Free. It’s a powerful word, that word free is. It can make a five-year-old child light up (“free candy?”) and it can topple governments when expressed by the masses in a country deprived of it.
In technology, the word free almost immediately conjures up a notion of open source and the promise of no more license fees and a whole host of other things (free beer/free speech, etc. depending on where you fall on the licensing debate.)
In fact, in my job at Lucid Imagination working with customers on Apache Lucene and Solr, I often hear during pre-sales meetings, from the CIO/CTO on down to the engineers in the trenches, the lack of licensing cost as one of the top three reason why they were attracted to the open source search engine and other open source tools (quality and speed/scalability are the other two.)
Interestingly enough, once they are in production, free, while still on the list, is further down.
What’s moved up the list to stand alongside quality and scale? As the title of the post implies, it’s flexibility. Choosing quality open source solutions allowed them to remain flexible in light of the continuing economic pressures of the day. What does that flexibility mean in the day-to-day work of building out production-ready systems that meet requirements while staying on time and on budget?
While there are the obvious advantages like being able to open up the source and make changes and fix the inevitable bugs in all software, there are many other gains, both technical and business-oriented.
For starters, saving a cool million or more on licensing costs gives you a lot more flexibility to innovate on your core business needs. Since popular, high quality open source projects rapidly accelerate the commoditisation of the markets they serve, businesses no longer need to invest large budgets in baseline, mid-level and even some high-end functionality.
Moreover, I believe that this flexibility to innovate, to try new things, to experiment, to feel like one can contribute is what today’s IT worker, especially developers, wants from employment (once basic economic issues are addressed) and vice versa.
In fact, I’ve even seen cases where they’ll take less money to do it! (See RSAnimate’s You Tube video on what motivates people for some reasons why.)
As a corollary to the first point, no single software solution (proprietary or open) is great at all the things it claims to do. However, if your core foundation is based on open source, you can plug-in best-in-breed technologies (proprietary or open) instead of being stuck with an add-on module that is less than stellar just because it ships with the package or is just a few dollars more.
Naturally, with the built-in domain expertise in your organisation, you can also write your own and make it work seamlessly with your open source foundation. Thus, flexibility in sourcing solutions is also important.
Quality open source also lets organisations scale economically, which is often the ultimate in flexibility. In today’s world of data that begets more data in a never-ending upward spiral, organisations need solutions that scale not just technically, but also economically.
Sure, the big expensive proprietary database solution might be able to scale to meet your needs, but can you afford to pay for it at that scale? Furthermore, quality open source gives you support flexibility, ranging from community driven content to independent expert consultants to large companies.
I’ve seen questions answered and bugs fixed in a matter of minutes in open source communities, but I also know that the community model has gaps: you can’t run mission critical software on the hopes that the community will answer your question at 3 AM on a Saturday night (even if sometimes they do!)
Hence, most widely used open source projects have at least one company offering support, paying for developers to work on the project, providing training and other services. With both the community and the companies, you may not need them, but it’s good to know they are there.
Last, but not least, open source gives you purchasing flexibility. I first came across this phenomenon when working with a large Fortune 100 company investigating Apache Solr.
Up to that point, they were spending about $1 million per year on a large commercial search engine. Some of their developers, meanwhile, had come across Solr and started trying it out on their data - and it looked promising.
They took it to their architects and product managers and got buy-in for a proof of concept. Again it passed. So, they did a full bake-off. Apache Solr won, hands-down. At this point, they told the vendor the results.
In a last ditch effort, the vendor somehow was able to drop their price to less than $75,000 per year. Talk about a discount! In other words, open source inspires flexible pricing models for proprietary vendors. So, even if you don’t choose open source, you can still use its power to find out what the proprietary stuff is truly worth.
Oh, for the record, the company still chose Apache Solr.
Blog post by Grant Ingersoll, Project Management Committee (PMC) Chair Apache Lucene, Committer on Apache Lucene, Solr and Mahout as well as a PMC Member on the Mahout project.
Grant is co-founder of Lucid imagination, and works in the areas of search, machine learning, and natural language processing. He is the co-author of the upcoming "Taming Text", as well as several articles on search and machine learning. Grant is the Track Chair for Lucene, Mahout + Friends/Search at ApacheCon NA, 1-5 November 2010.