Economic Impact of Open Source on Small Business: A Case Study Mike Hendrickson Roger Magoulas Tim O'Reilly Published by O’Reilly Media Beijing ⋅ Cambridge ⋅ Farnham ⋅ Köln ⋅ Sebastopol ⋅ Tokyo Preface This report was inspired by a conversation between Tim O’Reilly and Hari Ravichandran of Endurance International Group (EIG) Hari remarked that his web hosting business had been enabled by open source software, and that he wished he could find more ways to give back to the communities that have made his success possible Tim suggested that a novel way to give back would be to work together on a study making clear just how much of a role open source software plays in the hosting industry, and by extension, in enabling the web presence of millions of small businesses Hari graciously agreed to provide data collected by EIG’s Bluehost division as the basis for this study What we have tried to here is to help broaden the discussion of open source software’s impact beyond "the usual suspects" like Red Hat, MySQL, and other pure-play open source companies, and instead to focus on how open source has been a direct enabler not only of Internet companies, but of any business that enjoys an Internet presence One of our sayings at O’Reilly is "Create More Value Than You Capture." Open source software is one of the great examples of that principle Rather than measuring the value captured by companies that provide open source software or services based on that software, this report tries to hint at how value is instead captured downstream by users of that software Open source developers are unsung heroes of our economy It is not always the business titans, who have managed to extract a significant fraction of the value they have created, whom we should celebrate as the wellspring of economic growth Those who have put their work into the commons, enriching the soil from which other businesses can grow, have an untold economic impact This study is an attempt to initiate a deeper conversation about hidden sources of value creation in our society In a time when traditional ways of doing business have resulted in economic chaos, as financial companies focused more on extracting value from the economy rather than on creating it, it is worthwhile to reflect on what open source software can teach us about how to build a world in which, as Eric Liu and Nick Hanauer say in their book Gardens of Democracy, "We all better when we all better." Chapter Executive Summary Because the Internet’s infrastructure is largely dependent on open source software and open protocols, it’s fair to say that open source has played a significant role in the economic impact of the Internet This impact is both direct—in the revenues of Internet companies themselves—and indirect, in the increased revenues attributable to businesses because of increased visibility, efficiencies, or commerce on the Internet But because open source software is not monetized directly by most of those benefiting from its use, it tends to disappear from economic analysis Web hosting shows how open source software impacts the economy: web hosting companies rely on open source software to provide the infrastructure required to generate monthly subscription fees, small and medium sized businesses (SMBs) get low cost access to the Web and the open source tools to build a web presence—extending access to customers and transactions for existing companies, and creating a new breed of company that could not exist without the Web In this report, we use a unique data set, as well as survey data, provided by the Bluehost division of Endurance International Group, a web hosting company that provides hosting services to over million customers, most of them small and medium sized businesses There are areas where we believe Bluehost’s data provides a reasonable proxy for the SMB hosting market The Bluehost data is used to analyze the actual software used by their customers, estimating the cost savings provided by open source software We project the total economic impact of open source software as an enabler of the web hosting market as well as the economic impact on the SMBs who make up its customers Bluehost offer customers a web infrastructure and programming tools rich with open source components like PowerDNS, Linux, Apache, and MySQL, and open source programming languages like PHP, JavaScript, Perl, Python, and Ruby Bluehost also offers open source tools like Content Management Systems (CMSs), dominated by WordPress, and e-commerce tools like osCommerce, Magento, and PrestaShop While offering open source software via web hosting and domain name registration is a multi-billion dollar business, we show how open source generates even greater economic impact by creating a platform fostering the increased success of the small and medium sized businesses that make up the majority of web hosting customers Here is some of what we learned from the Bluehost data and follow-on research: 60% of web hosting usage is by SMBs, 71% if you include non-profits Only 22% of hosted sites are for personal use 75% of customers build their own site using simple site-builder tools Another 6% have it built by a family member Only 13% use professional web developers The majority of the websites are informational; only 14% have an online store (including nonprofits that take donations online) Nonetheless, nearly 20% of businesses in the survey say they derive more than 50% of their revenue from their website The majority of hosted businesses are very small Only 15% have revenues in excess of $50,000/year, yet collectively, we believe these businesses