is already there… the largest ad blockers are collecting literally hundreds of millions of dollars (and growing VERY QUICKLY) and keeping it all for themselves… we believe that the consumers are the rightful owners of the data and as such should receive the lion’s share of the value…

  43. PhilipSugar

    JP Morgan did.