Windows 2003 vs Linux

[Linux_cat]

I've just noticed the advert at the top of the page, does windows 2003 server really outperform Linux, or are they just talking s*it?.

[je_fro]

If you use mozilla (or a derivative thereof) :
Try a forum search using "usercontent.css" posted by Hayl. You won't be bothered by that trash anymore.

[blingbling!!]

Anything outperforms anything else as long as you've decided in advance what the oucomes of your tests are. I like to quote Vic Reeves in these situations (UK Comedian....)67.8% of all statistics are made up on the spot

Statistics ..... BAH!! I knew there was a good reason that i failed my Maths degree ;)

--Robin

[hard candy]

Also, if you use Konqueror, go to settings, Stylesheets, Use user-defined stylesheet, and point to a copy of the file jefro was writing about. That will cut out most of the ads.
Here's a copy, but look at the thread, you can make more filters:
/*
* This file can be used to apply a style to all web pages you view
* Rules without !important are overruled by author rules if the
* author sets any. Rules with !important overrule author rules.
*/

/*
* turns off "blink" element blinking
*/
blink { text-decoration: none ! important; }
*

/* remove Ads */
*[src*="mds.centrport.net"],
*[src*="images.deviantart.com/shared/topads.gif"],
*[src*="view.atdmt.com"],
*[src*="www.justlinux.com/RealMedia/ads/click_lx.ads/intm/linux/www.justlinux.com/index/706227077
/468x60-2/House_Dealtime_Promotion_1e/dealtime_750_2.html/38653$"],
*[src*="www.justlinux.com/RealMedia/ads/Creatives/House_Dealtime_Promotion_1e/bg-left.gif"],
*[src*="www.justlinux.com/RealMedia/ads/Creatives/House_Dealtime_Promotion_1e/sp.gif"],
*[src*="www.justlinux.com/RealMedia/ads/Creatives/Hoovers_Online_4e"],
*[src*="www.justlinux.com/index/1238038594/house_ribbon/House_Ent_Linux_Fall03_1n/elforumsave-03.gif"],
*[src="www.justlinux.com/index/1238038594/house_ribbon/House_Ent_Linux_Fall03_1n/elforumsave-03.gif"],
*[src="http://www.justlinux.com/RealMedia/ads/click_lx.ads/intm"],
*[src="http://www.justlinux.com/RealMedia/ads/click_lx.cgi"],
*[src="http://www.justlinux.com/RealMedia/ads/Creatives/Nortel_Networks_Email_1b/sp.gif"],
*[src="http://www.justlinux.com/RealMedia/ads/Creatives/Nortel_Networks_Email_1b/right-bg.gif"],
*[src*="hardwarecentral.dealtime.com"],
*[src*="http://http.edge.ru4.com/smartserve"],
*[src*="www.justlinux.com/RealMedia/ads/Creatives/House_Dealtime_Promotion_1e/go.gif"],
*[src*="www.justlinux.com/RealMedia/ads/Creatives/House_Dealtime_Promotion_1e/logo.gif"],
*[src*="www.justlinux.com/RealMedia/ads/Creatives/House_Dealtime_Promotion_1e/bg-right.gif"],
*[src*="www.justlinux.com/icom_includes/special/cdxpo/80211-ribbon.gif"],
*[src*="www.justlinux.com/icom_includes/"],
*[src*="www.justlinux.com/RealMedia/ads/click_lx.ads/intm/linux/www.justlinux.com/"],
*[src*="www.internet.com/corporate/"],
*[src*="http://hosting.verio.com/index.php/vpsmanaged_vps.html"],
*[src*="hardwarecentral.dealtime.com/BannerIn"],
*[src*="www.verisign.com/images/landing/lan6031/ecom_form_160x209.gif"],