represent a trillion dollars of economic output PayPal is the dominant payment mechanism, outstripping direct credit card payments by nearly 2:1 WordPress is a far more important open source product than most people give it credit for In the SMB hosting market, it is as widely used as MySQL and PHP, far ahead of Joomla and Drupal, the other leading CMSs Languages commonly used by high-tech startups, such as Ruby and Python, have little usage in the SMB hosting market, which is dominated by PHP for server-side scripting and JavaScript for client-side scripting Open source hosting alternatives have at least a 2:1 cost advantage relative to proprietary solutions Chapter Introduction By most traditional measures, pure-play open source companies appear to have had a relatively modest economic impact Red Hat, the largest open source software company, had revenues of $1.13 billion for the fiscal year ending February 29, 2012, and a market capitalization of $10.8 billion as of June 2012 MySQL was sold for $1 billion to Sun Microsystems in 2008 After the sale of Sun to Oracle, MySQL has estimated revenues in the range of $171 million, and MySQL ecosystem revenue is expected to grow at a CAGR of 40% to reach $664 million by 2015 [1] There are many smaller companies that directly monetize open source software programs, but none have achieved the scale of either of these market leaders Yet there is no question that open source software has had a profound economic impact that belies these numbers Companies like Google, Amazon, Facebook, and Twitter are built on open source platforms All of them run on Linux, make heavy use of open source databases and languages, and contribute to the open source communities they rely on They have built a proprietary layer on top of the open source stack, analogous to the way Microsoft builds applications on top of Windows, or Apple creates operating systems and applications for its various devices from a common platform Even Apple bases many of its platforms and products on open source software and offers many Open Source languages, databases and other tools for developers And the economic impact is clear: Google had 2011 revenues of $37.9 billion and market capitalization of $189 billion Facebook had 2011 revenues of $3.8 billion and market cap of $82 billion, and Amazon had 2011 revenues of $51 billion and a market cap of $103 billion It is difficult, however, to apportion the value of the Open Source databases, programming languages, tools and Web software infrastructure in creating the success of companies like Google, Amazon and Facebook, as they all mix Open Source components with proprietary software Fortunately, we know another market where we can more directly tie value to Open Source software: the Internet Service Providers (ISPs), domain name registrars and web hosting companies that rely heavily on Open Source components to provide access to web services for their customers According to IBIS Research, the total size of the Internet service provider market in 2011 was $79 billion [2] ISPs typically provide both a data pipe and a set of basic Internet services, including email and web access They may also provide web hosting, and with it, domain name registration services It is difficult to allocate the value provided to customers of an ISP between "the dumb pipe", or access to free (but not open source) services such as Google, Facebook, YouTube, Twitter, and millions of websites, versus use of open source email and web tools, but there’s no question that without the existence of the free and open source software of the Internet, this huge and profitable market would not exist In the case of the domain registry market, the allocation of value to open source software is simple Domain name registration is essentially a service business based on the open source Berkeley Internet Name Daemon, or BIND, which (along with a few related programs, like PowerDNS) provides the Internet’s Domain Naming System, or DNS It is fairly reasonable to ascribe the entire size of this market to open source software Many domain name registrars also offer web hosting services based on Linux, Apache, MySQL, and a variety of other open source tools We estimate the domain registration market revenues as roughly $5 billion in 2011, based on market leader GoDaddy’s last reported revenue (before going private), and on estimated market share for the top providers, including GoDaddy, Endurance International Group (EI), Enom, Network Solutions and 1&1 These direct Internet businesses are just the tip of the iceberg According to a recent McKinsey study, Internet-related businesses now represent approximately 3.4% of GDP in 13 countries they studied (including the US), and 21% of GDP growth over the past five years in mature economies [3] The report also notes that 75% of the Internet’s economic impact comes from tradtional business; that SMBs get a 10% productivity boost from internet usage; and, that SMB’s with heavy investment in web technologies grow twice as fast as those with little or no web presence Because the Internet’s infrastructure is largely dependent on open source software and open protocols, it’s fair to say that open source has played a significant role in this economic boom But because open source software is not always tied directly to revenue, it tends to disappear from economic analysis In 1975, environmentalist Steve Baer described in his book Sunspots (Zomeworks) this kind of indirect economic impact by way of an analogy he referred to as the "Clothesline Paradox": If you take down your clothesline and buy an