Parasitic computing is a technique [...] that uses the legitimate function of computing hosts to perform some other computation.

From Gian David Perrone‘s Master’s Thesis [PDF]

According to a Nielsen report broadband-connected Americans spend 2.6B hours online every month. That’s about 300 computing-years/month in the US alone while people are actually in front of the screen. The global internet population is 2B people! That is a lot of largely unused processing power.

There’s projects that use idle processors to solve difficult problems (SETI@Home, Rosetta@Home, SuperDonate, etc). Most of those projects are based on BOINC and require the installation of a client software. All BOINC projects have a combined all-time user-base of 2M users. That is 1/1000 of the internet population (probably less, given that BOINC runs on servers as well). For a mind-blowing list of volunteer computing articles click here.

I think we can do better.

Parasitic JavaScript and MapRejuice

This weekend we (@shiftb and @ryanmcgrath and me) built MapRejuice for the Node Knockout competition. I finally found some time to reflect on what we learned and where we’re at, so here it is.

MapRejuice is a distributed Map Reduce implementation in JavaScript. A website owner can insert a small piece of code into his website and his visitors will start cranking away at MapRejuice jobs. This computing power can be sold and the webmaster gets a cut. Goodbye advertisements? Not quite yet. There’s a few things to figure out first.

Distributed computing in JavaScript has been attempted numerous times before:

• http://jsdc.appspot.com/ Is a distributed reverse-hashing implementation that can be embedded into websites. More recently the author built http://distributed-pi.appspot.com/static/beta.html
• Andrew Collins uses the BBP algorithm to compute digits of pi in visitor’s browsers. The algorithm automatically verifies the digits.
• Nick Jenkins found a way to dynamically adjust computation speed in distributed JavaScript application based on processing power. I also stole the term “Parasitic JavaScript” from him.
• Collaborative Map Reduce in the Browser. A proof of concept Map Reduce implementation in Javascript.
• Misco, a MapReduce framework for mobile devices. This is actually implemented in python but the mobile aspect makes it interesting.

The problems commonly associated with Parasitic Javascript are

• Security: The worker scripts could modify the host site. Also the data and algorithms used in the computation are visible to the clients.
• Performance: A few years ago JavaScript implementations were  still rather slow.
• Latency: Most distributed computing applications are not l because the input data is so large. The latency from sending this data to workers would kill any large-data computation.
• Reliability: Workers in this kind of setup are not trustworthy and fail often (when the user navigates away form the host site)
• Restrictiveness: The in-browser JavaScript runtime is not sufficient for complex calculations. There is no binary protocol and data cannot be encrypted. Data cannot be streamed. Data cannot be requested and sent back across different domains.

Using the upcoming WebWorkers standard makes it impossible for workers to modify the host site and the performance of modern JavaScript engines with JIT compilers is not too bad. Some of the restrictiveness may be solved by technological advances, but let’s not count on that.

What now?

So what have we got? We can run jobs on large numbers of unreliable, non-trustworthy workers with slow transfer speeds in a restricted, sandboxed environment. Now we’re looking for a class of problems that are well suited for that scenario.

Most Map Reduce jobs seem to run some sort of analysis on terabytes of data with chunk sizes of 64MB. That’s not going to work in our case. But what if the data sizes were generally small, or the data was publicly available, or the data could be randomly generated? How about a distributed count of words in Wikipedia? Or a Monte Carlo simulation for investment portfolios (thanks Tobias!)? Or an analysis of a person’s interests and habits on the internet?

As I said, there’s a few things to figure out and the technology may not quite there yet/widespread. But that doesn’t mean that it’s not worth a shot. And before we know it the next wave of technologies such as Google’s native client will be available and we can run BOINC in everybody’s browsers :)

We’d appreciate any input on the topic since we’re new to distributed computing.

Update 1: Voting at Node Knockout has ended and we came in second in the innovation category and seventh overall. Not too bad :) Thanks for all the support and congratulations to the winners!

Update 2: We got covered in the MIT Technology Review (errr, the blog at least). We also got a shout-out on the changelog (starts at 8.30min)

Share More

No related posts.

Property Heatmaps For Places

I want to have heatmaps that I can lay over places to reflect certain properties such as how nice the place is, how expensive it is or how many girls there are around. I want to be able to animate this data like weather-forecasts and to detect patterns and to predict how it’s going to change.

I don’t have much on this yet, but check out this mockup:

Coming up after the jump:

• Credit Card Meets Online Game
• Mass-Customization of Physical Products + Gaming: Facilitating Shopping Addiction
• Lab Notebook for the iPad

Credit Card Meets Online Game

This is my favorite. It’s sort-of a spin-off of the next idea and is best used in conjunction with it.

