Slashdot Mirror
The Real Problem With Alexa
As the defacto 'Guy in Charge' of a reasonably large web site, I am routinely asked questions by a variety of people that lead inevitably to Alexa. It might be a question from my Boss at SourceForge about traffic. Or it might be a sales guy asked by a possible advertiser why some other random website is bigger or smaller than Slashdot. Most often it's a random reporter doing background for a story that has nothing to do with Slashdot. Why I'm considered an expert is very confusing, but why they always regard Alexa rankings as meaningful is even more so.
Here's the problem: Alexa doesn't work because of who will install it, and perhaps more importantly, who won't. Let's start with a place I'm very familiar with: Slashdot readers. Until recently Alexa didn't work on Firefox... instead only IE users participated. On the internet as a whole that's fine: like 80% of users run IE. But on Slashdot only like a quarter of you do.
What about re-installing the plug-in after you update your browser? When Firefox 2.0 came out, almost a third of Slashdot readers upgraded within a few days. You upgrade Minor Firefox releases overnight. Even IE users of Slashdot update relatively fast, from 6 to 7 or even minor revisions. New versions often break old plug-ins. When you get that alert that a plug-in is out of date do you just forget about it? I know I do. And that's not even counting clean OS installs. But if I went to random non-technical friends and family installations, I frequently see versions of software so dated it makes me cringe.
And that's not even talking about the fact that Alexa's toolbar is pretty much spyware. How many Slashdot readers are giddy to install spyware? You either? Big surprise. Because of who we are, and what it is, our population will self select out of consideration.
Did you know Alexa excludes SSL? How many etrade users do you think there are? Now personally I'm glad that they aren't tracking my browsing at my credit card company, but it's just another factor reducing accuracy.
Equally perplexing is the accounting of iframes. Let's look at someone like double click's alexa rating. Now it's hard to say, but I don't think I've ever visited their website. Have you? But according to Alexa, they have nearly a 1% share of the internet. I'd tend not to believe it... but they have iframes on zillions of web pages and counting those sure would account for this huge ranking. What about all those badges for the popular social networking websites? What influence are those iframes having on Alexa rankings? Alexa's FAQ says they don't count, but I'm skeptical.
In Fact, Alexa KNOWS that it is a flawed metric for measuring. Have you ever tried actually looking up alexa on alexa? Unsurprisingly, it is unavailable. Why? Visitors to Alexa.com would be the most likely of any user population on-line to have installed their plug-in. I don't know what their 'Rank' would be, but I bet it clearly would be an apples to oranges comparison against ANY other site on-line.
Of course who do you think actually will go out of their way to install something like this? I have a good guess... if you are obsessed with acronyms like SEO or terms like PageRank you are very likely to care very much about these things. I spend a real percentage of my week dealing with people flooding my systems with garbage content designed to screw with these ratings. And you know they all have the toolbar installed so their zillions of worthless spam websites are being counted.
This problem has parallels elsewhere of course: The Nielsen ratings struggle to account for PVRs. Since you got a TiVo, when was the last time you watched "Live" TV? This is part of why Science Fiction shows struggle on TV... scifi fans are early adopters. So we stopped getting counted and our favorite genres are butchered by networks and lost to the void. PVR users tend to be wealthy (those boxes are expensive) and educated. Now I'm not saying that the dumbing down of TV is exclusively the fault of Tivo, but it sure didn't help that we weren't being counted as excellent "Smart" TV shows get canceled while we keep getting more seasons of Survivor. Who we are and how we live causes us to not be counted, and this has unintended consequences.
So what do we do? I wish I had a good answer to this. My first suggestion would be that if anyone mentions Alexa to you that you freak out and go on a 5-minute rant about how Alexa is stupid and anyone who is using it to seriously make a business decision should be fired. It doesn't actually help, but i estimate that every time I do this, I burn the same number of calories as I might on an elliptical trainer. I assure you the beer gut ain't getting smaller on its own.
Alternatively you could just install the toolbar on every machine you can find and skew the numbers ridiculously towards people that are likely unrepresented. Of course, the conspiracy theorists amongst you will just bitch that I'm trying to fudge Slashdot's own rankings in a system I'm claiming to hate. But that only helps proves my point... the conspiracy theorist is a demographic strongly represented on Slashdot that is unlikely to trust this software. We all ignore a broken status quo "Gold" standard that would fail a 100 level college science class on the grounds of flawed methodology. And this only leads to us not being counted.
Isn't Alexa considered spyware?
It baffles me how people actually look to them for information, considering how they get it.
I should install something that you claim is broken?
I love the smell of astroturf in the morning... at least it almost smells like that.
That's all true, but unless someone's got a better alternative, it doesn't matter.
It isn't surprising that people who spend money on advertising want to have some metric by which to predict (estimate, guess, what-have-you) the impact of each dollar spent on web advertising. Assuming the people spending the money are, as a class, either stupid or ignorant is a mistake. Odds are good that many of them know that Alexa is flawed, but also consider any information better than nothing. If nothing else, Alexa rankings demonstrate the relative popularity of a web site among Alexa participants - which is at least a concrete demographic, and the stats are inarguable on that basis.
What's being missed is that there's a fundamental problem, here. Populations which refuse to share information with such aggregators will always self-select against representation. It's no different, really, than stating that populations who do not vote self-select against being represented in government. That doesn't stop us from using elections as a way to select people into government.
In the specific case of slashdot selecting against itself, it's debatable whether we're a demographic many organizations would even want to target (with web advertising) if they could. How many comments on how many stories have included someone claiming that he's either unaffected by or negatively affected by advertising? That he's less likely to buy a product he sees advertised? Broader yet, how do you suppose the median number of lifetime banner ad clicks for the slashdot user compares to that of the web-using population at large?
