big teddy
  • Mar
    18
    My three-fisted economic recovery package.
    by mike at 11:49am

    Alright I’m not Mark Cuban or Paul Graham I’m not a self made millionaire ready to invest in early stage startups.   What I do have is a bit of time and a few  good ideas that I’ve kind of left behind that might be able to be resurrected by a few people who may be unemployed with some time on their hands and an interest in starting a business.   These are two ideas that I started building on my own and then lost a bit of interest in.  I still think they’re viable ideas and you’ve got a headstart because a certain chunk of the early development is done.

    So in one fist I have A Different Engine my new Interactive Television application development company.   This is my bread and butter, but its a new services company with few clients right now so I might have some down time - and idle hands are the devil’s playground.  So I have two other ideas for more traditional intenet startups both these apps have working prototypes and one has been soft launched even (with no traffic ;-).

    My goal is to really get a couple of teams up and running and then just provide whatever guidance I can.

    The Ground Rules

    Note I’m stealing a bunch of these from PG and at least one from Mark Cuban.

    • The Goal is to build successful businesses NOT an exit.   There is the possibility of taking some intial funding and maybe down the road some sort of exit event, but that’s not the goal - that’s the journey.   This goal is to build a business that keeps a small, competent team paid and happy to come to work every day.  Get revenues, then profitability and grow organically.
    • Get Ramen Profitable.  If you haven’t ready Paul Graham’s 13 rules for startups do so now.
    • Start with 5 people per team.
    • Take no more than $100,000 in seed funding - if any.  You may ask “why take any?”  The answer is that funding allows you to get through some rough spots starting up.  Infrastructure is expensive and credit cards suck.  Now this may end up being only $5-25K for some servers but that’s fine.  The second thing some investment brings is exposure to that investor’s network.  Ask most what the most valuable thing they got from the Ycombinator program was and its likely going to be “Investor Day” when they get introduced to the founders network of potential investors and possible acquisition partners.  You may ask “why take so little?”  the issue here is that taking significant VC capital comes with a whole  world of expectations,  and requirements that I think are too high.   At the end of the day though this will be your product.  Make the decision that you think makes sense.
    • Everyone works on the product.  You’re either building the app, keeping it up or maintaining its content.  If you’re not doing something that affects what users see then you’re not that useful.  The Managment structure is pretty much flat.
    • Work agile - ideally on monthly iterations.  If you’re not familiar with the agile process yet, get there, its pretty straightforward and places a heavy emphasis on getting the product out.  Both these projects fit this model well.   And note I’m not an agile nazi, far from it, but I think working off a prioritized product backlog and getting release quality code out each month is a great way to work.
    • You can live anywhere.

    Equity.

    Note there may be no way this concept could work, but I haven’t thought of why not yet.

    So the way I see it I’d hope to take a small bit of equity, since I have a lot of early work done - this will be negotiated with the team but maybe 5%?  Then I was thinking on the most equitable way to distribute equity when teams may be in flux and the only thing you’re working for is equity.

    What I was thinking that for the first few months or years - until staff starts getting paid, that equity would be based on full months worked.

    The concept I have in my head is that at the end of every calendar month, everyone who worked the entire month gets one share of equity.  So say 5 team members start on May 1st, on May 30th everyone gets a single share out of a total of 5. End of second month everyone gets one more share but the   pie is out of 10.   If at the end of the second month someone leaves they walk with their two shares, if someone new comes on they, like everyone else gets a share at the end of the month and the pool again grows by 5.

    The thing is that while shares get diluted each month, the initial founders would still hold the same 20% as long as they stay with the company (though this could dilute with any investment)

    Anyone have a better way?  I’m not sure how to handle asking someone to leave (a firing) yet, but that would need to get worked out.

    The Apps

    wesayitiz.com

    The Team Needed

    • 2 Ruby on Rails developers
    • 1 designer
    • 2 Content Editors and Product Managers.