Imagine you have a credit card that will reward you not only with air-miles or bonus points but also with other virtual points/achievements/levels/community props. Basically, you’re combining the addiction of shopping with the addiction of online games!

You can even have challenges like hitting a certain series of stores in one day or spending more than 100$ at least once every day. I want the system to say “Spend 100$ today to reach level 34 and get the super rare achievement XYZ”.

You could do this together with a bank and collect wicked interest rates or sell it to existing credit cards as a value added service that will make them more money :)

Mass-Customization of Physical Products + Gaming: Facilitating Shopping Addiction

People are talking about how mass-customization of products is an incredibly lucrative market.

So let’s make a website where you can buy customized things. Now let’s also make it into a game where you can earn points and achievements etc. that allow you to customize your product in more interesting ways. People should be able to tell by the size of your shades that you’re on level 27 or killed some super monster or whatever.

This general concept doesn’t translate 1-to-1 in every market. Here’s two suggestions of how to implement the idea.

• Luxury, rich people’s markets: This is jewelery, accessories, maybe fashion etc. According to my benchmark movie “Up In The Air”, rich people pride themselves in having many air miles. Fine, let them spend air miles in product customization! Or even better, give them a credit card, that is connected to your store and allows them to gain “levels” and what not based on their shopping habits (see previous idea). Their wives can then buy expensive products that will let everybody know that her husband has bough three items worth more than 10k with a credit card in the last 4 weeks ;)
• Livestyle markets: Create an iPhone or Facebook or other “modern” game (most likely some roleplaying-thing) and allow them to pimp out their boring real-life selves with in-game achievements

I don’t have time to write my story around these links, so I’ll just add a list of links here that are sort of related:

• Web 3.D presentation by Joseph Flaherty
• Webkinz lets kids buy stuffed animals from their online game
• Shapeways sells personalized products made by 3d printers and such
• IIRC Chris Anderson’s latest book is about the paradigm shift from bits to atoms or something

Lab Notebook for the iPad

This is hilarious. So I search for iPad laboratory notebook and find this software called iPad ELN (Electronic Lab Notebook). I nearly discarded the whole idea b/c of that but then I look at the screenshots. Turns out, iPad ELN is a desktop application and  its name actually is “iPad”. BOOOM! Lawsuit!

Anyhow, apparently when scientists do lab work they have to maintain a lab notebook that is subject to strict rules. That way you can’t alter your experiment results after the fact or rip out experiments that are not in line with a later conclusion. I guess the exact rules depend on the lab and industry. This seems to be a big pain point.

So let’s just put all that in an iPad app. Here’s a list of advantages:

• Connect to lab equipment automatically, no more manual recording
• Upload results to servers and working groups automatically
• Full-text searchable notes/data etc
• Audio/Video/Photo notes
• Integrated utility functions such as timers, audio/video triggers etc.
• Maybe integrated with some sort of controlling software depending on your industry
• In general, you can try to automate as much of the book-keeping process as possible.

I have no actual market data (a manual LN costs 9 bucks, though), but knowing the medical market and how the iPhone has been adapted in there, I guess this could be highly interesting. For reference, the self-proclaimed market leader in the ELN space, EKM,  has a product that is called EKM LABTrack that is so expensive that you have to email them for a quote. Also, there seems to be an actual ELN device called LabTrack and there seems to have been some sort of lawsuit.

All of this I gathered from two minutes of Googling and a discussion with Neuroscientists a year ago in Munich, so take it for what it’s worth. They wanted me to make an iPhone app but I found the iPhone to be insufficient for the task. Well, the iPad’s better. Go for it.

Business Idea of the Week Disclaimer: I just write these ideas down because it’s good training, I like them, don’t want to forget them and will most likely never have the time to actually implement them. If you know of a similar venture or are interested in the idea, I’d love to hear about that!

Share More

Related posts:

1. Business Idea Of The Week: Make Work More Like Games
2. Business Idea of the Week: Artificial Luxury Fashion Labels
3. Business Idea: Picture-Taking-Machines for Sights and Landmarks

Team DormDorm.com at the Launch Party: Justin W, Amir, Justin I, Mick, Igor, Chris and me (Karolis is not in picture but he was awesome, too)

I’m happy. It’s a fantastic day and I’m back on a plane to SF. I can’t believe how amazing the Startup Bus experience has been for me and everybody involved. These are my fresh, unorganized impressions thus far :)

Why was it awesome? Because people who do such things are awesome! Example: The team put together the fantastic website and pulled an all-nighter after B***** deleted the entire thing one day prior to launch. Everybody got along fantastically and in the end, the bus probably launched more friendships than startups.