*[src*="hardwarecentral.dealtime.com/xKW-"],
*[src*=".qz3.net"],
*[src*="ads.pointroll.com"],
*[src*="mirror.pointroll.com"],
*[src*="www.justlinux.com/RealMedai"],
*[src*="a.tribalfusion.com"],
*[src*="/icom_includes/special/cdxpo/cdxpo-top.gif"],
*[src*="fmads.osdn.com"],
*[src*="www.justlinux.com/icom_includes/special/cdxpo/cdxpo-top.gif"],
*[src*="www.justlinux.com/icom_includes/special/cdxpo/"],
*[src*="www.justlinux.com/icom_includes/special/cdxpo/80211-ribbon.gif"],
*[src*="/icom_includes/special/cdxpo/80211-ribbon.gif"],
*[src*="www.tinyminds.org/images/banners"],
*[src*="images/banners"],
*[src*="adfarm.mediaplex.com"],
*[src*="images.deviantart.com/affiliates/"],
*[src*="images.deviantart.com/shared/winzip.gif"],
*[src*="www.justlinux.com/RealMedia/ads"],
*[src*="images.deviantart.com/banners/"],
*[src*="images.deviantart.com/shared/trillian.gif"],
*[src*="sfads.osdn.com"],
*[src*="ads.doubleclick.com"],
*[src*="cache.burstnet.com"],
*[src*="ads.doubleclick.net"],
*[src*="ad.doubleclick.net"],
*[src*="ad.doubleclick.com"],
*[src*="us.a1.yimg.com"],
*[src*="www.theregister.co.uk/media"],
*[src*=".qz3.net"],
*[src*="a207.p.f.qz3.net"],
*[src*="a207.p.f.qz3.net/7/207/2554/v001/eyewonder.download.qz3.net/2554/114720"],
*[src*="/RealMedia/ads"],
*[src*="www.burstnet.com"],
*[src*="/ads"],
*[src*="ar.atwola.com"],
*[src*="spd.atdmt.com"] {
display: none ! important;
visibility: hidden;
}
A:link[HREF*="ad."] IMG { display: none ! important }
A:link[HREF*="ads."] IMG { display: none ! important }
A:link[HREF*="/ad"] IMG { display: none ! important }
A:link[HREF*="/A="] IMG { display: none ! important }
A:link[HREF*="/click"] IMG { display: none ! important }
A:link[HREF*="?click"] IMG { display: none ! important }
A:link[HREF*="?banner"] IMG { display: none ! important }
A:link[HREF*="=click"] IMG { display: none ! important }
A:link[HREF*="/ar.atwo"] IMG { display: none ! important }
A:link[HREF*="spinbox."] IMG { display: none ! important }
A:link[HREF*="transfer.go"] IMG { display: none ! important }
A:link[HREF*="adfarm"] IMG { display: none ! important }
A:link[HREF*="bluestreak"] IMG { display: none ! important }
A:link[HREF*="doubleclick"] IMG { display: none ! important }
A:link[HREF*="iclick_lx.cgi"] IMG{ display: none ! important }
/* disable ad iframes */
IFRAME[SRC*="ad."] { display: none ! important }
IFRAME[SRC*="ads."] { display: none ! important }
IFRAME[SRC*="/ad"] { display: none ! important }
IFRAME[SRC*="/A="] { display: none ! important }
IFRAME[SRC*="/click"] { display: none ! important }
IFRAME[SRC*="?click"] { display: none ! important }
IFRAME[SRC*="?banner"] { display: none ! important }
IFRAME[SRC*="=click"] { display: none ! important }
IFRAME[SRC*="/ar.atwo"] { display: none ! important }
IFRAME[SRC*="spinbox."] { display: none ! important }
IFRAME[SRC*="transfer.go"] { display: none ! important }
IFRAME[SRC*="adfarm"] { display: none ! important }
IFRAME[SRC*="bluestreak"] { display: none ! important }
IFRAME[SRC*="doubleclick"] { display: none ! important }
IFRAME[SRC*="www.verisign.com"] { display: none ! important }

