Choosing Open Source Solutions

Part of my job is finding and testing open sourced solutions for already prominent commercially available software. The concept is simple: If it's open sourced, it can be customized, be platform independent, and it can be free. In the business world, this poses two key benefits. Having software that can be customized means fewer problems and more functionality. Getting it for free means lower cost for the services we provide to our customers, and having more money to spend on employees and infrastructure. As simple as this sounds, finding the right fit can be a laborious task of trial and error. Part of my job is minimizing the trial and error.

Finding the right open source product is just as important as the decision to use an open source product to begin with. In every business software environment there are a few common components. There is the commercial product we've all used for years. There are the two or three popular open source alternatives, and there is a list of migration headaches we all experience that eventually becomes the list of reasons why we should stick with what worked before. Occasionally there is a product that inspires us to stay in the fight. A classic example we can all relate to is a product most of us use all day, every day: the email client.

Transitioning from the industry standard email server and its client software has its pros and cons. The obvious benefit is cost. Purchasing server software can get very expensive very quickly. This cost usually grows as your company grows. The client software is usually purchased with a per user license that will also continue to grow in costs as your company grows. Or it's purchased by site license which is often only slightly less expensive. The greatest challenges faced with open sourced alternatives are functional dependence and data migration.

The majority of office software users in general have been using the same solution suite for several years now. Shortcut keystrokes have been memorized and feature dependency has become ingrained. This becomes a real challenge for open source solutions. Often, when new solutions have been introduced to market, deployment was met with end-user opposition. When the first prevalent open source email clients were released, a few were championed as the replacement cure-all for the mainstream standard. It wasn't long before users began to learn that features they had become accustomed to were either moved and renamed, or missing all together. The lesson learned was that the economic benefits of a software package are lost on a user who cant get past how much they dislike using it.

In choosing an open source solution, often there will be the commercially available product that meets all of your needs, and the open source products that will meet most of your needs. Occasionally there will be an open source product that either matches the functionality that the end user has become dependent on, or that adds a function that makes the loss of a feature acceptable. When our company made the decision to begin using the Zimbra email server and client, two of the deciding factors were its platform independence and its migration ease. The email client is available for a wide range of operating systems, including the commercial ones. While migrating emails, folders, calendars and contacts can get a little tricky, depending on which client you are importing them from, they've all made it so far. After that, migrating an account from one machine to another (regardless of the operating system) is an absolute breeze.

When you're evaluating open sourced software you have to remember that there will always be two perspectives you have to keep in mind. If you're considering it from an IT/Management perspective, if you look hard enough, there will often be a program that meets your platform requirements, is easy to install, and is free. Doing your homework also means evaluating them from the end-user perspective. It's easy to say "they're just going to have to deal with it..." but this more often than not leads to a decrease in productivity in the least, and at worse a full fledged mutiny. Few things are more frustrating than pitching a change and then having to go back to the drawing board when what seemed like a good idea fails and requires reverting back to something you decided was worth leaving in the first place. A little expansion on the evaluation will usually lead to much less trial and error, and ultimately a better fit in the long run.


Chase Crum is the IT Infrastructure Manager for Voicenation and a self-proclaimed Linux FANATIC.


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

But Open Source has no vendor on site...

PerfMonk's picture

This is common in big organisation to have a Microsoft tech or rep.
They always have the ear of the management, since they invite them to costly dinner, tour in California to visit installations, etc. In brief they have a lot of marketing money in hand... Alas,this is not the case for Open Source.

But the truth will prevail in the long term ...

I have the same job!

JohnP's picture

I have the same job. I'm constantly looking at FOSS solutions and psuedo-OSS solutions, you know, those companies who market themselves as FOSS, but only release limited versions as FOSS. Or companies highly dependent on FOSS under the covers, but so tightly integrated that swapping in a newer, minor version of, say, the MTA will break the total solution.

For my/our small business, we obviously strongly prefer FOSS solutions and use them internally. Sometimes there is no viable FOSS solution available for a given requirement, but at least my clients know it and can be confident they've done the research before signing a $5K-$200K or more license agreement. Often 90% of the needs can be met with FOSS, so 90% of the users don't need the commercial product. License costs are reduced in that way as well.

My business also uses Zimbra for the last 2 years and are mostly happy. No solution is perfect, so we're always evaluating other players, including the main commercial guys.

Sometimes the commercial version of a FOSS solution adds features that are critical to the client and they decide to spend the money. Usually that project cost is much less overall than the commercial product costs and it avoids all the other "infrastructure things" that the other provider requires too. It also usually avoids 15%/yr maintenance and forced upgrades to remain under support.