We left SF at 6am on Tuesday. At 7am we got a prep talk by Phil McKinney at HP’s executive briefing center and executive breakfast. We pitched over 20 ideas and came together in teams on the drive to LA where we talked to strangers. (and some dude tried to run off with Igor’s notebook) I can’t remember most of the next 40 hours; I was running an experiment on sleep deprivation.

Apparently everybody was really concerned about that (good times, though) and we had lots of technical problems and wrote lots of nutty code and stayed at this place and that and Elias drank four cups of Texan truck stop sauce. In the end, though, it all came together. We had six brand new startups. We arrived at a massage parlor with beer, girls and Josh Bear.

The following two nights of Austin were nothing short of amazing and I’d love to have stayed longer. The teams pitched in front of a fantastic VC panel and some teams are already following up on the ideas (definitely including us; We are DormDorm.com – the Hotels.com for summer dorm rooms).

Sooooooo… nobody took us and our startups seriously. We can’t create a startup in 48 hours. It will only be worth 48(x size_of_team) hours of work and the businesses will by construction have massive (black) gates to entry (instead of, you know, barriers to entry). I’m down with that and I won’t feed the trolls.

But understand this: The bus was a win for everybody. The sponsors and winners got heaps of great exposure (this, that and the other), the Buspreneurs got all the stuff in the next paragraph for free and we might just have inspired other people to go out and do things.

So what about the Buspreneurs? They got everything! Friendships, business connections, a new startup to work on, a fantastic and mostly hilarious time, personal insights, a broadened horizon, new skills, you name it! There is not  single one of us who regrets this trip and once we got going, we received fantastic feedback.

Here’s what I learned from The Startup Bus.

• Randomly mixing technologies together will lead to disaster.
• In the end, it will be ok, though
• When all else fails, sleep.
• Mental and physical constraints make you work harder and be more creative or grumpy.
• 48 hours are longer than you think.
• When it comes to decisions you’re passionate about, don’t make a compromise if you think it’s a bad decision. Nut up or shut up. Nobody’s gonna be passionate about the compromise. (this is an actual lesson)
• Just do it.
• Don’t'cha ever forget! We’re on a bus muda******.

Finally, big thanks to Elias Bizannes, the crazy aussie nutcase who came up with this in the first place.

Elias, you and the Startup Bus are the stuff that legends are made of.

Share More

Related posts:

1. 6 Reasons To Have A Blogger On Your Startup Team

I was a World of Warcraft (WOW) addict. The fishing activity is a tragic example of what that encompasses: Click, Wait, Click on the Bobber, Click, Wait, Click on the Bobber, … Some people do this for hours.

Why do people click and wait for hours? Because after a while a fanfare sounds and a banner comes up: “Achievement Earned: Super Fisherman 200“. Other people see that, too. That is why WOW works.

Fishing is routine work, but in contrast to other routine work, people enjoy fishing in WOW (sort of). Work can be like that. Hit the jump for some ideas I’ve had on this topic.

Example User Story

You’re Ted. You don’t have any skills to speak of. At your old job, you slacked around for 8 hours before going home to enthusiastically play (aka do routine work) WOW until late at night.

It’s your first day at your new company, the Bor Ingsurance company.

As you sit down to write your first email, a fanfare sounds that everyone in the office can hear! The screen shows a badge “Typewriter Level 1, 20 strokes/min”. You don’t understand what’s going on until a couple of hours later, you achieve “Typewriter Level 2, 25 strokes/min”. You click on the banner and see on your profile that you only need a couple of more typewriter levels until you can either get a free dessert in the cafeteria or a designated “TypeWriter” ball-pen.

Your first day at Bor Ingsurance was the last day you ever used only two fingers to type. The End.

Other examples:

• In a retail store, whoever makes the highest individual sale on a day gets an achievement. Whoever does that for three days in a row is on a “streak” etc. The possibilities are endless.
• Coming in early on your birthday will give you an unexpected round of applause + the “Super Discipline” ;)
• Answering your officemate’s phone while he’s out might give you an unexpected companionship point.
• Contributing to a project that you’re not staffed on will give you some achievement.

The Core Issue Is Employee Motivation

In large companies with many employees the inhibiting factor in efficiency seems to be employee motivation. From my experience at government agencies, large corporations or even fast food restaurants, the workers just don’t seem to want to go any faster or be any more productive/helpful.

The above story was exaggerated, but only slightly. What reason does a lowly employee currently have to learn touch-typing? None. Just give him one and he’ll learn it.

