February 17, 2019

Lou's Views: Zend Philosophy: Finding Another Way - page 2

Reevaluating the nature of open source development

  • January 23, 2001
  • By Lou Grinzo

As Jagielski explained it, the Zend people did not just sit in a room and conjure up their best guesses about what the market needed or wanted, or which itch they wanted to scratch, to use one of the less appealing cliches from open source. Instead they took the radical step of going out and talking to companies that might be customers for PHP-based products and asked them not just what they liked about PHP or wanted in a new release, but what was keeping them from using PHP. I can tell you from experience that the distinction between these two questions is far more than semantic hair splitting. Understanding exactly why an individual or company isn't using a product, whether it's software, underwear, or Tupperware, is easily the best possible start to knocking down those barriers to entry and turning that interested tire-kicker into a happy, paying customer. As addiction counselors like to say, acknowledging you have a problem is the first step in overcoming it.

When Zend asked this question, they found that potential customer concerns focused on three areas: performance, security, and support. And that's what they, in turn, addressed with their products: Improving the performance of a PHP server by caching the compiled PHP scripts (Zend Cache), a way to allow companies to distribute binary-only versions of PHP scripts (Zend Encoder Unlimited), and more comprehensive support and pre-packaged PHP options (Zend SOS).

OK, you can argue that there's nothing remarkable here, and it's just another case of a company that's smart enough to do good market research and use it properly before it committing to a product line. Not all companies are that adept at such things, obviously (New Coke, anyone?), but it's not exactly earth shattering, either. And I won't even speculate about how many open source developers actually talk to their intended users before they develop a product. Still, given all the companies I speak with that do think they've figured out the perfect product plan without talking to anyone outside their company, it was refreshing to hear what Zend did.

The most interesting part of the conversation came when Jagielski detailed his view of how commercial companies fit into the open source picture. He said that he sees companies as being in a position to form a vital link in the open source food chain, to act as an on-going communications conduit between users and the developers. At a time when companies are struggling to prove that they "can work with the open source community" (read: they're looking for ways to make money with open source software without turning open source developers into an angry mob with pitchforks and torches), the best answer might be for them to do take their supposed strength (working with the customers) and apply it in this new role.

Think of the opportunities this could open up. Instead of merely taking existing open source software, possibly modifying it, and then figuring out how to turn it into money, companies can take a more enlightened approach that helps everyone, including themselves, to profit from open source. They can use their experience in customer relations to create the most vital link of all, the one that's missing in the vast majority of open source projects: the bridge between end users and developers. Done effectively, this could have a huge impact on the usefulness of major open source software projects. The developers get good feedback from the users on what's really wanted (and no, raw e-mail and web site postings are not a substitute for this, particularly for the more mainstream software packages), companies get something more marketable, and best of all, users get software that's better tailored to what they want.

Of course, this is all just theory, and we know how often beautiful theories are sullied, warped beyond recognition, or destroyed outright by ugly facts. How well Zend or any other company can execute this plan, and how receptive the developers of any given open source project are to outside input, will vary widely from one situation to another. But it's still perhaps the most interesting and promising answer to date to one of the questions we all get asked by people outside the open source world, "How do you make money on free software?"

Most Popular LinuxPlanet Stories