/* turning some false positives back off */

A:link[HREF*="download."] IMG { display: inline ! important }
A:link[HREF*="click.mp3"] IMG { display: inline ! important }
A:link[HREF*="newthread"] IMG { display: inline ! important }
/*
* For more examples see http://www.mozilla.org/unix/customizing.html
*/
/* block some flash */
embed[type="application/x-shockwave-flash"][src*="spd.atdmt.com"] {
display: none !important;
visibility: hidden !important;
}

blink { text-decoration: none ! important }
/* marquee { -moz-binding: none ! important } */
:link, :visited { text-decoration: underline ! important }
img[src^="*/ads/*"] { display: none ! important; }
*[src*="bs.servingsys.com"] {
display: none;
visibility: hidden;
}

[Pafnoutios]

Windows 3.1 was Microsoft's best chance at outperforming linux. And at that time, it probably did. Every version since then has been worse than the previous.

[rocketpcguy]

it's faster than linux 2.4, yes. but not 2.6

if you use firefox, get the adblock plugin

[lagitus]

Originally posted by rocketpcguy
it's faster than linux 2.4, yes. but not 2.6

WTF???

[Parcival]

Statistics ..... BAH!!

It's always the same story - those who don't understand statistics are the first ones to cry that all statistics are faked. :rolleyes: Yes, it's possible to fake and bias statistics, but it's also possible to do honest and clean statistics. Probably not in this case, though. :)

[Icarus]

Originally posted by Parcival
It's always the same story - those who don't understand statistics are the first ones to cry that all statistics are faked. :rolleyes: Yes, it's possible to fake and bias statistics, but it's also possible to do honest and clean statistics. Probably not in this case, though. :) Finally! Someone else in the world that understands statisics (besides my college professor)! :D

I always like how someone sees a % and thinks it was made up and pulled out of the air (it does look that way to someone that doesn't understand it, I guess)

Anyway...why are you looking at the banners?? Arn't most people now immune to them and don't even notice? I know I don't see them and I'm not blocking anything!! :p

And thanks Joel, for messing up the formatting of the page with a long URL! :p

[Stween]

Originally posted by Icarus
I always like how someone sees a % and thinks it was made up and pulled out of the air (it does look that way to someone that doesn't understand it, I guess) I always laughed when a lecturer we had last year was running over some stats stuff (the primary aim of the module wasn't stats, but we used a lot of the stuff none the less).He summed it up as "a complex way of lying".

He had some great stories though, such as the story from when he worked at Nokia and they released the Nokia 9210 (http://www.nokia.com/nokia/0,,147,00.html) without doing any market research beyond three guys in one office. When the things started shifting much, much quicker than expected and they ran out of stock, Nokia had to hurriedly do some market research and throw some numbers together to make it look like they knew what they were doing ;) Slightly off-topic, but amusing.

[Icarus]

Originally posted by Stween
He summed it up as "a complex way of lying". I prefer the "accurate statistic with the available data" :D
What is available could be anything!

[mrBen]

Statistics are only as good as their 'meta-data' - It is useless to just read the raw figure, you need to read the associated text that explains how the figure was calculated and what it actually means.

For instance, the statistic: "100% of women surveyed think mrBen is the sexiest man alive" looks completely unrealistic. But it _could_ be true, if the associated text indicates that women surveyed actually was only my wife.

When it comes to Windows statistics, generally what they have done is used circumstances where Windows runs faster / is cheaper or whatever, and ignored the rest. By using a very specific data set, they can indicate what they like - it's not a lie, it's a selection of the truth.

[hard candy]

But it _could_ be true, if the associated text indicates that women surveyed actually was only my wife.
Was this asked before she found out about the new computer you had bought? :D

[cybertron]

Originally posted by mrBen
By using a very specific data set, they can indicate what they like - it's not a lie, it's a selection of the truth.

Yep, and this is why I generally don't trust statistics much. Unless they support my preconceived notions, in which case they're absolutely true:)

[Linux_cat]

Thank you, i feel alot better now. I can now get on with my life.

[Parcival]

For instance, the statistic: "100% of women surveyed think mrBen is the sexiest man alive" looks completely unrealistic. But it _could_ be true, if the associated text indicates that women surveyed actually was only my wife.

Note quite, because there is no such thing as a "true figure/result" in statistics. Let us call back what statistics is about:

statistics is the quantitative analysis of data with the goal to determine if the frequency of certain cases can be marked off from mere chance.

Example: we start a vote called "Who's your favorite mod of the day?" and Bon Bon happens to win with 67% of all votes while mdwatts goes with meager 5%. Now we can test this result for significance, i.e. test it for the probability that this result arrived because of a true preference within the voting population or if it happened just randomly. (imagine every voter flips a coin and posts the vote according to it - the given result above could be a possible outcome of a coin toss)

Even worse, besides chance we have also to deal with errors in statistics. One possible error can stem from our test being screwed (e.g. because the question isn't worded carefully), our test-person not paying enough attention, typing mistakes when we punched the data into the keyboard, etc. That's why we also have to indicate the probability of an error when we publish results in statistics.

For all those who didn't understand anything above, I put everything into a few simple words:

Statistics deal with chance and probabilities and thus only produce results of a certain probability.

Ethically good statistics are transparent about the uncertain aspects of their results.

When you publish a statistic analysis in a good journal, you always talk about the sample your analysis is based on and the statistic methods you use, plus you also name the author where you can receive the data file to analyze the sample yourself - science is about falsification, baby. :cool:

Bill, could you send over your serverstudy datasheet? Thanx.

[hard candy]

Bon Bon happens to win with 67% of all votes
I would not be assigning this to chance, this has happened before when Bon Bon was running against Herphes for class president in the 6th grade. Bon Bon got 70% of the vote while Herphes got the other 70%. (See we had a class of 10 in our special ed class and we had 4 visiting students that day, so we had 140% in our class the day we voted. Luckily, Bon Bon had brought a pack of gum that day so he could get the 4 extra students to vote for him). Well, we had a runoff since it was a tie. In that vote Herphes won with 50% while Bon Bon got 30%. 2 students were sick that day so we only had 8 voting. If they had been there Bon Bon would have gotten their vote since he had traded a possum to one for his vote and he promised to not call Sushi Smelt if she would vote for him.