electric clothes dryer, the electric consumption of the nation rises slightly If you go in the other direction and remove the electric clothes dryer and install a clothesline, the consumption of electricity drops slightly, but there is no credit given anywhere on the charts and graphs to solar energy, which is now drying the clothes So too it is for the Web and for open source software Tim Berners-Lee put the Web into the public domain, the most open of all open source licenses; as a result, like the ecosystem services provided by the sun, the open source software that makes the web possible disappears from our accounting (In an interesting example of the Clothesline Paradox, it’s worth noting that while Internet users are often accused of enjoying free content, and being unwilling to pay for what they receive, they in fact typically pay substantial sums to their ISPs as a monthly subscription for access to these "free" services Simply by comparing the economics for the Cable companies and other ISPs for consumers who use Internet services to those who watch television, you can see that it is the ISP, not the consumer, who is getting the free ride On the television side, the Cable company must pay for content; on the Internet side, the Cable company gets content for free, with much of it created by the very customers who are paying them for access.) Web hosting shows how open source software software impacts the economy: web hosting companies rely on open source software to provide the infrastructure required to generate monhtly subscription fees, small and medium sized businesses (SMBs) get low cost access to the web and the open source tools to build a web presence - extending access to customers and transactions for existing companies, and creating a new breed of company that could not exist without the web Web hosting firms (including the web hosting component of Internet Service Providers) effectively provide access not only to a web server but also to domain name registration, programming languages, content management systems, e-commerce apps and other software components for which there are both proprietary and open source alternatives However, they overwhelmingly provide open source software as their first alternative According to Netcraft, 85% of all websites are hosted on open source web servers, including Apache, nginx, and Google Web Server Fortunately, we have access to a unique data set provided by Bluehost, a subsidiary of web hosting company Endurance International Group At the time this report was written, EIG was the second largest hosting company in the world They provide hosting services for over million accounts and million domains, most of them SMBs This data contains statistics and usage data from both Bluehost and other affiliated EIG companies Due to the size and the diversity of this data set, we believe that Bluehost’s data provides a reasonable proxy for the hosting market as a whole In this report, we use the Bluehost data to analyze the actual software used by their customers, estimating the cost savings provided by open source software We project the total economic impact of open source software as an enabler of the web hosting market as well as the economic impact on the SMBs who make up its primary customer base [1] Matthew [2] Internet [3] Internet Aslett 451 Research Service Providers in the US: Market Research Report, April 2012 Matters: The Net’s Sweeping Impact on Growth, Jobs and Prosperity, by Matthieu Pelissie du Rausas, James Manyika, Eric Hazan, Jacques Bughin, Michael Chui, and Remi Said; May 2011 This means that Julie spends approximately $1,020 per month to generate $4,000 to $5,000 in return Remember this is a person who enjoys what she is doing and started her business out of her passion for knitting Case Study of a SMB #2 Now let’s look at another SMB that has taken a little less expensive route (a full open source solution) in their costs to run a website In our second dicsussion with a SMB, the husband and wife team of Nicole and Brian Sidwar formed Castlechat.com in March of 2012 They are located outside of Orlando, Florida Brian and Nicole wanted to start their own business They were on vacation at Disney World in Orlando and began thinking about what kind of business they could start that would be somehow related to the Disney properties Their idea was to create a place for other small businesses to advertise They discovered that there were numerous other sites in the same space, but that most of the existing sites would only take one advertiser from each industry category Also, their competitors offered information only on their sites, so Brian and Nicole took it a step further and set up a fullservice travel agency to assist people in booking vacations They saw an opportunity to create a site that was everything about Disney They offer unbiased news, information, and reviews of the Disney properties: user-contributed content that will help customers to evaluate restaurants and other local services, or help them to research to plan a better vacation Neither Brian nor Nicole have any programming experience, yet they have been able to everything themselves Brian mentioned that when they want to install a plugin or something similar, they go on the Web to forums or websites and get detailed and oftentimes tailored instructions on how to accomplish what they need to Since they just started their business in March of 2012, their revenues are not something they want to share at this time However, they are averaging about 7,100 unique visitors per month and are on track to see that move to 