I posit that we pose a particularly galling challenge to marketers. On the one hand (if you'll allow me a bit of net-cultural hubris), we're a demographic of above-average intelligence, above-average income, with an above-average tendency to spend money on brand new technology, and who have an above-average impact on what other people will buy. On the other, we refuse to share our habits with "big brother," we're easily offended (eg, we hate proprietary formats solely because they're proprietary), comparatively hard to bamboozle, and have a cultural predisposition towards "free" (both beer and speech). That is, on the one hand, we're a fantastic demographic to succeed with, but on the other, we're a tough nut to crack.
The point is that Alexa is flawed, without a doubt. But it seems more flawed from the point of view of a group which deliberately makes itself all but impossible to measure. And frankly, if we're not willing to provide the information necessary for advertisers to make informed choices, we're going to continue to be ignored, both on the web and on television. (Yes, I do realize that Nielsen is specifically flawed with respect to DVRs - but even if they weren't, how many members of this site would voluntarily install habit-tracking software on their TiVo? How many members of this site would call for a boycott of TiVo if it installed it for them?)
Reality has a conservative bias: it conserves mass, energy, momentum...
That's probably why it's filed under 'Editorial'.
... but what is Alexa ?
Will have to reread this, but it doesnt come off as news but a rant. And no I wont install the toolbar.
"Rant" ?
CmdrTaco is being rebel, anti-establishment, rage against the machine, fuck the system! This is what he's done here, and he deserves *respect* old man.
Back in the days, when we were pissed about religion, wars and social injustice, we dressed like goths and sang bad rock and roll and emo music.
But today, thanks to the world wide web, we take the next level, and all this unrelenting energy in today's youth comes in the form of a rant against a toolbar that rates sites. And I say, bravo.
Of course it's a rant, it's an editorial. /rant.
The tags were there before TFA.
Furthermore you will need to re-read it because of your race to FP you likely only read the front page blurb.
-nB
whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
...because digg.com is beating slashdot.org :)
It might not be news in the sense of something that just happened, but its still worth being reminded about once in a while.
Its true - the people who obsess over PageRank and SEO are the ones driving this - everyone already knows their own web stats just by looking at their server logs.
Kevin Smith on Prince
When I used to administer a website (b2b, you've never heard of it) my boss loved Alexa. I told him time and again to uninstall it, and even did so myself a number of times... but he'd put it back every time. Then, one day, all dynamic content on the main page just vanished. I brought it back from backup, and chocked it up to a bug. Then, it happened again a little while later. I started snooping around our logs.
Turns out, Alexa's spiders were ignoring the robots.txt file, and capturing usernames and passwords. It logged into the administrative area, and followed the "delete" link for every entry. My dumbass boss still didn't want to uninstall Alexa. Could have strangled the man.
Don't worry, you'll be able to re-read it tomorrow when Zonk makes a dupe post.
Better metric...
I wonder if Google publishes the click-through rate for keywords to sites in the search index?
you know they have the numbers from adwords, but still....
whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
Nope read what he wrote. Didnt notice is was CmdrTaco though, so sorry. Guess as an editorial it's interesting and sums up a lot of feelings people here have. Just when looking for a news post it seemed out of place. There are to many "I hate X software" rants and seemed odd that the very application being told about suggests installing it.
"I have to come up with an excuse for why Digg is 'beating' Slashdot." LOL.
My first suggestion would be that if anyone mentions Alexa to you that you freak out and go on a 5-minute rant about how Alexa is stupid and anyone who is using it to seriously make a business decision should be fired.
I've been doing this for years. The problem (or actually just what marketers perceive as the problem) is that there is no generic public way to compare web site traffic. The only true way to get traffic metrics is from the web site owners. And they could easily make it up to take in more advertisers. So people in advertising look to Alexa as the only third party source.
The biggest sites don't have as much of a problem because they can work closely with advertising partners. Medium and small sites, however, don't get as much personal attention. So proving themselves as worthy web space for ads is more difficult.
The only people I've heard of that install the Alexa toolbar are web site owners because they want to see their rank often. Ironically so few people have the toolbar installed that they drastically boost their own rank.
We need to convince marketers that Alexa is pointless. But I'm afraid that without a good replace they'll keep using it.
Developers: We can use your help.
Maybe this is why: http://www.alexa.com/data/details/traffic_details? url=slashdot.org
http://www.alexa.com/data/details/traffic_details? url=digg.com
Come on folks, it's time to be counted!
Now where can I download the Alexa plugin for lynx?
Oops.
I hate to say it, but that really proves not as much that "The only way advertisers can get accurate data as people opt in", as it proves that they have not elected to find new methods to track data properly/independantly. If you were able to develop a way to get honest and accurate data of the number of hits on a site to site basis, would even that be more accurate? (assuming you started to collect an enormous list of sites). Say check all the news aggregator websites language by language (I'm sure there's thousands in each), but rank them by who is getting the most unique hits in a day, etc? Of course a site could skew their own results which creates its own problem but would this not at least be more valuable than alexa data?
So Alexa says they are not spying on the user. Big surprise.
How can I verify what this toolbar is really doing unless I have the source code? IMHO the problem lies there: There is no trust for Alexa because nobody can really say for sure how it works and that it doesn't harm the user.
How is Alexa different than any other selective-survey system? The Nielsen ratings are acquired via 'diaries' (or occasionally set-top boxes). Radio 'listener share' is determined similarly by Arbitron. The NY-Times bestseller list is based on books sold to distributors, not books sold to the public (millions of unsold 'bestsellers' get pulped or donated to libraries every ear).
Just come to terms with the fact these organizations are in bed with advertisers and move on with you life.
I judt got a nre Kinesis keybiartf so please excusr ant egregiou typos.
"Hi boss, I've ordered that new Expensive Server, it arrived, and I installed it on our network."
"But Taco, you've already had Expensive Server installed three days ago! I thought the Purchasehose was supposed to eliminate dupe servers!"
"Boss, this one is like, totally different. Ask Zonk and kdawson, they know."
"Of course since most of you are Slashdot readers..."
hm... given the context I would say 'most' is an understatement
"It isn't surprising that people who spend money on advertising want to have some metric by which to predict (estimate, guess, what-have-you) the impact of each dollar spent on web advertising."
