Posted on 02/04/2007 7:40:12 AM PST by ShawTaylor
February 2, 2007 -- (WEB HOST INDUSTRY REVIEW) -- Internet research organization Netcraft (netcraft.com) announced on Friday the results of its Web Server Survey for February 2007.
Netcraft says it received responses from 108,810,358 Web sites, an increase of 1.93 million from last month.
According to the survey, Apache had a decline of 442K Web sites this month and saw its share of the Web server market slip by 1.47 percent to 58.7 percent. Netcraft says this is the first time Apache's market share has been below 60 percent since September 2002.
Microsoft-IIS on the other hand gained 935K Web sites, continuing its advance towards Apaches leading position. In the February 2006 survey, Apache held a 68 percent market share, giving it a lead of 47.5 percent over Windows, but this month, Microsoft's share has increased to 31 percent, narrowing Apache's advantage to 27.7 percent, says Netcraft.
Last I heard, Microsoft was paying folks to run their "parked" web sites (reserved domain names but no real content) on MS servers, just to boost their position in this race.
IIS, a hackers paradise.
"IIS, a hackers paradise."
Is that why ever more web sites are using it?
Laugh all you want, but here is what NetCraft themselves said about it:
This month's survey brings one of the largest one-month swings in the history of the web server market, as Microsoft gains 4.7 percent share while Apache loses 5.9 percent. The shift is driven by changes at domain registrar Go Daddy, which has just migrated more than 3.5 million hostnames from Linux to Windows. Go Daddy, which had been the world's largest Linux host, is now the world's largest Windows Server 2003 host,
Your assertion is wrong. The following paragraph from the site you linked,
In the February 2007 survey we received responses from 108,810,358 sites, an increase of 1.93 million from last month.
Shows why, Apache did not lose current market share, Vista was introduced and inflated the numbers for "New" servers, not current servers.
If apache spent the kind of money MS did for it's Vista release, I dare say it would have 90+ percent of the 1.93 million new boxes mentioned.
Just let Vista settle in for a few months then come back and tell me that.
It hardly matters whether Microsoft has the best web server or not to people like me. I am off the reservation for good, running Fedora, Kubuntu, Mysql, Netbeans Java, PHP, Apache, Tomcat, Gimp, Open Office. I still have two Win 2000 machines for legacy things, but basically the monkey is finally off my back.
Anymore, improvements in hardware and software barely move my operational efficiency - we are already "good enough" in most areas. Therefore, what floats my boat is 1) licensing issues, 2) long term code stability (not changing Visual Studio every three years for example), 3) grid and clustering, and finally 3) cost.
Vista and Microsoft don't offer me squat in those four areas, so I could care less about fancier buttons or "better" DRM and my security is already fine. So, who cares if IIS picks up a point from Apache, let someone else pay the microsoft tax.
As far as MS paying for it, it was all over the net. I'm not up for more research to prove it, I just don't care that much.
These figures that Netcraft uses are not even close to being reliable. For instance, they said "Websites" that responded. Most servers have anywhere from 1-1000+ different websites, so 1 server could count as 1000 responses.
By a margin of 1000 to 1, PC's are sold with MS(IIS) installed when you buy them. So by my calculations Apache should have lost a heck of a lot more then a few percent, again if Netcraft's numbers were reliable.
Now if you really wanted to use reliable numbers you would compare the sales of Linux type Software sold(either alone or on a new PC) versus Microsoft.
Those are hard numbers, and those numbers show MS on the losing end every month/year for at least the last 5.
Now you may ask how I know netcraft's numbers are not reliable, I manage over 15,000 sites on 6 servers and just about 1/3 of those sites I manage got the "Survey". All six of those servers run Unix/Linux, so that means if I took the time to complete their survey that would count as 5,000 responses.
Surveys are useless unless you can guarantee that all respondents are,
(1) Telling the truth.
(2) Using information from different servers for each OS.
I refuse to let any company get any information from my servers without my permission. I will purposely have my servers send a fake response if someone trys. It's my business unless I want them to know.
I wonder if a survey of business has been done. In Alaska, at least, LAMP has become the dominate OS/architecture for big ISPs. Others have been looking to OSX as a server OS. The school districts here are, of course, MAC reservations.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.