What many IT shops fail to understand is that deploying FOSS requires maintenance, patching, and all the things that commercial solutions require. There are headaches involved, but at least YOU GET TO CHOOSE when and how those things are handled AND you've usually avoided the cost of software licenses too. If a FOSS tool has an issue that you consider a show stopper, you aren't stuck waiting for 9+ months so the vendor can work it into their release schedule just after your annual maintenance is due either. You can take action and create a solution or work around, then provide it back to the community. Sorta like community service and great for advertising your support of a FOSS project.

Having an in-house trial for each product is essential to gaining internal business support, but upper management needs to be on-board to push it. Productivity is what software is about after all and just because something costs less doesn't automatically mean it is cheaper.

Choosing open source solutions

Anonymous's picture

For a certain employee who is unwilling to switch to Floss, he/she may not feel like he/she is having an immediate financial motivation to perform better by using FLOSS but would feel the immediate pain of the learning curve. It's not like it's his/her own money that's at stake here, right? It's JUST the company's money, right? It belongs to faceless/nameless shareholders. A way needs to be used to make what would be perceived like the future possible pain of retraining more of an immediate painful reality, like the loss of a promotion or a bonus.

Perhaps the correct title for this article could be "Migrating..

Chetan's picture

..from Proprietary software to Open source."

The classic example in the article is really good case of "migration" from software X to software Y.

Rather more important aspect of making "End Users" happy is tricky one Proprietary not necessarily provide intuitive experience often its the case of people getting use to a particular software interface and the "new" software may be difficult to getting used to.

and as far as "Choosing Open Source Solutions" for business is concern commercial open source is way to go! not true blue open source.

Couldn't agree more

ian's picture

As one of those end users who are " just going to have to deal with it...", I've had to enjoy a piece of image software that requires 3 different user names, one put onto the same page twice, and two different passwords to get one image. Strangely, we had identified this might cause a bit of distress prior to its introduction; it seems to have been overlooked ( I believe the main reason is this is one of those commercially sourced packages that requires the sacrifice of an arm or a leg to change anything from defaults).

I'd have put some of it down to a healthy paranoia, except one of the passwords has a (proven) lower limit of no more than 2 characters. Given that the package has a webpage frontend, if it was open source the solution would have taken a day to rough out if not implement. Instead, I'm wasting up to 1/2 an hour of every day repeating myself.

right on

apexwm's picture

As the others have posted already, having open source and GNU/Linux in the business today is a great advantage. I own my own company which was formed in 2001. We originally started using Linux and Microsoft products, but dropped Microsoft soon after because it would have put us out of business with the extremely high costs and maintenance involved. We've been using Linux 100%, and as such we've managed to operate very efficiently while keeping costs down. I've also migrated all of my main home computers to Linux. I became tired of the constant maintenance. It's one thing to be paid at work to constantly fix things, but I don't have time to fix Windows at home. Linux just works.

What a lot of people don't realize is the huge array of software available for the GNU/Linux distributions. Alternatives to common proprietary programs made by Adobe and Microsoft have equivalent programs that are free, open, and NOT bloated like the proprietary ones are. I've managed to replace many programs like Adobe Premiere (with Kino and OpenShot), M$ Office (with OpenOffice), Adobe Acrobat (with CUPS-PDF writer), and many many more.

Excellent post and I agree,

r0000t's picture

Excellent post and I agree, I'm glad to see a subject like this get some attention. If more companies would use this approach, it would push the OSS to the next level, which in the end, would only benefit....well, everyone!

The biggest hurtle I see with OSS in an enterprise level business is support. Companies who have a decent investment in a project, and choose open source, often like to have some sort of vendor supported backing to the software.

Open Source Solutions

GNUguy's picture

I'm really glad to see this subject is getting more attention. There have been two other blogs recently that discuss this. I don't have the links handy just now. Email me if you'd like them.

IMHO, this is the most important subject to address while marketing GNU/Linux (...once for Richard) and open source solutions.

Marketing issues as I see them are...

1. Cost. This has been thoroughly examined and proven as an advantage for FLOSS although, no doubt, there will always be those who will refuse to accept it.

2. Stability. This has been addressed by the community (depending on which distro you use and the target hardware).

3. Security This has been addressed. (See number 1 for disclaimer.)

4. Games. Games are an issue for the consumer market. The business community doesn't care about games. And if FLOSS makes more inroads to the business community, then hardware companies will pay even more attention to Linux and develop Linux compatible drivers.

If you're interested, contact Zonker [ ]. He and some other folks have organized the Linux Dairy Council. The group hopes to find a way to support the marketing effort for FLOSS. I think they might be interested in your findings.

Just my 27 cents. (Opinion adjusted for inflation.)