    The Concept

    Alright the name sucks, it came from “Its news because we say it is”.  The basic idea is that as successful digg, reddit and fark are I think there is still a space for a real meta news site. The idea is that news stories either come from users, or the content editors browsing the web and finding the best and most current stories.  The more I hear about digg or reddit or even hacker news being gamed to get on the front page the more I feel that a site that’s maintained by a few plugged in individuals may be the way to go.  While the community thing works, I think each of these sites skew a particular way, and I still end up at sfgate or nytimes for my “real” daily news.   Why not focus on the news and manage the content that way?

    Once stories are approved for the front page, the stories are ranked on the page by clicks.  More clicks, the higher on the page.  This is my user generated bit but could be removed.

    I’d love to see this site end up a bit like the NYTimes site, except the content is all externally linked.

    One thing I was going to do that was relatively unique was to be as public and transparent as possible with this site.   Why not let users know what kind of traffic its getting and the revenue being generated.  This whole thing is a bit of an experiment why not make it public?  More details on that

    Current Status

    The site is up, user authentication and it’s got an adminitrative interface, and even harvests RSS feeds in order to jumpstart finding content.   An Adsense account is already in place for revenue and there’s a comment system.   It could use a sharper design and once it gains traction I think you’re going to break out the content categories (content is categorized but there’s no display function behind it yet).   It still needs some QA and my test coverage is pretty sparse, I’d recommend the first thing done.

    Also its hosted on a slicehost chunk that it will likely need to move off of ASAP.

    HeyThere

    The Team Needed

    • 1-2 mobile application developers (got to be multi talented, iPhone and Android are the initial client platforms).
    • 2 Backend Apps developers - Java and Rails
    • 1 designer
    • 1 Product Manager

    The Concept

    HeyThere is not as developed as wesayitiz but it may have longer legs.   HeyThere is a location based messaging app, kind of like Loopt’s little brother. If I’m at a bar (back in my single days) and saw a cute girl at the end of the bar and wanted to find out who she is, how could i do this?

    HeyThere is essentially a hybrid IM and Chat client that broadcasts messages to the 20 other users closest to you.  No mapping or tracking, just broadcast chat/IM based on proximity - there are similar services that do this via a bluetooth broadcast mechanism.  There isn’t yet a revenue stream - but it couldn’t be hard, allow local vendors to send an occasional message for a fee and you’ve got something.

    Current Status

    The app is based on XMPP/Jabber and its been a while since I tabled this one so it may take some jumping from some hoops to get these running.

    I have 2 client prototypes in some kind of working state, the first runs on flash (the user inputs location) the second works on Android with GPS enabled.

    I have a backend working that is a plugin for the Openfire Jabber server which is Java based - I’ve also patched that server with something so I have a version of it in the repo (at this moment in time I can’t remember why).   I also have a few hacks in there that allows you to use any Jabber client for testing - you send location in text to a bot instead of using a built in GPS mechanism.   Scaling this piece may be a difficult task, I played with some caching mechanisms but they’re really hacked together. You may want to take a look at porting this work over to ejabberd.  Its Erlang based so its not as easy to find competent devs for this but it may scale better. Openfire can be clustered however so maybe just stick with that.
    All code is in a subversion repo and I can give access if you’re interested.

    This one is a bit of a beast to get running but I’d work with anyone interested on the different issues around it.

    The Rest

    So that’s it really,  two apps with working prototypes looking for the team that can take them to the next level.  They’re likely not commented enough but hey I’m here to answer questions.

    Think you have what it takes?  Then email me at recovery@bitdamaged.com

    Any questions or comments, then leave them underneath!

  • Mar
    09
    Today I quit my job
    by mike at 9:52am

    The Shivers

    Oh dear effing jebus what have i done?

    That’s the question rolling around in my head like the ball on a roulette wheel.  I just quit a great, highly paid job at one of the largest companies in the US, in likely the worst economy of my life in order to start an application development company.  How did I get here?