There are several easy ways:
If you're so naive as to not insist on hard numbers for actual views (the log files are best , you deserve to get hosed - you can analyse the log files and factor out multiple views per host ip to get the actual number of real views, and reduce fraud; ditto with geolocation of ip addresses to factor out bots in 3rd world countries; ditto for bots that crawl every link on a page; ditto for pages that are loaded then immediately dumped for another page).
As an advertiser, I'd want unique eyeballs - real human eyeballs - that can be verified.
Kevin Smith on Prince
The reason why DoubleClick has so much traffic is because it's an ad company. And afaik every time you load a page with ads (and you don't use an adblocker), some content from DoubleClick's page is transfered. This way DoubleClick gets so much traffic.
In my experience, a lot of PHBs are only too happy to have information. They don't really care if it's valid information or not, just so long as it's there and that it sounds good.
It was a massive wake-up call to realise how many middle-managers and the like will quite happily swallow any old crap as long as they perceive that it's authoritative. Has anyone ever tried to tell them about how bad the information is? (real question btw - I'm interested in seeing if other readers experiences were as bleak as mine).
bang goes my karma... again...
It clearly has biases, and (worse) these seem to change slowly with time, but for the web sites I host, there is a nice correlation between their Alexa reach and their
hit count.
It is certainly good for a crude ranking of sites - Slashdot's rank right now is 558, and that clearly means a lot more traffic than some site than a rank of 5 million.
So, like many other measures on the Internet, it is flawed, but it has value.
I forget the url just now (something like data.alexa.com) but all you have to do to skew the alexa stats is pass your redirects or ping a certain url with your url at the end.
Then alexa thinks the toolbar has tracked you to a site. Your stats go up.
No install needed. Just check a firewall to find the hidden url.
It's very different than the Nielsen ratings. Nielsen does not let random users join their system. They seek our a statistically valid sample population, and pay them for using the system. That is extraordinarily different.
I would LOVE to have a similar scenario for other ad-driven media. Imagine if I could flag TV commercials with "not interested" and then never see that commercial again, or any commercial for a similar product. Once it got a good feel for what I really like and don't like, I probably wouldn't feel the need to skip commercials. The same could be said of web ads. If I could cherry-pick which ads I was interested in and which I wasn't I might not be so inclined to block ALL of them.
Ads are useful to me sometimes, but picking the signal out of the noise is usually such a hassle that I'd rather just skip the whole process. If everyone could make a very personal statement about what they want to see ads for and what they don't, I think the benefit for both parties would improve.
Symantec corporate flags the alexa toolbar as spyware, so I couldn't run it if I desired to.
. jsp?docid=2004-062410-3624-99
http://www.symantec.com/security_response/writeup
I'll just use my special getting high powers one more time...
I imagine ratings like these are mostly of interest to advertisers. Thanks to a little firefox plugin this means the ratings are even more skewed. Or maybe the two cancel one another since people reluctant to install Alexa is also the users likely to install ad-block, thus giving advertisers exactly what they want. It is effectively a number which says where naive fools that will opt in to stuff like this can be found... In either case slashdot need not care since page-views do not translate into ad's viewed anyway. Especially not among slashdot users. In particular, thou I guess slashdot has ad's ( there is a big white space at the top of this page ) I wonder how many slashdot readers actually see them.
Alexa targets a demographic which are more likey to click on banner ads and buy the junk which they advertise. So for the advertisers targeting those demographics I'm sure it works out ok.
You can't take the sky from me...
Of course a site could skew their own results which creates its own problem but would this not at least be more valuable than alexa data?
No, it wouldn't, and you've already stated why. Everyone knows that web site logs are the single most accurate way of measuring web site traffic. And no one uses them anyway - not because they think Alexa collects better data, but Alexa doesn't have a vested interest in making a given site look better than it is.
A system which counts on the person selling to give you an honest evaluation of the worth of their product is never going to be more accepted than one involving a third party.
You're right, however, in that what's really needed is a better way to track visits to web sites. The problem is that we can't trust the buy-in of the owners, because they're (obviously) biased. Also, we can't trust the opt-in of the visitors, because so many of them don't opt-in. So the question becomes, what sources of information do we have?
I don't have an answer to that question. And, based on the lack of third-party ratings systems other than Alexa, I don't know that anyone else has that answer, either.
Reality has a conservative bias: it conserves mass, energy, momentum...
I'd argue it is rather different. TV is one way. Your television browsing habits are slightly less revealing than say, your banking activities or the blog entries you post.
Also, Alexa claims to give you some value in exchange for letting them piggy back on your browsing. Nielsen is more public and more respected. This helps mitigate the sampling problems.
Suck it up and find a better metric for your boss.If his "boss" (or any of the other scores of people who accost him about the popularity of websites) would let him pick the metric, he wouldn't have this problem.
The point of the article is that he has to defend someone else's choice of metric.
Or perhaps, the point is more of an "Ask Slashdot" sort of thing...
As in, "Hey all you /. geeks, what's a better way to do this?" Taco's comments on the flaws in Alexa's system and Control Group's comments on some of the particular challenges against this demographic in general support that.
Heck.. it seems like an interesting enough problem to me, but then again, I don't have a sig like yours:
/.: "Anti-Microsoft Rants, Apple and Google d*ck sucking." Pathetic.If you hate it that much, why are you hanging out here?
(Sorry, I really need to stop feeding the trolls...)
Erskin
geek.
about your S.O.!
File under 'M' for 'Manic ranting'
Besides, can't Tivo gather viewing data, aggregate it and then tell the networks what shows people are watching? Can't Nielsen factor Tivo into their calculations? Shouldn't the Tivo actually mean more sci-fi, not less if the people who own such boxes are more likely to be recording sci-fi on it and therefore tipping Tivo's gathered data in their direction?