10,000 per month soon They plan to provide either data reports from iPage analytics to their advertisers or when they get Google Analytics running they may supply that to their partners They have not looked at their geographical distribution of customers yet, but they are big fans of the know as much about your customers as possible school So they will be mining this information shortly They have had some struggles with Google Analytics working because of some meta-tag issues, but their hosting provider has bailed them out with a nice set of analytics They are obviously interested in seeing seasonal trends for their site and will start to target their marketing efforts based on what they find in the data According to the couple, the more they know, the more they can tailor their content to their specific customers Sounds like a sound plan Different type of start-up Nicole and Brian represent a slightly different type of SMB startup They are rolling up their sleeves and building their site component by component They are using the Web to learn enough to change and modify their site They are doing incremental changes, but in a thoughtful way that does not break other aspects of what they offer They have one order form at this time for their Castle Connection where you can sign up for a personalized package of more than 20 items that will provide you with a better vacation experience The form uses PayPal when you check out SMBs #2’s technology stack This SMB uses a Linux-based hosting option with three domains attached They have a few mailboxes installed for correspondence with their customers They have four MySQL instances installed and a bunch of PHP and JavaScript files on their site Brian mentioned that they pretty much everything with WordPress They will research a plugin extensively and find detailed explanations of how to install it before they move forward with a change like that Hosting base is Linux Databases are run by MySQL Plenty of PHP on their site and they only change pieces they know won’t break other things JavaScript on their site and they only change pieces they know won’t break other things WordPress is used to manage all their content and serve pages to their customers Online sales As mentioned earlier, there is a link to PayPal from a PHP form built with WordPress The form takes you to a PayPal checkout that is written in JavaScript This may change as their business grows and when the day comes that they sell merchandise as well as the packages they now offer CastleChat also has opportunities for third-party advertisers, and they expect this to be a growing area of their business Programming languages and tools Again, the site is mostly run by WordPress, but there is plenty of evidence of PHP and JavaScript Brian has an infectous attitude of can do, and we are sure he will what he needs to to run his site He is gung-ho about learning any programming he needs to evolve his website It really is refreshing to listen to small business entrepreneurs with so much passion for learning more so they can build and improve their business CMS As much of this report shows, WordPress is the key tool It has the lowest investment in time, money, and learning curve; it absolutely is the best ROI for a SMB Brian did report that he and Nicole evaluated other CMS solutions and WordPress was the most suitable because of ease of use, plugin availability, and public support on the Web, specifically in forums discussing installation and maintenance issues Marketing and sales Nicole and Brian have a pretty smart strategy for getting their start-up going They have saved lots of money with their choice in hosting services, maintenance costs, and add-ons, so they are investing in SEO services from their hosting provider Think about that While you are starting out, you need more customers, not whizbang technology SEO will help increase their reach and will build their customer base Brian certainly seems like he will learn the basics of SEO quickly and will be running the show soon They were not aware of the spend $1 and get $8 back in the Google AdWords promise, but I am sure they will evaluate that too The two owners plan to advertise on other sites that will generate more traffic and customers for themselves I am sure they will use the same analytical approach to finding the best sites to advertise with How you manage the costs? Costs? Wow, so here is a major difference in these two case studies Brian and Nicole have a threeyear plan that is less than $5 a month for hosting They use WordPress to deliver their content and PayPal to securely book sales So all this amounts to under around $44 a month This is so low they can afford to add the SEO package on now while they ramp up But the SEO package is not a fixed cost; it’s a variable cost each month that can be assigned to a marketing cost When you add that on to their package they still come out at $544 per month to run their SMB Let’s take a little deeper look at the economics of an open source approach versus a mix of open and proprietary Return on investment comparison Stack Cost Proprietary stack $1,020 Open source stack $44 Savings $976 (month) Scenario one: Open source customer does nothing with the savings and keeps more than $11k otherwise spent on a technology stack $976 * 12 = *$11,712* per-year Scenario two: Open source customer goes with SEO Package to increase customer acquisition and awarness of their company Monthly SEO cost = $500 Monthly savings = $976 - (monthly SEO cost) = $476 [ $476 + $500 * ( * 0.7 * 5) * (m) ] * 12 = $16,212 per-year The factor (2 * 0.7 * 5) ⇒ is from Google