Writing down the entire concept is out of the scope of this blog post, but here are some ideas:

• Companies can compete against one another. Great for employee-employer bonding and identification.
• Achievements are given to teams. This is similar to WOW, where you sometimes need up to 40 people to receive an achievement.
• In large companies, a designated Game Master controls the “game”. He has t make sure, that nobody goes without achievements.
• You earn points that you can use to receive special perks and treats.
• You have an intranet (or maybe cross-company) portal. (comp. Enterprise 2.0)
• There can be seasonal achievements and awards.
• Most achievements are not known before somebody gets them.
  

One core consideration has to be that every feedback from this system is meant to be positive. No employee should be exposed in front of others (e.g. because he can’t touch-type). If possible, team-achievements are preferred.

Implementation: Workflow Engines

Implementing the idea in a running, possibly very large company is not possible. There are far too many different programs involved that would have to be interfaced with. Luckily, a window is opening: the adoption of workflow engines.

In the last couple of years (largely thanks to the rise of ITIL and ISO 20000), companies have introduced methodical process management. This means modeling their internal business processes (not computer processes) in a format such as BPMN and managing them according to best practices.

BPMN modeled processes can (theoretically, with some tweaking) be executed in a workflow engine. Workflow engines guide the people involved in the processes through the steps necessary to complete them. Instead of writing an email to somebody, a staffer enters his message into a designated field in the workflow engine. It takes care of sending the information to the other participants in the process.

I will not go into detail about the pros and cons of workflow engines. In short, they facilitate controlling and improve reliability and efficiency while taking away the worker’s freedom. One day workers will do routine work on thin clients that can only run the workflow engine so that the employees can’t procrastinate (on Facebook) anymore. (In fact, accessing Facebook, flash games et al. could be a reward for points in that scenario)

The switch to workflow engines is the window we need to wait for. It is the ideal time to put an employee motivation program in place.

Final Considerations

In a way, my proposition is similar to the mostly forgotten “employee of the month”, but automated. Games have shown us how people can be motivated to do boring things. The WOW example has spawned a number of very successful imitators, such as Microsoft’s X-Box Live, Steam’s Achievements or the myriad of Facebook games with their levels and badges.

Implementing a similar system in corporate settings yields the following benefits:

• Employees will work more efficiently because they are motivated
• The atmosphere in the company will improve
• There will be no added marginal costs (except for the possible game master and the software)
• Employees can be conditioned to learn new skills (such as touch typing or using a new software)

All of this sounds great, but make no mistake: It is also difficult.

There is a reason that Blizzard employs a team of psychologists. Before this idea can be sold to companies, studies proving it’s success need to be conducted. This is probably something that can best be done in a university setting. In fact, this idea was the reason I accepted (and later turned down again) a graduate position in Neurocognitive Psychology last year. Given how obsessed I used to be with this idea, I’m glad I finally wrote it down :)

What do you guys think? I think this idea has a lot of potential and the payoff – if done correctly – is huge. Just writing this post makes me want to start. Now.

Business Idea of the Week Disclaimer: I just write these ideas down because it’s good training, I like them, don’t want to forget them and will most likely never have the time to actually implement them. If you know of a similar venture or are interested in the idea, I’d love to hear about that!

Share More

Related posts:

1. Business Idea of the Week: Artificial Luxury Fashion Labels
2. First Business Idea of the Week: The Social Media Alarm Clock
3. Business Ideas Brain Dump

Hacker News is flawed.

I started reading HN a few weeks ago and already think it’s conceptually flawed. A lot of extremely smart people frequent HN and the community’s potential is being wasted.

This is not a rant. I appreciate everything that News.YC does for the startup/hacker community and will continue reading it. But I might not engage in active discussions. Hit the jump for my reasons.

Update: I was just made aware of the archive.org pages for HN. Takes you back.

Update 2: Some new points (“this link has expired” et al.) have come up. Check out the discussion over at HN for details.

I realize that most of the constraints on HN were consciously put into place but as the sit grows, the features have to adapt.

1. It takes only 1 hour for a post to disappear from the “New” page

On a side-note, it takes only around 10 hours (or less, at the end of the day) for a post to disappear completely from the page, unless it has made the (one of the) front-pages. This leaves too much room for chance in what posts get the most exposure.

2. The Point system sets the wrong incentives

Most people seem to be in it for the points. They comment more on posts with good exposure, because there the chance of gaining more points is higher. A commenter with an average of 5+ points per comment would not comment on an interesting article that is currently on the last page and about to get lost.

Also, some people seem to continue threads that are so far down the list that they cannot be “revived” in more popular threads to gain more points.

3. There is no way to search

This is probably a very conscious decision, but still I feel like potential is being lost here. If I am interested in a specific topic, I would rather look for it on HN because the successful articles are usually smart and insightful.