Anyway, perhaps the upshot of all this is that maybe Nielsen need to expand out into web browsing. Alexa is just a poor man's Nielsen with no control over who installs their spyware. Perhaps there is a need for an industry trusted source that installs browser monitors on a representative sample of the population to gathers stats in strictest of confidence.
Slashdot is an extremely popular website with great demographics. It should be a huge money maker but it probably under performs.
It doesn't show up all that well in Alexa because very few people that go to Slashdot use or would use the Alexa toolbar.
It probably doesn't show up all that well with the advertisers because Slashdot readers are technically very sophisticated.
What percentage of Slashdot users are blocking the ads on Slashdot? 80%? Slashdot should be the "Myspace" of the technical crowd. Heck it had the friends list long before Myspace was around. We have our Journals "aka" blogs so yea it is a little Myspace full of bright people with money to spend. But it doesn't make that much money. Slashdot should be worth many millions but it isn't. The real problem isn't Alexa but how can Slashdot live up to it's potental for that evil word. Profit. After all I am sure the Slashdot crew would like to make the big bucks.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
And add to this mix that we collectively HATE advertising. So we all use ad blockers, flash blockers, script blockers, image blockers, and anything else we can find which reduces or eliminates advertising which gets in the way of reading the content of a web site.
So even if we do get "counted" and the advertisers can determine what it is that we browse, the current method of "in your face" ads will quickly push us towards a way of either blocking the ads, or simply not going there any more.
And I DO click on ads, but only if they are:
- NOT in the way of the content
- NOT blinking, flashing, moving
- NOT trying to distract my eye towards them
If ANY of the above happen, I am gone from the site, and will NEVER go there again.
(Hey, this is my 1,000th post. Woo Hoo!)
- - - - - - - - - - -
I am a programmer. I am paid to produce syntax not grammar. Deal with it.
Yes the reason digg has a higher page rank on alexia is because the average digg user is almost universally less technically savy then the average slashdot user. No I am not being sarcastic, they are dumber.
A slashdot debate on bush and the war, for example, will use complete word and sentances and sometimes include facts. A crazy rant it might be but a crazy READABLE rant.
The same debate on digg...
bush = leet haxor
STFU, WAR IS BAD, OBAMA 08 WOOT.
This says it all: my work web proxy blocks access to alexa.
Reason - "Spyware Suspected"
I need a new sig...
"I hate to say it, but that really proves not as much that "The only way advertisers can get accurate data as people opt in", as it proves that they have not elected to find new methods to track data properly/independantly."
And one of his points is that this audience would resist it. So no there's no uber-method just waiting to be found. The price for not being part of the system (like illegal down loaders)* is to basically have no influence.
*"Voting with one's dollars" vs no vote with anything.
Sure, but that all presupposes you've already bought ad space on the site in question. When you're trying to select which web sites to purchase ad space on in the first place, you don't have access to any of those metrics. If we were talking about a handful of key sites, that wouldn't be a problem - test the waters, go with what works.
But given the huge number of web sites out there that run ads, you need some way of doing an initial selection of which ones to pay. Hence Alexa.
Reality has a conservative bias: it conserves mass, energy, momentum...
If you hate it that much, why are you hanging out here?
I'm this close to leaving. I keep hoping things will make a turn for the better. But it sure doesn't look like it will.
Quite right. Not Alexa's fault.
As a statistician, I can reassure you that the only thing that's worse than no data is flawed data. When you have no data, you know something is wrong and you start correcting that. When you have flawed data, you don't. Instead you use that data and build on it, never knowing that what you measure, calculate and estimate has nothing to do with reality. In other words, it can be dangerous, to your job and the company you're working for.
Imagine the (flawed) data you have tells you that almost 100% of the people visiting your geek-gadget page are fans of some rock group. Why? Because they use a proxy that was written by some fan of said rock group whose proxy subtly alters the meta information sent by your browser to tell everyone you surf to how much you like said rock group. You analyze it and invest heavily into marketing crap from said group, hoping that your customers will buy it since they all appearantly love that group.
Result? Big desaster. Nobody buys it. Nobody even knows that group. They just all used the same proxy/plugin/younameit, not even knowing that whoever wrote it wanted to advertise his favorite band.
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
You're telling me - rants drive me absolutely nuts, especially on this site. They don't make good reading, they pointlessly waste your time, and they use up valuable screen real estate that could be occupied by other, more interesting stories. The methodology behind rants us usually utterly broken but, for reasons unclear to anyone, are regarded as 'postable material' on all too many sites. I mean, let's not draw the line at Slashdot. Rants show up on:
That we obviously need to abandon rants is clear, because they're almost always pointless, but there are so many of them these days that it gets to the point where the only metric you're using to compare sites is the quality of its rants. This is entirely flawed and meaningless, and leaves me wanting a stiff drink. Still, don't get me started on their frequency on /. You're all Slashdot readers, most of you just go ahead and prove my point anyway.
So say you go to some random site and end up reading a rant. What have you learned. After you close your browser, are you any more complete as a person? Have you grown intellectually. Let me think: no ... no. I'm not some some expert on rants and why I'm writing about them is very confusing, but I think I have as much to say about the dumb things as anyone. And if that bothers people, at least I got the point across.
Here's the problem: rants don't work. If you RTFA, and start with a place I'm very familiar with (namely Slashdot) like a quarter of you write rants anyway. And that's not even talking about the fact that any rant, and not all posts are rants, is going to take up people's time and not get modded very well anyway. How many Slashdot readers would mod a rant up? You either? Big surprise. Because of who we are, and what it is, our population will self select out of consideration.
Did you know rants can get posted by ANYONE? How about Anonymous Cowards? Now personally I'm glad of that, free speech and all. But anyway, those are my (heavily edited) thoughts on this.
If digg is "beating" slashdot, let it win. Maybe the YouTube popularity blog can suck away the idiots from slashdot.
Don't think of it as a flame---it's more like an argument that does 3d6 fire damage
I don't think that your story is a very good indicator of how rubbish Alexa is, it just highlights issues with your own system.