m = multiplier (effectiveness of SEO relative to Google), in this case m = 0.25 (we make a conservative assumption that SEO is one-fourth as effective Scenario three: Open source customer continues on with SEO Package, AND invests rest of savings into AdWords [ $476 * + $500 * ( * 0.7 * 5) * (m) ] * 12 = $49,896 per-year is the AdWords multiplier m = multiplier (effectiveness of SEO relative to Google), in this case m = 0.1 (given the AdWords campaigns in place, (we make a conservative assumption that SEO is one-tenth as effective) Google AdWords multiplier: "… for every $1 a business spends on AdWords, they receive $8 in profit through Google Search and AdWords" So with a back-of-the-envelope calculation you can see that scenario of investing in AdWords and SEO could generate two-thirds more revenue than scenario and more than four times as much revenue as scenario But of course, scenario is just saving money without reinvestment in growth But then again, that $11k may help you make more widgets or pay for other expenses that will help grow your business Chapter In Conclusion While it is widely recognized that major Internet sites are built on an open source platform, and that much of the Internet’s infrastructure is based on open source software, there has been little attention paid to the economic impact of open source In much the same way that solar energy and other "ecosystem services" are ignored in our economic accounting, open source and the open protocols of the Internet are taken for granted It is easy to discount the value of open source software in the success of sophisticated Internet applications such as Google or Facebook; after all, it is their proprietary services that allow them to derive outsized revenues and stock market valuations In the case of Internet Service Providers, which should properly be considered subscription resellers not just of bandwidth but also of services based on open source software, it is clear that a more significant part of the revenue should be attributed to that software But the argument is clearest in the case of the $5 billion/year domain name registry and domain hosting industry Here, the majority of what is being sold is subscription access to open source software Yes, there are bandwith and data center costs, but the service being sold is access to that software Most compellingly, though, the economic impact of open source software is seen "downstream," in the increased revenues of the Small and Medium Sized Businesses who, because of ready access to open source software platforms via web hosting businesses such as Bluehost, are able to advertise their business on the internet, and if they actually sell products online, to expand their customer base far beyond local markets, reaching customers around the world We estimate that Bluehost customers alone experience a $12.4 billion dollar economic impact on their business Given that SMBs are widely thought to generate as much as 50% of GDP, the productivity gains to the economy as a whole that can be attributed to open source software are significant The most important open source programs contributing to this expansion of opportunity for small businesses include Linux, Apache, MySQL, PHP, JavaScript, and Wordpress The developers of these open source projects and the communities that support them are truly unsung heroes of the economy! Chapter Appendix AdWords Impact Model Assumptions used to build the model: Total ad spending by SMBs ⇒ average of these two $5.4 billion in 2010, according to this article [14] from BIA/Kelsey $2,300 * # of SMBs in 2011 This survey estimates online ad spend at $2,300 per SMB in 2011 Share of ad spending on Google AdWords ~29% This is a product of the numbers below By 2010, 37% of online ad spending will go to paid search [15] Google owns 80% of ad spend on paid search [16] 37% * 80% = 29.6% Google multiplier = [17] We conservatively estimate that for every $1 a business spends on AdWords, it receives $8 in revenue through Google Search and AdWords Thus, to derive the economic value received by advertisers, we multiply our AdWords revenue on Google.com search results in 2011—what advertisers spent—by Google owns 80% of ad spend on paid search [18] Model '(Total online ad spending by SMBs) $5.4 billion * (share of ad spending on Google AdWords) ⇒ 37% (online search) * 80% (Google share) * (this is the Google multiplier: $1 of ad spending ⇒ $8 revenue) = Economic impact of Google AdWords = $12.8 billion US Census Data for SMBs in 2010 [19]: 5.3 million We looked at a number of sources to estimate the share of SMBs who are off-line The most creditable, up-to-date source estimated that 30% of SMBs were offline [20] SMBs offline = 5,300,000 * 30% = 1,590,000 SMBs online = 5,300,000 * 70% = 3,710,000 (# of SMBs without a website ) 1,590,000 * ( EconImp_Goog_SMB / # of SMBs with a website ) ( $12.8 billion / 3,710,000 ) = (AdWords economic impact left on the table) = $5.5 billion This is an estimate, because we are assuming that EconImp_Goog_SMB came solely from SMBs with websites (You still can use AdWords even if you don’t have a website, but you probably need to be able to optimize your landing pages to maximize conversion and get to the 8x multiple from Google.) Open Source Cost Differential Model The assumptions, methodology and equations used to build the Open Source Cost Differential model Assumptions used to build the model Amazon EC2 pricing provides a reasonable proxy for the difference between Open Source and commercial software infrastructure, i.e., operating systems and databases Amazon is equally adept