You can search on Google with site:news.ycombinator.com but the result listing is much less informative (no points, no date, no poster)

4. Using HN as a discussion forum doesn’t work

It is possible to post questions to HN that are then supposed to be discussed. The problem is, that these posts rarely gain a lot of points although I think they are much more interesting than a link to an article. According to my very scientific study I have found incidental evidence that about 3% of all posts making the front-page don’t contain a link.

Here’s an outrageous theory: Maybe URL links get more points on average, because people make their friends vote them up on HN to get exposure? (just like at Digg!)

A lot of potential is being wasted here. Or am I just overestimating the quality of the HN community?

5. There is no way to revive old discussions

If I come across an interesting discussion that is far down the list (or even out of it) I may respond to it, but the original author will probably never see it and neither will the rest of the world.

Conclusion

I admit that I’m a HN newbie and that I have not tried very hard to engage in an actual discussion there. This is because to me, participating feels a little too much like Digg (albeit a Digg with smart people) and there are too many “Thanks, I have nothing to say” posts.

Keen observers might notice that this is also a personal story about a discussion (that I care about a lot) that got quite a few good comments but few points and was thus dropped. So take everything I say with a grain of salt.

Some of the above issues have relatively easy fixes. Some might require rethinking. Consider dropping the points system in favor of a discussion-based rating system or introducing a new “upcoming” page with a random sampler of new posts.

I’m interested in the rationales that have made HN what it is today.

Share More

Related posts:

1. 6 Reasons To Have A Blogger On Your Startup Team

I was doomed from day one. I settled for the wordpress.com blog instead of a self-hosted wordpress installation because it seemed much easier and I was lazy. I took the blue pill (which is the one that doesn’t free you from the matrix, btw). Hit the jump for a quick rundown of what I did.

Reasons For Switching

Among others and in no particular order:

• Crappy Social media integration. WordPress.com does not allow custom JavaScript anywhere on the site. Thus, adding social media buttons (other than the supported ones) only works with static anchor images which direct the users away from my site.
• No feedburner integration
• Hilarious pricing (if you want to host your blog on more than one domain, you pay 10$ per domain)
• Insufficient analytics features. The wordpress.om analytics show nothing expect for the number of hits.
• No custom plugins.
• No custom theme modifications (except for css) and small selection of themes

How to Switch

I won’t go into detail on how to perform the switch since other bloggers have done so well and extensively. Check out these posts to get you started:

• How to Migrate your Blog from WordPress.com to a Personal Domain
• Migrating from WordPress.com to WordPress.org
• Six Easy Steps DIY: Move Your WordPress.com Blog
• How To – Move From WordPress.com To WordPress.org
• Migrating from wordpress.com to wordpress.org or self-hosting

Things to Consider

• Comments by wordpress.com users will no longer include their Avatar or link back to their blog.
• You will probably waste hours of time configuring your new blog
• Blog name/description/settings are all lost
• All wordpress.com features that have you include a shortcode (such as [digg=someurl] or [tweetmeme] or

  [/sourcecode]
  

  ) will initially not work.

• Embedded youtube videos won’t work.

Things to Take Care Of

• After you’re done, make sure to have your old wordpress.com (yourname.wordpress.com) link to your new one
• Install a nice theme (hundreds of themes are available)
• Configure your sidebar just like it was
• Install the Webmaster Tools Verification plugin and reconfigure webmaster tools settings for your new blog
• Activate Akismet and enter your wordpress.com API key
• Install WP-Stats and enter your wordpress.com API key
• Create a FeedBurner account and install and configure the FD Feedburner Plugin
• I also installed these other fantastic plugins:
  • AddToAny share button
  • After the Deadline (as seen on wordpress.com)
  • Broken Link Checker (as seen on wordpress.com)
  • Digg Digg interactive social media buttons
  • Feed Stats
  • Google XML Sitemaps
  • Greg’s SEO
  • Syntax Highlighter Evolved (as seen on wordpress.com; so you don’t have to chance your code blocks)
  • W3 Total Cache
  • Wickett Twitter Widget (as seen on wordpress.com)
  • Yet Another Related Posts Plugin (YARPP)
  • Zemanta

If you have any questions about any of the above steps, please let me know.

All of this being said, it’s been a rough day but I’m happy that I moved and I’m very content with the result. Now this page can grow :)

Share More

No related posts.

When I was done, I noticed how I had just described a typical blogger. Hit the jump for 6 superpower reasons why we should start founding companies with bloggers.

Disclaimer: The reasons not to hire bloggers will not be discussed in this article. I can not be held responsible for damages incurred by hiring bloggers.

The links below each entry can help you acquire the discussed superpowers!

1. Bloggers Are Interested In Everything

Most bloggers are looking for stories that might be of interest to their readers and are thus interested in a large number of topics. When working on a startup with limited resources every founder should expect to be involved with everything.