This is why you shouldn't use HTTP GET for 'delete links'. Anything that changes content should be POST, which will stop bots crawling your site just by following links from breaking things. We have standards for a reason..
As for alexa crawling your site as a logged in user, what? As far as I know the toolbar itself doesn't do any crawling, only reporting. Maybe it was providing links to Alexa that later got indexed, but if they were properly secured then you wouldn't have any issues. The fact that you seem to be relying on a robots.txt for security indicates bigger issues. The only time I've heard of a 'Toolbar' doing this kind of thing is when Google released their proxy service (which they later withdrew), as it automatically preloaded all pages - and again poorly designed pages using GET to modify data encountered problems just like yours.
trackware, not spyware, from your link: "is a program that installs a toolbar and gathers Internet browsing and search information." which is EXACTLY WHAT IT IS SUPPOSED TO DO in order to aggregate site popularity.
That's true if the person using the data is unaware that it's flawed. But an educated decision can be made to use data that's known to be flawed, if one evaluates what those flaws are, and what they'll mean to whatever it is you're doing.
In fact, as I think about it, I'm not sure "flawed" is the right word. The information is incomplete; whether that's a flaw depends on whether or not you recognize that you don't have all the information.
I think that assuming all the people using Alexa rankings to make purchasing decisions are stupid is misguided. I think it's a much safer assumption that the distribution of stupid, average, and intelligent people among that population is fairly close to that of the population at large. Many of them are making decisions based, in part, on having information that they know to be incomplete, which they judge to be preferable to making decisions based on having no information.
Reality has a conservative bias: it conserves mass, energy, momentum...
Alexa is definitely not a good way to compare all websites with regards to traffic but it definitely has uses (for us to look at but not to install and help populate). Since the average type of user that visits a website stays constant looking at Alexa stats is a good way to determine trends with regards to content. When the whole HD DVD key fiasco hit Digg you can see the immediate spike as well as the slight depression in number of visitors afterwards. Alexa is great at helping those running high traffic websites and absolutely useless for my blog/blag.
Alexa is also useful for comparing websites that cater to the same sorts of folks. So you should compare Digg and Netscape's version of Digg or compare CNN to Fox to ABC to NBC (with regards to websites) and you should see somewhat accurate comparison numbers. What CmdrTaco is really getting at though is that for niche sites (and it is still niche) like Slashdot Alexa falls well short of capturing true traffic numbers.
internet like monkeys'
i estimate that every time I do this, I burn the same number of calories as I might on an elliptical trainer. I assure you the beer gut ain't getting smaller on its own.
Stress causes weight gain, not loss. It also sets you up for disease. In this way, M$ and other stupid bullies are litterally killing us all. Sudden weight loss or gain is a bad sign. Keep riding that bike.
Friends don't help friends install M$ junk.
I just read a rant.
My irony meter just broke, and you owe me a new one.
Going on means going far
Going far means returning
Equally perplexing is the accounting of iframes. Let's look at someone like . Now it's hard to say, but I don't think I've ever visited their website. Have you? But according to Alexa, they have nearly a 1% share of the internet. I'd tend not to believe it... but they have iframes on zillions of web pages and counting those sure would account for this huge rankin
I think the editor left out some kind of link... that or slashcode ate their URL.
And yeah I also think Alexa is broken... although for me it is completely meaningless. I believe a better ranking would be possible if Google, Yahoo and the likes published statistics of "most clicked page" or the like... at least the statistics would be OS independent.
Ubuntu is an African word meaning 'I can't configure Debian'
Ben Hocking
Need a professional organizer?
Hmmm. Marketing and metrics having a distinct diconnect from reality... For those UF readers out there (and I KNOW many Slashdotters are UFies as well), if you've ever worked with marketing people, Stef will look WAY too familiar. Something being better on technical superiority isn't going to catch the attention of the drones dazzled by shiny, flashy ads, and marketing slogans that tell us that we will be worthless in the gene pool if we don't have the flashiest car, freshest breath, and designer XYZ suits... And those of us in the knowledge professions, are not immune to a certain amount of stupidity. How many times have we performed upgrades on our own software or hardware, not because the old stuff didn't work well, or do the job we wanted, but because the new stuff is, well, new... Now if you'll excuse me, the latest release of my favorite Linux distro is out, and I need to do an upgrade whilst sipping my Frappuccino...
A rant about rants. How about that!
Eternity is a time bomb.
Let me save you some breath, I deal with non-technical small online business owners all day, every day, and I have for the last 7 years - they are obviously concerned with Alexa rankings.
.NET FRAMEWORK.
I *HAVE* been telling them that the stats are bullshit, not only for the reasons listed above but a few others - but eventually I gave up and developed a better strategy:
Don't bother explaining highly technical concepts to a monkey, it frustrates you and annoys the monkey.
If your pointy haired boss wants your Alexa ranking to improve I would suggest you:
1) Call a meeting, invite as many department heads as you can.
2) Make the problem your own, and phrase it as *MASSIVE*, *DIRE*, *EXTREME* (e.g. if we don't fix this, we could all be out of a job soon)
3) Suggest IMMEDIATE ACTION be taken, suggest hiring an offshore team of workers (China $0.37/hr) to install the Alexa toolbar and surf around your site.
4) Recommend that the company consider an immediate payout a Ukranian hacker with mob ties named "Ivan" who will pwn machines and install alexa and then randomly pop your site on his botnet for a reasonable fee.
5) Finally tell them that bribes to key employees in Alexa may be necessary - tell them you may have a contact and tell them to be ready to authorize six digit sums of money in a 24 hour period if necessary. [this can be useful for other reasons]
Trust me - as soon as the first mention of money (and specifically who's budget it will come out of) is made the general attitude toward how important Alexa is will change. They'll backpedal, claim you're being overly-proactive. They'll produce some rant they found on a website called dot-slash saying how Alexa rankings aren't important.
Tell them it's all propaganda, proceed to ignore whatever they say -- pronounce your undying love for Alexa - and it's relevance to the web.
DEMAND THEY RESPECT YOUR AUTHORITY.
