ASA Slams Microsoft For Anti-Linux Campaign

by Hinesh Jethwani    Aug 26, 2004

The British Advertising Standards Authority (ASA) has slammed Microsoft for using unethical advertising tactics in its Â’educational’ campaign framed against Linux, called “Get Your Facts Right”.

The advertisement in question appeared in a UK computer magazine, with the catchphrase, “Weighing The Cost Of Linux v/s Windows? Let’s Review The Facts.” A graph accompanying the ad compares the cost (in US$) per Megabit per second of “One Linux image running on two z900 mainframe CPUs” with “One Windows Server 2003 image running on two 900 MHz Intel Xeon CPUs”.

Underneath the ad, a highly controversial statement declares, “Linux was found to be over 10 times more expensive than Windows 2003 in a recent study audited by leading independent research analyst META Group, that measured costs of Linux running on IBM’s z900 mainframe for Windows-comparable functions of file serving and Web serving. The results showed that IBM z900 mainframe running Linux is much less capable and vastly more expensive than Windows Server 2003 as a platform for server consolidation.”

The asterisk linked to a footnote at the bottom, hilariously maintains, “Results may vary outside the United States” (So theres a fat chance that you could save oodles if you run those MS servers in Bangladesh!). The complainants maintain that the comparison was misleading, because the operating systems were run on different hardware.

Back home, the campaign has not seen any opposition from advertising circles whatsoever. A CXOtoday investigation conducted in June revealed that all CIOs quoted in the “Get Your Facts Right” campaign initiated by Microsoft India had been handpicked from archived case studies (some of them going as far as 4 years back), making their significance in the present day extremely diluted (Read the complete story here).

The advertisers in the UK example maintain that they intended the advertisement to compare competing file-serving setups that met the same needs and were intended for the same purposes. They said they had prepared the advertisement in response to an advertising campaign by IBM in which Linux running on an IBM mainframe was tested for file serving and web serving.

They further claimed that their advertisement was based on results from a benchmark study and the advertisement informed the public of the results from that study about the relative performance and cost of one Linux image running on IBM’s z900 mainframe CPUs and Windows Server 2003 image running on two 900MHz Xeon CPUs.

According to the advertisers, the benchmark study was a network load performance test that was neither hardware specific nor operating system specific; they said the fact that the hardware and operating systems were different was irrelevant. They pointed out that the client PC did not determine the server used and that the server workloads were the same and were functionally equivalent. The advertisers explained that each server was tested to deal with increasing numbers of functions from client PCs. They said they took measurements from the client PCs to assess how fast the server would respond. They asserted that the study was audited by Meta, an independent consultancy firm, who reported that the study was a fair comparison.

The Authority noted that because the advertisement stated “Weighing The Cost Of Linux v/s Windows”, it implied the comparison was between Linux and Windows operating systems only, and not about the performance of operating systems on different hardware. The measurements for Linux were performed on an IBM zSeries, which was more expensive and did not perform as well as other IBM Series and that the advertisement implied running a Linux system was, in general ten times more expensive than running a Windows operating system.

ASA maintains that readers would infer that the advertisement compared Linux and Windows operating systems only and that the advertisement implied running Linux operating systems were, in general, ten times more expensive than running a Windows operating system.

Because the comparison included the hardware, as well as the operating system and therefore did not show that running a Linux operating system was ten times more expensive than running a Windows operating system, the Authority concluded that the advertisement was misleading. It advised the advertisers to amend the advertisement and advised them to seek help from the CAP Copy Advice team.

Tags: ASA