• 20 Blog Topics To Get You Unstuck
• 100 Blog Topics To Write About

2. Bloggers Know Their Internet (And It Knows Them)

The typical day for a blogger is sitting around in their underpants and surfing the internet. For a startup, this is great. Bloggers have seen everything on the net and know the dynamics of a good viral story. Also, some bloggers might even be influential or well-known in the blogosphere which further helps your cause.

• Blogging 101 – What Is It And Why Should I Care
• I Got The Scent Of Going Viral On Social Media – Now I Am A Bit Confused
• The Internet – A Small Town in Cyberspace

3. Bloggers Know How To Write (Viral Content)

Every successful blogger can write. But why do startups need writers? In my case (German founders, all-English communication), when writing any kind of public communication such as business plans, proposals, blog entries, documentation etc., we ran into problems. Having a blogger as copywriter would definitely help.

• How To Make “Viral” Content!
• Tutorial: How to Write a Viral-ready Article in 2 Hours Flat
• 9 Ways To Make Your Writing More Compelling

4. Bloggers Can Work On Their Own

One defining quality for great company founders seems to be the ability to work independently, to see problems and to prioritize. Bloggers have these abilities, otherwise they wouldn’t be able to run their blog on their own.

• Ten Stupid Mistakes Made By Newly Self-Employed
• 10 Reasons Why You Should Never Get A Job
• 10 Myths About Self-Employment

5. Bloggers Can Do Design

At this point it should be clear that bloggers are all-rounders. One of the most important aspects of this is design; As with other products and services, a blog can only be successful if its looks appeal to the readers. For startups who usually don’t need a full-time designer in their team this is incredibly useful!

• How to Blog Design Style Guide
• 43 Design Mistakes You Should Avoid

6. Bloggers Are Customer Oriented

I started blogging a couple of weeks ago. Although it started out as a writing exercise/notebook, I already find myself thinking: will this be relevant to anyone? Do my readers want to read this at all? Pro-bloggers do this to the extreme. As soon as a topic comes up (let’s say it’s trending on twitter), there are already articles on it. Needless to say, some IT startups are not customer oriented at all!

• Writing customer-oriented online copy: the ‘inverted pyramid
• “Customer Orientation”

Share More

Related posts:

1. 5 (Possibly Fixable) Reasons Why Y-Combinator’s Hacker News Is Flawed
2. Why The Startup Bus Was Awesome
3. Business Idea Of The Week: Make Work More Like Games

There seems to be a discussion going on about how newspapers can maintain their business model in the connected world we live in. They are suddenly competing in a global market in which the consumers spend little more than one minute reading news and expect that to be free. A couple of weeks ago Google Chief Economist Hal Varian held an interesting presentation at UC Berkeley. According to him, the decline in newspaper revenue is caused primarily because reading news has shifted from a leisure-time activity to a working-time activity. Thus, the goal must be to reinstate news as a leisure time activity. (Let’s call that the “iPad-approach”)

I disagree. The internet enabled news to become a working-time activity, but it is not responsible for the decline in leisure-time newspaper reading. In fact, according to Varian, the number of per capita newspaper subscriptions has seen a steady decline since the 50s that has only ever so slightly accelerated in the last couple of years.

So here’s a (difficult) question: How can newspaper publishers embrace their working-time-ness?

The iPad Approach Cannot Save An Entire Industry

Before we tackle the big question, let’s look at the industry’s plan to move reading news back to the living room. (Facts as-remembered from Varian’s presentation).

Apparently newspaper subscribers used to spend an average of 25 minutes per day reading their paper. Working-time online news readers spend 38 minutes a month reading online news. The largest proportion of newspaper ad revenue used to come from specialized sections such as finance or cars. These sections  receive the fewest online clicks because dedicated car- or finance-related sites have higher search engine ratings. Conversely, the sections that wouldn’t have a lot of ads in a real paper receive the most online visitors (disasters, war, politics).

Since advertising dollars are paid according to click-through rates and exposure (in time and quantity), it is obvious that simply putting articles up online supported by ads isn’t sustainable at all… If only people were to read news online as much as they used to read it offline, the publishers might be able to break even. This is why they try to make news a leisure time activity again.

Apple has shown their intention to do so by developing the iPad, which is ideal for leisure-time news consumption. Unfortunately, there are several issues that are not addressed by those plans:

• The main reasons why people don’t read newspapers in their free-time are TV and videogames (not a lack of leisure-time news-reading devices).
• There are a number of things people can’t do with electronic devices:
  • Put their cereal bowl on it
  • Throw it in a trashcan
  • Find it in a subway car and read it (and then throw it in a trashcan)
  • You get the point.
• Publishers will still be in a global market. In other words: Every single publisher has to compete with the NY Times and other giants. The internet does not need 1000 editors rephrasing the exact same press release and there will definitely be casualties.
• Only because they are reading news in the living room, people will not be more willing to pay for it. You can read Gawker publications or The Huffington Post on your iPad in the living room for free.

