After a sequence of pieces focused on software reseller or services businesses, let's cast our attention on how we can use Linux to improve the fortunes of resellers in the PC hardware assembly industry.
The first computer I purchased (a Z80-based CP/M system) came from Computerland in 1982. Back then, computer hardware resellers (like the Computerland franchise) were the only channel game in town. They were intimate with the product offering and in many cases were technology enthusiasts themselves who had a keen sense for trends - their businesses flexibly adapted to meet or precede customer demand. In the ensuing time, I've watched the industry grow from the pure preserve of the technologists to become something of extreme utility, which effects nearly every organisation in the country. The channel has generally done very well; it's a vital component in pushing the message out as well as helping users adopt successive waves of new technology.
In recent times, however, all is not so good. Economically speaking, this is perhaps the most deflated period I've seen in my 20 years in the industry. Part of the problem is that many customers feel they have ploughed a lot of resources and money into technology solutions, yet are often unhappy with the returns.
How do we in the industry change this perception about the lack of value in our offerings? What can we do to stimulate customers' interests and buying habits? If you're in the hardware assembly/white-box segment of the market, I have a simple business plan that might just help. But first, let me recount a story which I believe helps explain why I think that local assembly resellers are in the best position to benefit from Linux.
Recently, I was involved in pricing hardware for a couple of clients: 30 and 55 PCs from two of the biggest tier-one PC channel suppliers. Neither of these vendors could provide me with a workstation without Windows pre-installed or without the price of Windows included in the deal. Even after explaining that the intention for these systems was to run Linux and that the client had decided to opt out of paying for Windows, the suppliers feigned intractability - there was nothing their local reps could do; their hands had been tied by decision makers far removed. I therefore turned to a local assembly firm to meet my pricing requirements. They had no problems in delivering to my requirements, confirming in my mind that this flexibility in offering can be turned into a business differentiator. This group doesn't need to wade through a complex bureaucratic quagmire to produce customised or targeted products. Manoeuvrability and speed of adaptation are inherent strengths. By deploying Linux to accentuate these strengths, the local build channel can both find a niche, fill a need and help create both.
I posit that dual-boot Windows/Linux workstations are better value for your customers and can serve to differentiate your product. Nothing revolutionary here, but why install another operating system with the PCs you build? For starters, hard disk space is cheap. Linux and its applications need only 5GB of spare room - not much off that 60GB hard disk in the new workstation. Also, it costs you nothing but a few minutes of time to dump an image of the Linux distro to the PC's disk. Remember, this is free software; there are no additional licences to pay beyond the Windows licence you have already factored in to the cost. The chief selling point of adding a Linux distribution to the PCs you supply customers is to offer better overall value for money than the competition. Any Linux distribution you can freely download off the Internet includes several thousand open-source applications; spreadsheets, powerful graphical manipulation programs, CAD packages, personal finance packages, and so on. It's these that benefit the customer and leave them with a renewed sense of value for money.
On to methodology. We know that installing both operating systems with their sundry applications in a dual-boot arrangement is best done from the outset. How to do so has been recounted in hundreds of online tutorials and by almost every Linux systems book. Linux already ships with the repartitioning and resizing tools (GNU Parted and Disk Druid, for instance) or you can invoke gee-whiz utilities (like Partition Magic) which fully understand Linux partitions. As a last resort you can use ol' faithful fdisk. Linux comes with graphical dual-boot loaders which allow safe and easy selection of and interoperation with Windows 98/Me/2000/XP etc.
If you'd like to provide additional value, resell the boxed set of the full Linux distribution you wish to bundle with your dual-boot workstation. This will only increase the price of your system to the consumer marginally, and definitely makes an impression in terms of overall value. Expect a buy price of between $30 and $200 for the various boxed Linux sets, depending on manual count and number of packaged apps. A full boxed set of SuSe Linux 8.0 includes thousands of pages of manuals and sprawls across seven compressed CDs and a DVD. It almost reminds me of the boxed-set manuals that Microsoft used to ship with MS Office a decade ago, back when customers used to believe they were not getting ripped off. Suffice to say, it puts an end to the claim that Linux lacks desktop apps. For additional revenue, you can charge a fee for helping your customer through services such as installation of networking or firewalls; helping them configure PPP, PPPoE or Cable Internet links with their new Linux system or help them integrate the system into their small business LAN. Remember, Linux is new. New things drive industry revenue.
A side benefit of getting these apps into the hands of customers is that it may also help curb the problem of software piracy, a practice both unethical and illegal, which affects all members of the channel. By supplying hundreds of quality end-user applications to your customers - pre-installed, ready and waiting - you reduce the impetus for these users to pirate proprietary software, which they may not have been able to afford in the first instance.
If you're at all concerned about providing support for the Linux part of the dual-boot system, ensure that your customer understands that the Linux software is not supported by you and that, as with the Microsoft equivalent, the customer will need to contact the supplier (Red Hat or Caldera, for instance) and pay for support if needed. Customers expect nothing more nor less from Windows. The boxed-set versions of Linux ship with prepaid support options.
Finally, in case you were wondering if Linux remains the preserve of the technologists who once used CP/M systems, remember one thing: even CP/M (in fact, a cheap clone thereof, that was re-badged as MS DOS by a little company then known as MicroSoft) eventually morphed into the Windows we have today. Changes do happen; not only is change the core nature of this industry, it's what helps drive sales of the next big thing. Grab onto the Linux philosopher's stone and help turn your local assembly iron into gold.