at negotiating with their Open Source and commercial providers We chose specific pricing options that reflect typical choices for SMB customers, knowing that there is no typical SMBs and that their online needs vary widely: One-year term “Medium Utilization Reserved Instances” - providing a reserved instance and balancing upfront costs with lower hourly costs for a yr term — Average of Small (Default) and Medium pricing within the “Medium Utilization” tier The price differential between open source and commercial instances drops as usage/volume rises; these changes have a small effect on the model To calculate the cost differential, we allocate upfront costs by month and assume hourly fees for 18 hours of usage per day in a month EC2 pricing includes the hardware, system administration, network, HVAC, real estate and other costs required to run a network connected server instance Database pricing MySQL is included with the Linux option pricing — Three combinations of usage and price options are offered for SQL Server SQL Server Express - free for up to 10 Gb SQL Server (Web) - for work loads typical of web sites running a CMS SQL Server (Full) - full access loads and function typical of transaction systems like ecommerce and finance apps Ecommerce systems require full database access CMS systems require web database access — The mix of Bluehost business customers use of ecommerce apps, CMS apps and simple web presence represents a reasonable proxy for the US SMB market: 7.4% have an ecommerce system (all have a web presence, many have CMS) 70% use a CMS to manage content, but no ecommerce system 23% have a web presence only, i.e., no ecommerce or CMS instances — Many SMBs generate low data volumes, we assume: 50% of SMBs with an ecommerce have small enough data volumes to use free SQL Server Express 50% of SMBs with a CMS have small enough data volumes to use free SQL Server Express Our model combines the usage patterns based on the Bluehost SMB business user data with the price differentials between the various options to determine an overall open source / commercial price differential Adjusted ecommerce share (adj_ecomm) 7.4% of SMBs w/ ecommerce * 50% of SMBs too big for free comm’l database: 3.7% Adjusted CMS share (adj_cms) 70% of SMBs w/ CMS * 50% of SMBs too big for free comm’l database: 35% Web Presence (web_pres) 100% of SMB customers less adjusted ecommerce share (3.7%) less adjusted CMS share (35%): 61% General price differential formula price components upfront price per month (upfront_price_mo)= upfront_price / 12 hourly rate per month (hr_price_mo) = hourly_rate * 18 hrs/day * 30 days upfront_share = upfront_price_mo / (upfront_price_mo + hr_price_mo) price = upfront_price_mo * upfront_share + hr_price_mo * (1 - upfront_share) price differential = + ( (avg(windows_price) - avg(linux_price) / avg(linux_price) Full access database price differential (fulldb_diff): 9.83 Web access database price differential (webdb_diff): 2.15 No database price differential (nodb_diff): 1.57 Formula (adj_ecomm * fulldb_diff) + (adj_cms * webdb_diff) + (web_pres * nodb_diff) = (2.7% * 9.83) + (35% * 2.15) + (61% * 1.57) = 2.05 [14] BIA/Kelsey [15] eMarketer [16] ADV Media [17] Google [18] ADV Media [19] www2.census.gov/ces/bds/firm/bds_f_sz_release.csv [20] Parallels, May 2012 Chapter Programming Language Trends At O’Reilly Media, we have a data mart containing six years of job posting data from SimplyHired.com, a job posting site that aggregates most of the online job postings in the US Below, we take a look at job posting trends involving computer languages over the past three years This data provides context for the Bluehost data You will notice that in each of these charts there is a light green area plotted behind the bars Those green areas show the peaks and valleys of all job postings, not just technology jobs or postings for the specific language This way you get a sense if the whole US economy is trending up or down and does the language mirror that trend Note that the scale of these charts is not the absolute number of job postings, but is normalized relative to the number in January 2009, which is set at 100 A value of 200 thus means that there has been a 100% increase in the number of postings since January 2009 Trends JavaScript seems to be a close proxy to what is happening in the overall market There are few monthly instances where JavaScript is counter to the overall market March 2009 is an easy anomaly to spot Pay attention to the scale as the other languages are not quite as big, so the charts may look similar, but scale is important The next chart is for Perl There was a slowing in job postings starting in 2007 which is not evident in this view As you can see, Perl is still a widely used language, partially because of its alleged duct-tape capabilities If we were able to show job data back to the early to mid 2000s, Perl would have been at the top or near the top for Job advertisements PHP is on the same scale as Perl but has twenty weeks higher than 200 (so does JavaScript) on the X axis whereas Perl had two PHP is dominant in the Bluehost data partially due to the fact that so many of the popular web tools are written in PHP, and those tools tend to drive language adoption Also, PHP is a good general-purpose web programming language that lets you get data in and out of a MySQL table without a steep learning curve Python is measured on a higher scale than PHP and Perl, and is consistently above the 200 threshold Python’s job posting trend is very consistently going up at a average rate of 28% for the period reflected Python