IDENTIFY YOURSELF AS THE BIG DOG OF TECHNOLOGY.
ASK WHO ELSE GRADUATED FROM DEVRY LIKE YOU DID?
WHO ELSE IN THE ROOM IS A CERTIFIED NOVELL ADMINISTRATOR?
IF CHALLENGED BY ANYONE TAUNT THEM AND SAY THEY PROBABLY DON'T EVEN UNDERSTAND BIG "NETWORKING" CONCEPTS LIKE SECURE SOCKETS LAYER, TRANSPORT CONTROL PROTOCOL, AND
Then proceed to tell them that (in your professional opinion) your company won't be able to recruit good people because of your poor Alexa ranking. Tell them that search engines will stop spidering your site, and eventually your traffic will drop to zero. Without a good alexa ranking your email will get caught in more spam filters and you'll appear on blacklists and phishing filters more frequently. That means the SSL locks won't show up on browsers anymore. This will cause packet loss on your routers to increase. If it's not fixed immediately it's possible eventually your domain won't even work if somebody enters it directly into their browser. ALEXA IS THE MASTER OF THE INTERNET THEY ARE ALL KNOWING WE MUST SERVE THEM WITHOUT QUESTION.
ps> I *seriously* did have one customer who hired an offshore Indian firm to boost they're rankings (no bullshit) - feel free to mention that your competitors are already doing this, and the clock is ticking. WE NEED A DECISION NOW.
The next topic: PAGE RANK (umm.. wash, rinse, repeat)
Nielsen ratings depended on statistical sampling to measure what your "average"
consumer was watching. The "market" needs to do what Nielsen did and pay consumers
for information. Alexa, Doubleclick, Yahoo, Google and everyone else who seeks to
track your browsing habits do it for one reason: information is marketable and it makes
the consumer marketable.
The average slashdotter probably has the same bag of tricks to opt out of marketing:
All phone numbers on the DNR, one or more DVRs, use of flashblock and adblock, regular
cooking dumping, spam blocking/filtering on e-mail. Different classes of web sites have a
different user id, password and e-mail address if those things are required.
And yet out there somewhere must be slashdotters willing to install a tool, utility, whatever
to track their preferences for cold, hard cash. When marketers figure out a way to shell out
micropayments for each datum discovered along with an enforceable opt-out and expiry system,
let us know; some minority will install it. Until then we'll be the un/underrepresented technophiles.
There are worse fates.
Best rant ever.
+1 Ironic.
Slashdot: where don knuth is an idiot because he cant grasp the awesome power of php
is called compete.com
And it's only a rant because the flaw does not swing in Taco's favor.
As someone who isn't a statistician, but uses statistics on a regular basis, the first question you always need to ask is can I trust the data. You need to run statistical tests on the data to see if it makes sense. You need to understand the problem first, and that data second.
If you can't tell if the data is fooling you you shouldn't be doing statistics on it.
So I guess I'd amend your proclamation to read: "The only thing worse than no data is data that the statistician couldn't tell was flawed."
Hi there. I tend to like reading rants. At least good ones. Rants usually mean that someone is being stupid, and the ranter realizes this. Rant's also usually include why the other person is being stupid and how to remedy it. It's kind of like freedom of the press, but on a smaller scale. If you can't point out someone else's stupidity, then why bother to think that you yourself are a smart person.
To me, most rants are a wake up call by someone who sees a problem clearly.
Make a firefox add-on that collects anonymous info on what sites you've visited. Of course, it must be open source. It simply stores the count of visits on a table, where instead of storing the site's name, it stores the SHA-1 (or to be on the safe side, SHA-256 or SHA-512). (Note: Only the domain is stored, not the full uri). After a period of one week of data storing, it connects to say, "slash-rank.com" (i made up that name) using SSL and sends the data. Finally, the data is deleted from your hard drive. To prevent someone simply bombarding slash-rank with fake data, you're given a ticket and your slash-rank toolbar chooses a random password for that ticket (the password is stored at slash-rank.com and in your local configuration. When the data is sent, slash-rank dot com verifies that your password is coherent with your ticket number. If not, it just ignores the data.
The only problem that has to be solved is how to assure that users aren't generated automatically (captcha?). If you provide real life information, how to assure that the site won't track _YOU_ down? And it shouldn't be installed in servers, because it would become a traffic sniffer - and the iframes problem appears again.
So, accuracy is in conflict with privacy. How to solve this? It's an open question, it seems.
You missed the point - you're only talking about impressions and clicks AFTER the ad buy has been made. Obviously those are easily tracked. The point of "tools" like Alexa is to help the advertisers figure out which sites to buy from in the first place.
no longer working for cnet
So you're saying advertisers trying to reach the affluent trend-setters on the internet should advertise on the sites with the least (recorded) traffic?
Taco's got a case of the Mondays
I have gotten numerous sites into the top 75k of Alexa ratings by simply installing the toolbar on a couple of machines and regularly browsing through the entire site. On the other hand, I have sites that receive 3000 unique hits a day ranked around 300,000 on Alexa. That being said, I still use Alexa all the time to figure out which sites are well trafficked, and I imagineit is far more accurate than the author is giving it credit for. If you eliminate obvious exceptions (sites that cater to SEO folk and sites that cater to certain audiences such as Linux users) I think you will find that Alexa makes for a useful although not 100% accurate tool.
Alexa perfectly illustrates one of my biggest difficulties in dealing with businesspeople - they want numbers, and while they would prefer good numbers, they'll take bad numbers if none others are available. This contradicts my instinct to not report data unless I'm pretty damn sure that it's accurate. That aside, I think that Alexa is misleading, but not useless. The only people who install the Alexa toolbar are the type of people who think that its statistics are meaningful and worthwhile. Thus, it's a good way to tell if advertising and bizdev people are looking at your site, but it's not really good for much else.
to install spyware on every machine we find?
This is blasphemy. This is madness!
Taco is just trying to fudge Slashdot's own rankings in a system he's claiming to hate.
