Slashdot Mirror
How Do You Sell Linux Software?
smelroy asks: "My software company recently released a Linux port of our business instant messaging product, but we are not sure of the best way to sell it, since to many people selling Linux software is an oxymoron. Users on each end of the computer knowledge spectrum associate any and everything Linux with free. Even when we tried to get people to beta test it for us they said, 'It runs on Linux so why isn't it free?' Another comment from a reporter in response to our telling him of our Linux release was 'So it is open source then right?' So my question to the Slashdot community is when is Linux going to be prevalent enough on the desktop that people will pay for applications and not always assume they are free? Better yet, where are the people who feel that way now?"
Hopefully never - when people use GNU/Linux they will realise that Free software is better for a whole host of reasons. Once they've had their eyes opened to this, then using closed-source proprietary software becomes a much less appealing proposition.
Blaming GW Bush for the Iraq war is like blaming Ronald McDonald for the poor quality of food.
The world could use another IM system, or in other words: The world could use one working open IM system. What do we have now? There are ICQ/AIM, MSN-Messenger and Yahoo-IM on the proprietary side and Jabber on the open source side. The commercial systems work, but they are not interoperable and you can't have your own private IM-system on a detached network (desirable because of reliability and security considerations). Jabber on the other hand has yet to be scaled to publically usable proportions and is far from being considered "acceptable for mission critical applications". While I would certainly like to see one unified system which meets the requirements of a public system and a company-wide closed system alike, I realize that there is a market for a controlled IM-system which is independant of an internet service. I doubt that the advertised system could be that system, but I wouldn't dismiss this business idea so fast. The window of opportunity is probably very small though and the manpower which Microsoft and AOL can put into this to catch up or jump in before you makes it look like quite a gamble.
Lots of people have bought Zeus too, even with open-source free alternatives like Apache.
My department currently has bought Real (now Helix) Server, SPSS Web Server, VMWare, and a course content delivery system (mainly for the course content), all running on Linux. If you make a product that is better that free alternatives, those that need that improvement will buy the software.
Helix server is an excellent example of Linux software sellling, from a user stand point. I can stream all 3 major media types (Real, QT, and WMF) from one server, with one set of accounts, one set up issues, etc. and it's worth it to us for that alone.
Do somethig unique (as the parent has said) and it will sell, same as in any other field.
Linux games have sold - UT, Quake, as you're paying for something unique.
Cross platform is good to, as is lets those of us who like a product pick what we want to run it on. I hate finding a product is Windows only, as it means I most likely have to find yet another machine, with yet another OS license to run it on, instead of just putting it on our load balancing linux cluster.
I am, and always will be, an idiot. Karma: Coma (mostly effected by
--you offer it in different thresholds, and that's about it. Free to hobbyists, support for the payers. Software is worth paying for once it actually works as advertised for the customer, that's the pain threshold, that is usually classed as alpha or beta or full release-stable. Stable that is sold should come with the ability to actually get ahold of a human being in a timely manner for any troubleshooting questions or contracted/requested work on a feature enhancement, etc. Those with their cash transferred to you get their feature enhancements and problems bumped to the top of the line obviously. And the docs NEED to be written (the final draft) by an english major, someone who's task is to stand over the engineers once the coding is finished and it's time to finalize the man pages and docs with a clue by four until everyone is satisfied the docs are understandable and thorough and accurate. A small paragraph where every other word is an acronym is *not* a working doc to anyone except a coder. I've seen that phenomenon way too often with linux programs. Assume people who might be interested in purchasing the full release are not "all" coders, proceed from that point.
I know I pay for software for BOTH the convenience of having it on a quality made cd and ALSO because I then have a working dead trees manual that doesn't require me burning out my printer. I like the convenience of owning the cd, I like most of the time reading docs from a book, not on a screen. Some do, some don't care, that's just my "paying customiser" opinion. Free is free, a lot of people enjoy downloading constantly and compiling and whatnot, a lot more want the thing to work and be right there and easy to install (or reinstall) and have a reference manual. That's the part worth paying for, and keep the costs reasonable enough. This is like the clueless music/movie sellers. Instead of selling zillions of 5$ cd's, which they could, they make less money trying to sell 20$ cd's. Clue-less morons. It's called getting greedy, only works for a short time, then people do something else, yes?
Another thought, your program must be useful and functional to a degree at least two steps above whatever else is out there. Not one step, at least two steps better, or don't try to sell it.
Now that that's out of the way, I think it's better to just make your money admining linux in the commercial arena, then work on your custom coding and app development the same as almost everyone else is, spare time or allocated free time at your employers. The basic concept of "sharing" needs to be drilled into the bean counters heads as the long haul way to cut costs and get "more", but sharing only works if it's two ways. Everyone benfits by shared code, so the company can concentrate on building and selling their widgets. By using open source as the BASE of a company's IT, they save money. By finally bingoing that a little across the board sharing is what's allowing them to save money, perhaps they will cut some slack and allow a little company time to be used for your interesting app. That's the difference between leeching and sharing. Leeching-only as a concept is just not cool, either short term or long term. Your company benefits from open source in general, and they get first dibs on the product, useful for their business, and score brownie points as being a "friendly" and ethical company in the gestalt of society as it were. They develop a "good rep". With todays business climate, public perception is important. People are skeptical of businesses now, with dang good reason-a lot of them are shall we say less than forthright with their numbers and pretty dismal in the ethics department.
If there's no way to do this and you absolutely need to make your money off of your new application, then it must be customizable to a degree that large companies would pay for the application, it has to have a bona fide usefullness that is unique, perhaps security? That part I don't know, too many variables and wildcards. Usually in software there's the full release, then "release lite" to the non paying hobbyist public, that seems to be what is working now for some people. You offer the customized features for a customized premium in cash.
Welcome to the world of software. Many software products that used to command astronomically high prices can now be replaced with commodity Free Software products. In fact, this is what is driving Linux adoption as well. Many companies are finding that they can deploy Linux instead of a commercial operating system and save money, and so that is what they are doing.
People that are using Linux already are an especially hard sell. After all, once you start using Linux and the wide array of Free Software tools you are very likely to experiment Free Software solutions before paying for commercial ones. For example, instead of buying a proprietary instant messaging service they will almost certainly try Jabber first. For one thing, it is probably easier to "apt-get install jabber" than to purchase and test a commercial product.
I lurk on the PostgreSQL mailing lists and we get quite a few Oracle deserters, and the reason for this is simple. PostgreSQL has gotten to the point that for most uses it is good enough.