Anyhow, if you want to be one of the lucky publishers being saved by the iPad approach, you should do the following:

1. Get rid of your print copies (according to Varian that cuts the total cost by 50%).
2. Leverage the power of the internet and multimedia (or: make reading news more like interviewing a group of experts)
3. Specialize and use your resources. A company that has world famous columnists, Pulitzer Prize winning editors and an international network of reports does not need a “Dining and Wine” or “Easy does it for Zippy: Greyhound finds a home” section. (Found on the Sun-Times homepage) Maybe hey should even get rid of Reviews of any kind because there are dedicated websites for this, as well.
4. Don’t assume people spend more than one or two minutes a day reading your site. Make it easy. Give the masses what they need to know in order to have a rough understanding of current events.

IMHO, regarding the iPad approach, some publishers will get very rich and all others will struggle and maybe die. Let’s see what they can do about it …

How Can Newspaper Publishers Embrace Their Working-Time-ness?

Honestly, I don’t even know if it is feasible, but here are some ideas on the topic:

• Tailoring the advertisements for instant gratification (due to time constraints at work) or business related needs would be a good first step.
• If people have only 70 seconds to read their news, it has to be presented accordingly. That is barely enough time to read a whole article! How about a news feed with extended headlines that is tailored to you, your company and some crowd-sourced rating mechanism?
• I know of Germany companies that provide “education” services to large corporations by e.g. showing employees possibly relevant encyclopedic entries. Corporations are willing to pay for that sort of thing.
• Reading the newspaper broadens your horizon and educated employees with a broad horizon are of interest to all companies, IMOH (even the employees that do routine work).

I wanted to publish this post already, but I’ll keep adding thoughts as they come ;)

This Is What’s Going To Happen (Very IMHO)

In conclusion, I think this is what’s going to happen:

• People will read news in their living room more. (but not 20 times more than now)
• It will be free, ad-supported and running on devices similar to the iPad.
• A lot of newspapers will die or be bought by larger players. There will be layoffs throughout the entire industry. (As evidenced by the existence of the NewspaperDeathWatch)
• The decline in print newspaper subscriptions will accelerate.
• Some newspapers may transition into a more specialized magazine format.
• Whoever figures out how to leverage the working-time-ness will win big.

Whatever happens, this is bad for the newspaper publishers, but not for the consumers. Bright writers will still have employment and the access to news will be easier than ever.

P.S.: Google will almost certainly release a gPad at some point.

Share More

Related posts:

1. Business Ideas Brain Dump

In Germany, we have a luxury bag manufacturer called George, Gina & Lucy. The company has built an elite image that implies decades of experience and a bag costs about 100$. What few people know, though, is that none of the people involved in founding that company were called George, Gina or Lucy. Instead, the company was founded in 2004 by a couple of MBA students. (also note how the company has no Wikipedia page. Coincidence? I think not.)

I am proposing to take this to a new level by producing a couple of “artificial” brands under one central management.

Design of the Brands

When building the brands I consider the following (in no particular order) to be useful and important:

• Each brand has to specialize on one type of product.
• The brand name must imply a rich heritage and sound vaguely like the name of a famous designer.
• The real designers are talented students.
• The website must not have an online shop! (seriously, how many luxury labels do you know that name prices on the site?)
• The website must have a “store locator”
• Initially the pieces are manufactured manually and sold for ridiculous prices on eBay.
• Market entry is best achieved in cooperation with local boutiques.
• Capital must be invested in advertising in high class fashion magazines

Central Management

I believe the key to the success is having a small management team and otherwise little or no staff. The following tasks can and must be outsourced to professional services providers:

• Human Resources
• Accounting
• Back office/Secretary
• Media Design
• Manufacturing

What remains are the following mission-critical tasks:

• Management
• Strategy
• Development of the Individual Brands according to the above guidelines

Challenges

The biggest challenge is growth. As soon as manufacturing a limited number of items for eBay and local boutiques doesn’t suffice anymore, it needs to be outsourced, too. For that, a suitable manufacturing country and a reliable partner in that country has to be found.

Some other more tangible challenges include assembling the management team, finding a great advertising agency and obviously hiring talented and new designers.

In conclusion I think that a project like this is definitely doable and highly promising. In fact, I have been on the verge of creating that exact venture last year but then got distracted by things that appeal to me more. (after all, this project is heavy on the management side and has little or no engineering appeal)

Business Idea of the Week Disclaimer: I just write these ideas down because it’s good training, I like them, don’t want to forget them and will most likely never have the time to actually implement them. If you know of a similar venture or are interested in the idea, I’d love to hear about that!