"Emancipate yourself from mental slavery, none but ourselves can free our minds !"
Whenever someone brings it up, just point out that Alexa's demographic is restricted to non-technical users who click "yes" when their browser says, "That's a horrible idea. Are you really sure you want to do that?"
If the advertiser wants to reach technically adept users with a large disposable income, Alexa's numbers will not help them.
Moderating "-1, Disagree" is simple censorship. Have the guts to post your opinion.
Also respond to management requests for information with "The spyware site? No, we're not particularly fond of spyware. Look what it did to SONY..."
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
It's the quality, not the quantity of your audience that matters. Despite the occasional trolling and flaming that goes on at Slashdot, it still uphold its audience as the most informed and highly intelligent. I can't say that for Digg.com.
Life is not for the lazy.
Don't let the door hit you in the ass on your way out.
I find that (from experience working in large companies that make consumer electronics) marketing tends to use this data not to target a product feature/function, but to target advertising. My experience is that they can be quite in touch with our inner geek insofar as convincing us to buy a hunk of junk. When it comes time to where engineering time/$ is spent, it's always with the masses, which we will never be almost by definition. This is how mediocre large companies work.
All I would be doing by installing alexa would be enabling marketing to more effectively bamboozle me. Why would I do that?
this rant about alexa you're commenting under is just another flavor of wanting it all, and expecting to give up nothing in return
slashdot is full of privacy idealists, who regularly rant about intrusions into their privacy... and then turn around and rant about how how inconvenient some things are. here we have a rant about a lack of inclusiveness... surely followed by a round of "hear! hear!" from the some here who will turn around and say "hear! hear!" under the next slashdot rant about universal id cards or bad password policies
"i want total privacy!"... short time later, the same folks: "why isn't alexa counting me!"
pffft. hello? did you think before you opened your mouth? there is a fundamental philosophical tension between security and ease of use. between privacy and inclusiveness. between untraceability and convenience. take your pick. understand why?
dear idealists: you can either err on the side of convenience/ inclusiveness, or you can err on the side of privacy/ security. but you can't err on the side of both at the same time. so pick one that matters to you more: convenience or privacy, and if you choose privacy, the next time something seems inconvenient to you, kindly shut up. and if you choose convenience, the next time something seems like an intrusion on your privacy, kindly shut up. if you still want to complain, then you simply don't understand the subject matter you are talking about, and you sound ignorant and hypocritical
intellectual property law is philosophically incoherent. it is your moral duty to ignore it or sabotage it
It's obvious what needs to be done.
1) Reverse-engineer the Alexa toolbar protocol
2) Write a program which can fake being the Alexa toolbar
3) Raise the Alexa rank of lots of useless sites, like goatse.cx to prove how useless Alexa is
Heck yeah!
I completely agree with you!
No wait -- not really.
It may be that Your PHBs are as stupid as you think, or they may
:/
a) recognize that data with known measurement issues can be better than no data at all.
b) recognize that they are expected to use such data simply because everyone else does (not using it can be just as painful as trying to convince your team to let you code in some other language, etc etc, for your part of a project) and that arguing against it costs political capital and money.
c) just not really care that much because it's nowhere near the top of their priority list
d) not really feel like "wasting your time" conveying/discussing the issue--this happens a lot in situations where management would rather have workers focused entirely on their little cog-like tasks
i am not a PHB, I do have an Masters in Public Administration...and no, I don't work for the gov't.
however, i am frequently surprised at how poorly people rate their bosses on slashdot...do you folks all work for the gov't? (yeah i know that's a troll? All my development experience was in the private sector and I had exceptionally talented middle/upper management...sure they did things for business reasons that screwed the fuck out of the developers, but what they did made sense. Meah. should have gotten an MBA.
The funny thing is, you couldn't even be bothered finding a comparison that wasn't already used in Taco's editorial.
I know, I know... It wasn't worth your time to read his post before you complained. Why is it worth your time to sit here and complain if Slashdot sucks so much? I have to wonder how much different the world would be if people wandered off to find more positive things to do with their time when they stopped enjoying something.
Shinma
"I won't install Alexa; the market needs to adapt"
There's an implicit assumption in your header. That adaptation means that you'll get what you want. Adaptation could either mean the status quo, or a worse solution. So think about that before you tell the world to adapt to you.
I dislike pointless rants as much as the next person, but I feel like you can at least give a little credence to posts like this. I'd imagine it's extremely frustrating dealing with this type of thing; the general reader doesn't really have any idea that this is a problem (I've been reading /. on an hourly basis for the past 4 years, but it' always possible that I just never paid enough attention to hear about this), but apparently it's something that he has to deal with on a regular basis. If he's making a post on the main page, it's obviously something that he feels is a serious issue and he's looking to the community for support and feedback. Rag on Slashdot all you want, but if you're posting on this then you obviously read it a good deal and hopefully get useful/interesting information from it. Why can't a founder of such an excellent (in my opinion) site complain and ask for feedback on an issue that's obviously important and causes serious problems in a way that most likely the users and the admins never anticipated or know how to deal with? Maybe if we all were running 640x480 or browsing with Lynx we could legitimately complain about the post taking up important news space, but in the majority we're not, and in addition Slashdot caters to such a wide variety of readers that you're never going to be interested in every single news post on the front page, even with the customizations. (if you are then I want your job).
On that note, I don't actually have anything to say about the topic at hand, but then again, neither did the parent.
Then don't use Alexa. You can measure traffic in ways beyond toolbars. Try the following:
1) Compete. It is free. It uses toolbars AND panels AND isps. It's not that accurate compared to Comscore.. but maybe Comscore is wrong.
2) Buy something. Comscore uses a panel method with a careful demographic spread so they can extrapolate from their sample with a small percent-error.
3) Buy Hitwise for percentages. It doesn't give you unique visitors, but it can give you comparisons and ranks and whatnot. They lay on top of ISPs and use a few panels. It is 20-30K for a year or so.