    This whole plan was likely hatched about 5 years ago when I, like everyone else there, got laid off by TechTV when it got purchased by Comcast and glommed into G4.  At that point I was the Lead Engineer of the Interactive Television Group.   At TechTV the ITV Group’s job was to support affiliate sales by creating apps to support cable companies new interactive and on-demand initiatives.   We were a small group and when we got let go we had plans to start an Interactive Television start-up, but time passed, we never got traction and  Interactive Television never really took off so I took a job with a company called Metatv - which was shortly bought by Comcast and Cox and turned into TVWorks (The whole Comcast thing is a bit hinky - I’m not sure I’m ever going to escape them completely).

    The move to MetaTV/TVWorks/Comcast was likely a very fortuitous event.  At TVWorks we were building the infrastructure to support the ETV/EBIF spec (which started life as a Metatv spec) for Comcast and Cox.  There I led the Client Applications group building the first two generations of ETV applications for Comcast.  Here I became and expert at this platform and a believer in its abilities, I liked my job but about a year and a half ago I wasn’t quite loving it.  As Comcast’s corporate infrastructure began to work its way into TVWorks processes I started to get a bit of an itch that I wasn’t meant to be working directly for such a huge corporate entity.  As long as I was at TVWorks I’ve been doing an evening M.B.A at SF State. The impetus for this was that since my time at TechTV I thought I would be heading more towards a managerial/entrepreneurial career path, I figured an engineer with an M.B.A is a pretty good swiss army knife of an entrepenuer.   I mean here I was in the cradle of the internet civilization and had yet to work at a true startup.   I got completely hooked meanwhile on Hacker News and the Y combinator funding model and even submitted a couple of startup plans to their funding cycles. The thing with these plans weren’t that they were impossible - but I’m not sure I was committed to them.

    So what to do?

    Leap and the net will appear

    So at some point mid-last year I figured I should go back to what I know best and that’s interactive television. So what exactly do I do?  The space is starting to settle technically, on EBIF and tru2way as the dominant technologies on the set-top.   There are still a few holes in the ETV technology chain that could be filled with the proper product offering so the first thought I had was to create a product company.  The upside of creating a product company in this space is that generally your “exit” can be bigger and its easier and generally these types of companies generally have revenue projections that are more attractive to venture capital investors.  There are a few problems however, first I don’t really want to be in infrastructure development, its not my domain experience and passion.  If I’m going to do this then I damn well was going to do something that I’m good at and love doing.  Its not just a passion and experience issue.  Second, developing and selling a product means building a significant business structure in order to support it.  You need some form of venture capital, time and committed partners in order to get this type of business running.   All three of these resources are hard to come by in the current economy and/or with a full time job.   Finally you lose a large chunk of equity and control.  The equity I’m okay with, the control not so much.  Its not that I’m a megalomaniac micromanager, in fact quite the opposite,  the issue I have with taking large chunks of venture capital is that you are now creating an organization that must be very significantly focused on creating value for the shareholders.

    Not that that’s a bad thing.

    But I’m not sure that’s the kind of corporate culture I want to create.  Maybe the business should focus on creating value for the people that work there?  I’d prefer something more intimate, a small dedicated team - with a heavy emphasis on their craft.

    Maybe I’m wearing rose colored glasses.

    So lets see I want to stay in interactive TV, I don’t want to do an infrastructure product and I’m hesitant to try to go after venture capital. This leaves two choices. The first would be an interactive TV application, or an application development services company.

    A services company has its downsides. Its a permanent hustle for new work, you can’t escape having people out there constantly selling and drumming up new work. It also tends to place more pressure on the services team when there is always an 800 pound gorilla of a client who wants everything better, faster and cheaper. But there are several upsides for starting an ITV development company right now.

    First as bad as the economy is, this year interactive television is becoming a reality.  The technology is being deployed on a bunch of cable operators.  EBIF in particular is a large driving force behind this, and it just so happens (wait… I’m trying to pat myself on the back here) I’m one of the foremost experts in developing apps for this platform.

    Second in down economies some companies are more apt to outsource development services - especially in an area where they are not going to have in house experience.

    Third is that this still leaves the door wide open to developing products in the future.  If you take the history of the web, many large application companies spun out of web properties creating tools that they needed and then productizing these tools for others.  Because the ITV field is so nascent I think there is going be a similar arc.