Share More

Related posts:

1. Business Idea Of The Week: Make Work More Like Games
2. First Business Idea of the Week: The Social Media Alarm Clock
3. Business Idea: Picture-Taking-Machines for Sights and Landmarks

Introducing WaveConnector for GWT 0.0.1

WaveConnector allows you to develop Google Wave gadgets and to test them locally in hosted mode! Head over to the project site at Google Code for more info.

The turnkey archive that is provided on the project site has everything you need to get started! It already contains the little project I made for the tutorial last week, so you can just downloaded the waveconnector-gwt-turnkey archive and start from there. There is no need to download anything else. Hit the jump for a really short How-To that’ll get you going in no time.

And now… a screenshot of our WaveListGadgetGWT running inside a hosted mode browser!

WaveListGadgetGWT running in GWT hosted mode!

Update: Apparently the old gwt-gadgets gadget linker for GWT does not support power features such as code splitting and resource bundles. This is not acceptable. I am currently looking into whether its possible to use the xs linker instead. In that case the communication with the gadget js api would have to be reimplemented (or maybe taken from gwt-gadgets).

Bummer.

Really Short How-To

Make sure to have the latest Google Plugin for Eclipse installed.

• Download the turnkey archive (you don’t have to extract the zip)
• Select File -> Import
  
• Select “Existing Projects into Workspace” from the import dialogue
• Select “From archive file” and choose the turnkey archive you downloaded earlier
• Click Finish. You now have three projects in your workspace.
  • GadgetTester is used to test your gadget in hosted mode
  • YourWaveGadgetDeployment is the project that compiles to the actual wave gadget
  • YourWaveGadget is the project that contains the GWT code for your gadget.
• To test the Gadget locally, run the GadgetTester. It will load YourWaveGadget in a hosted mode window
• To deploy, I suggest you just deploy YourWaveGadgetDeployment to the AppEngine. Otherwise you can compile it and upload the war/yourwavegadgetdeployment folder wherever you want.
• I have deployed the turnkey gadget to http://connected.latest.wavelistgadgetgwt.appspot.com/yourwavegadgetdeployment/com.thezukunft.wave.client.YourWaveGadgetDeployment.gadget.xml if you’re interested

Feel free to ask questions how all this works. Some might also be answered on the project page (including: “Why is this great?!”). I haven’t gotten around to really documenting things yet.

FAQ/Tips and Tricks and Random Notes

Taken from the project page.

• All code is released under the MIT License
• If your refactor stuff, make sure to adjust the module definitions. Eclipse doesn’t update them.
• The YourWaveGadgetDeployment project has AppEngine nature for easy deployment. This is not really required and you can remove the dependency.
• All events will run through the injected eventBus. Do not try to attach handlers to the Wave object. This is the core difference from the JS wave API.
• If you want to understand the dependency injection going on, note how the GinModules in the different projects inherit from one another.
• If you add classes to your main YourWaveGadget project that need to be injectable (like sub-widgets that need to be created by a provider), keep in mind to add them to GinModuleWaveGadget.java
• Use the Wave objects log function to log messages that are visible in the wave sandbox
• You can cut the compile time of your gadget (because you will still be deploying it often) by adding something like <set-property name=”user.agent” value=”safari” /> to your module xml. (replace safari with your browser)
• All gadgets make use of the DynamicHeight feature. If you want to adjust the Gadget’s height to fit the content fire a GadgetUpdateEvent on the eventBus.

That is all.

Acknowledgements

I just sort of pushed this out on my own, but really there were some other people have helped me.

• Most importantly, Hilbrand Bouwkamp and his superb cobogwave library for GWT on which WaveConnector immediately builds up on.
• Once again Alex Moffat over at Lombardi who first gave me the idea of creating Wave Gadgets in GWT.
• Everyone at Google for their superb tools ;)

What Now?

I should probably document things and I’m also planning to add some features. Most importantly I want a UI to manage the simulated wave’s participants and perhaps even the option to switch the “perspective” of the gadget to that of another user (maybe even side-by-side).

So although it’s not 100% done, I felt this might be a great share for some people and I think that this is the closest we can currently get to comfortably developing wave gadgets in GWT. Also, now you can take any GWT application (restrictions on rpc calls and such apply ;))  and make it into a wave gadget.

Please leave a comment if you liked this and tell your GWT coding friends :) Developing for wave is really fun!

Share More

Related posts:

1. A Wave Gadget + Robot using GWT 2.0 and Google App Engine, Part 1
2. An Editable Label for GWT with UiBinder and EventHandlers