is reaching into other industries like life sciences, financial, and scientific settings were there is a significant need for mathematical equations This may be part of the growth in job advertisements for Python When you look at job postings and growth, Ruby surprised us with weeks higher than the 400 threshold for job postings per week None of the languages measured here even have one week above 400 postings So there is considerable action with Ruby and that could be attributed to Rails being a hot web framework for web developers There is probably a loose correlation with start-ups quickly prototyping with Rails and Ruby makes sense as the driving language Job advertisements Ruby does not show up at all in the Bluehost data C# as a language has moderate growth and only a few weeks above 200 I think this language needs to be compared against Java and other enterprise type languages We are not convinced that moderately experienced Web Programmers think of using C# when beginning a new project ASP is also a surprise that is is as low as reflected in the chart below It does follow the seasonal patterns of the overall job market, but has only three weeks above 200 Makes us speculate that PHP and JavaScript are continuing to eat into the ASP market Analysis When looking at these languages there are a couple of points that stand out First, they are not all targeted for Web Programming C# and to a lesser extent Python are general purpose languages Python is likely the most versatile in finding a home in many different settings, including life sciences, computer science, health IT, and other areas where either numerics or engineering requirements are strong C# is much more of a strong enterprise language for Microsoft platforms Although Ruby has some similar characteristics to Python, it is more closely related to PHP, JavaScript, and Perl for its use in web programming At least from a job posting perspective, Ruby is the clear leader From a language usage perspective, though, PHP and JavaScript are dominant among Bluehost website builders Many of the software packages that website owners install will use PHP, JavaScript and MySQL So while not every SMB will program with these tools, they may need to hire a consultant who does, or learn enough of the tool to make minimal edits Either way, these tools are a major part of the open source web development ecoysystem We have noticed in the past couple of years that JavaScript seems to be the most important language to learn if you want to program on the Web Libraries, tools, learnability, and flexibility are some of the factors helping this language become number one in use and importance The net-net is that the Bluehost data shows that programming language in the SMB market does not match the programming language job market as a whole Languages such as Ruby and Python, which are in vogue among high-growth technical startups, are hardly used at all The hosting market is dominated by PHP and JavaScript About the Authors Mike Hendrickson has held a variety of positions in the publishing industry including, Product Development Manager, Editor, Executive Editor, Editor-in-Chief, and Associate Publisher Two constants are that he has always enjoyed managing people and being involved with cutting-edge technologies At O'Reilly, he is the Publisher for the Open Tech eXchange division, where he is working to grow existing print market share while expanding into new online and in person markets Roger Magoulas is the director of market research at O'Reilly Media Magoulas runs a team that is building an open source analysis infrastucture and provides analysis services, including technology trend analysis, to business decision-makers at O'Reilly and beyond In previous incarnations, Magoulas designed and implemented data warehouse projects for organizations ranging from the San Francisco Opera to the Alberta Motor Club Tim O'Reilly is the founder and CEO of O'Reilly & Associates, thought by many to be the best computer book publisher in the world O'Reilly also publishes online through the O'Reilly Network (www.oreillynet.com) and hosts conferences on technology topics Tim is an activist for open source and open standards, and an opponent of software patents and other incursions of new intellectual property laws into the public domain Tim's long term vision for his company is to help change the world by capturing and transmitting the knowledge of innovators Economic Impact of Open Source on Small Business: A Case Study Mike Hendrickson Roger Magoulas Tim O'Reilly Editor Mac Slocum Revision History 2012-07-18 First release Copyright © 2012 O’Reilly books may be purchased for educational, business, or sales promotional use Online editions are also available for most titles (http://my.safaribooksonline.com) For more information, contact our corporate/institutional sales department: (800) 998-9938 or corporate@oreilly.com Nutshell Handbook, the Nutshell Handbook logo, and the O’Reilly logo are registered trademarks of O’Reilly Media, Inc Many of the designations used by manufacturers and sellers to distinguish their products are claimed as trademarks Where those designations appear in this book, and O’Reilly Media, Inc was aware of a trademark claim, the designations have been printed in caps or initial caps While every precaution has been taken in the preparation of this book, the publisher and authors assume no responsibility for errors or omissions, or for damages resulting from the use of the information contained herein O’Reilly Media 1005 Gravenstein Highway North Sebastopol, CA 95472 2012-07-18T08:43:02-07:00