    I guess there it is.  A ITV application development company and service provider that will (hopefully) spin off some products.

    Introducing

    A Different Engine

    So this is the reason I left my job today.  A Different Engine aims to be an interactive television development company.   In the short term we will focus on consulting and EBIF application development. Longer term is really hard to map out right now.  The whole television landscape is going through a huge seachange right now in how television content is delivered and monetized.  I’m not sure how its going to shake out entirely but I can pretty much guarantee you this - there will be interactivity involved and we’ll be there to make it happen.

    Behind every great (or halfway decent) man…..

    One of the biggest reasons I can do this now is because I have the support of my wife and family.  April I couldn’t do this without you, you’re my rudder.  Thank you sweetie - I love you!

  • Feb
    12
    Rails 2.3.0 update Gotcha
    by mike at 10:24am

    So I’m working on a Rails app that’s been going swimmingly until last night.

    Messing around on my Mac laptop I ran

    gem update

    This ended up killing about 3 hours of my night and 20 minutes this morning when I finally figured out what happened.

    The symptom was that I kept loading my main page and getting an error that my ApplicationController was missing.

    uninitialized constant ApplicationController

    This was strange because I still had my application.rb file, I could include the file in my controllers with “require application.rb” and get things to work, but this obviously wasn’t right.

    I correctly thought I’d somehow hosed my gems which I spent about 2 hours mucking with last night. I realized I had Rails 2.3.0 which was new but couldn’t figure out what the heck was going on. Finally this morning I created a new rails skeleton to see if that would run out of the box. On doing so I realized that “application.rb” was now “application_controller.rb” in the skeleton app. A sensible change, but that meant I needed to rename my application.rb file to application_controller.rb. to get my rails app to work.

    doh!

  • Nov
    12
    ETV across the board in 2009
    by mike at 2:14pm

    Today both Comcast and TWC have claimed they’re going to have ETV clients in all homes by the end of 2009.  Interactive TV in the US?  Woot!

  • Oct
    30
    The Future of TV Advertising.
    by mike at 3:40pm

    Currently, The Death of the 30 second spot.

    So way back in the day, Advertisers pretty much owned the TV airwaves,  the 1.0 version of TV Advertising had brands sponsoring and even producing and programming the shows that aired on the networks, the networks were essentially “dumb pipes” carrying the advertisers programming. Then  came the great Quiz show scandals of the late 50s which changed the arbiters of on-air programming from the advertisers into the hands of the networks themselves. Of course nothing in life is free, the networks needed to get paid and so was born the 30 second spot on TV - TV Advertising 2.0.

    And its been great right?  Outside of a slight bump wth the advent of the remote controI  (changing channels without getting up!  Argh!) and an initial fear of the VCR, this medium this had pretty much a 40 year run with no problem.  Advertisers would churn out these little spots, and the masses would sit on their couch and consume and all was well.

    But in the late 90s a company called Teleworld Inc. was formed and they created a new home media device called the “TiVO”.   This box had a strange history - it wasn’t understood by those who didn’t own one, and absolutely adored by anyone who did.  Suddenly more astute advertisers began to worry a bit about their precious television ad spend.  People with these new Digital Video Recorders (DVR - sometimes “Personal” Video Recorder) were able to record shows, play them back and skip the ads!   Heck future versions would even include the ability to skip in 30 second chunks to make it a bit easier.  Despite the popularity with their owners fortunately these devices have been fairly slow to catch on, but now,virtually every new set top box going out to consumers has a built in DVR - and these are getting out quick as more people want to take advantage of their shiny new HD TV.   Combine the growing uptake of DVRs, the new ruling on Network DVR’s and the rise of On Demand programming and you begin to get into areas where the 30 second spot starts to become severely threatened.  Yikes! What to do?  Where should advertisers send their dollars?

    The Addressable Ad, The Canoe And You.

    While the ad unit would stay the same, the effectiveness could be multiplied if, say, the ads you saw (or tried to skip through) were more tailored towards our tastes right?  Well in the U.S. this ability isn’t far off. Already many cable operators are looking at ways to implement addressable ads across their networks.   Technically its not terribly hard, the biggest issue is primarily a bandwidth concern - you have to pump multiple ad spots in the same broadcast bandwidth.   But this concern is being addressed as cable companies begin deployments of Switched Video.    Now all they need is to know what the Set Top Box’s users watch and suddenly you have a pretty good profile of which ads may be most effective BAM! Tailored TV Advertising.   The thing is though, to really get an effective ad buy across a single demographic you need to purchase spots across all the cable networks, this means you need some sort of cooperation between the cable networks.

    Getting all the cable companies’s paddling in the same direction is the goal of the Canoe Project.  The Canoe project is a cross cable company initiative that should allow advertisers to buy ads from multiple cable networks from a single point of contact.  (If you want I have a quick primer on Cable Advertising).   The Canoe project has a few goals but one  is to expand the reach of cable advertising.  Even without addressable ads the Canoe project will allow advertisers to get increased value by targetting ads by geographic demographic across the different cable comapnies.   Or in laymans terms they can say I want this Ford Flex ad to go to affulent urban areas and this Ford F150 ad to go to the hicks in the sticks.  With the increased targetting of ads the job of the media buyer is going to get a whole lot more complicated.  Something Google may be looking to address (Sheer speculation here - but I know they have some designs on STB software).   The Canoe project has two other goals however as well.    The first is to implement systems to deal with addressable ads.  The second is to make the ads enhanced.

    Enhanced Ads And The Forward Lean.

    One of the biggest drivers of ETV and interactive television in general is the ability to add value to the traditional 30 second spot by actively engaging the user - getting them leaning forward from the couch and making a traditionally passive experience into something more engaging.  While the possibilities will become vast there are a couple of primary types of enhanced ads that are being pushed by CableLabs and Canoe.

    Request For Information (RFI)

    RFI ads are one of the more ubiquitous forms of enhanced advertising.  It allows the user to request additional materials to be mailed or emailed to the user.  Some of the cooler more advanced versions can also tie into a Digital Voice offering or something like Jajah to set up a call with a customer service rep.

    Telescoping Ads

    Telescoping here means ads that take you away from your linear programming (telescoping into the content).   There are essentially two flavors here, the first type simply telescopes into a VOD offering.   In this scenario the viewer can be watching an ad for the new Batman movie and then choose to view the trailer on demand.   The more advanced version would telescope into a more robust interactive offering (sometimes called a “microsite”) which in some cases may just be an enhanced VOD stream.   This way the user could click into a microsite for Dominos an instantly be able to order a pizza, or into a CBS branded space with links to VOD previews of other shows. The hardest part of enhancing ads is adding interactivity to a 30 second window in time, which is why telescoping units are the most attractive.

    The Future or “How can I buy Lauren’s cute shirt?”

    While enhancing and addressing ads can both make them more relevant and more engaging,  neither directly address ad skipping.  The real solution to this will end up being some form of in show advertising.   The next generation is already becoming the age of multitaskers, and they’re going to want more and more out of their television experience.   Adding an interactive layer is really the next innovation that will happen to television shows, enhanced overlays and telescoping will really add new ways for advertisers to attract users from within the shows themselves - timeshifting be damned!  Even with relatively basic interactivity this kind of in-show interaction can add new methods for increasing brand reach.   This could be enhaced by tieing the interactivity directly to the show content as well.  Imagine a popup during “The Hills” that says “want to know what Lauren got her shirt?  Click Here”.  This could lead to an advertising spot or even the ability to purchase the product directly.

    Similarly shows like Heros and Sweden’s The Truth About Marika, are turning regular television viewing into an immersive interactive experience.   For those who wish, they provide the opportunity for users to actually interact with the show on the web - with additional content and plot development going on outside of the normal television confines.  The Truth About Marika even took this a step further by creating an entire alternate reality game where users could actually get up and out and experience the content outside in the real world.   These types of fully interactive experiences are most likely the future of all types of content, and advertising in this new immersive medium will span interactive applications and viral media to a wholly ground up interactive experience - ads included.