4) Wait awhile until the IAB's audit leads to some common definitions and standards among the aforementioned companies. The Interactive Advertising Bureau and Media Rating Council are auditing Nielsen and Comscore to make sure there is more transparency into what defines all these metrics, how they are counted, and how they should be counted, forever after. In a few years, there might be some consistency in the industry, which will at least stop you from comparing apples and oranges once you get beyond over-counting SEO spammers.
If you are concerned about the demographics and unselfconscious web surfing, you need to go with a company that looks at ISP data. That's right, everyone -- your service contracts with larger ISPs allow them to anonymously watch your traffic and sell it to companies like Hitwise with your demographic information. Suddenly, the 35-54 white male demographic with a 80K income in the south can be fully represented in the balloon-popping video site genre, until they start hiding behind a proxy. Because it is anonymous, it is even better than a panel, because they don't know they are being watched and don't change their behavior.
I have never looked at the "Rank" before. This site is about 9 months old.
Alexa shows that the 3 month change is -67% ... a decrease in traffic of about 2/3rds over the past 3 months.
This is ironic, since we had to upgrade servers and disks due to the rapidly increasing traffic load.
Webalizer (with common IP's like development machines ignored) shows a roughly 105% increase in traffic since April of 2007.
The Google Adsense traffic analysis shows a similar uptrend in page views.
But Alexa says the site traffic is down 67%.
And the site is certainly no more technical or cerebral than the average website and receives about 74% of hits from an IE flavor.
For what it's worth ...
There's a neat thing in journalism. Editors retain the privilege of being able to commandeer any space they want in their publication, and say just about anything they want. In the format of Slashdot, the editorial would take the top most position on the page, until a newer story filled the position.
I have been known to do the same thing on my site. It may be a "thank you" to our users. It may be a birth, death, or wedding announcement. It may just be that a particular topic has infuriated me to the extent that I needed to put my opinion in big bold letters on the front page, because no matter how much we may report on the topic, people still don't have a clue about the meaning.
If Cmdr Taco had posted a news story on the poor metrics used by Alexa, would that have received the same attention that his own personal account did?
Unfortunately, he's echoing what many of us already feel. Boss type people feel the need to rank high with Alexa. If the ranking goes down for any reason, they want it brought back up. Even on some of the lesser technical sites, discussions start about spyware, and people start removing these utilities. When that happens, the score for those sites drop, and the score for AOL.com and Disney.com go up. (when's the last time you de-spywared the kids computer?)
I know from reliable information, that my news site is read heavily by those in the intelligence services around the world. I sincerely hope that they wouldn't have the Alexa toolbar on their machines. Most of our users are very aware of what's happening around them. They're the ones that are careful to keep their machines clean of viruses and spyware. That leaves us with the random users who follow links from other news sites, or find us in search engines. Maybe they'll stick around. Maybe they'll even learn something.
Serious? Seriousness is well above my pay grade.
I am guessing Alexa does not show data for its own site because the toolbar posts there. Also, Alexa chart images in other sites might skew. The traffic would be almost 100% toolbar data posts, chart requests, and relatively little actual site browsing.
Yes they could be clever and filter that stuff out. As I said it is just a guess. Good a reason as any.
You sir, have put together a fantastic post. I tip my electronic hat to you.
the average digg user is almost universally less technically savy then the average slashdot user. No I am not being sarcastic, they are dumber.
So not being technologically savvy => dumb? That's a nice way to single out the 99% of the population that doesn't enjoy your hobby and/or profession.
Yes, Alexa is flawed because they just take their numbers at face value, they don't do any statistical analysis of their numbers. Instead of just freaking out as taco suggests, show them numbers from a company that does care about demographic distribution and statistical analysis. Use Compete.com. Here's Alexa's graph for slashdot vs digg using Alexa's metric called reach which is apparently percent of people with their toolbar installed who visited the two sites; and here's Compete's using a statistically computed number of unique visitors in a month to each site. There are other metrics available from Compete as well, but this is the core metric.
We always knew Comcast was corrupt, here's the proof: http://tech.slashdot.org/comments.pl?sid=1909890&cid=34545432
Anyone who has to clean out spam of blogs, and understands why that spam is there probably is a bit like me and considers "SEO" in the same frame of mind as "Spammers" and "Folks deserving of the noose. Opening up 100+ connections to the web server , leaving it nearly comatose in the process , to pump pharma ads is just fucked in the head.
NO SERIOUSLY I DONT WANT YOUR DAMN VIAGRA AD ON MY SITE.
And alexa doesnt help one damn bit
Agent: Alexa
Deny: /
(Or however it goes)
Excuse the Unicode crap in my posts. That's an apostrophe, and slashdot is busted.
I banned Digg from my machine (via my hosts file), and my life has been better ever since. Digg has become a wasteland of blog spam and teenage Bush-bashing trolls. The site is overrun. It's a total mess.
s h&start=0&ie=utf-8&oe=utf-8&client=firefox-a&rls=o rg.mozilla:en-US:official
Look:
http://www.google.com/search?q=site%3Adigg.com+bu
I mean, who would subject themsleves to this garbage?
I think you lost the meaning. He doesn't expect anyone to install it. He asked to make a point.
STFU, WAR IS BAD, OBAMA 08 WOOT. RON PAUL FTW!!11!
That won't work well with ads in general, because your desires change over time. For example, you'll be interested in car ads only when you're considering buying a new car.
Also, people like me would just vote every ad down until we didn't have to see anything. If I want to see advertisements for a product I'LL GO LOOK FOR IT.
Marketing Hype trumps Technological Efficacy.
Wow. Has this ever happened before?
I think you lost the meaning. He doesn't expect anyone to install it. He asked to make a point.
I think you missed the point. He asked people to install it. Admittedly, he said that he didn't expect most readers to do it, but he did ask.
"Accept that some days you are the pigeon, and some days you are the statue." - David Brent, Wernham Hogg
You mean https.
Always was, always will be. After decades, there is no agreed upon methodology for tracking the effectiveness of marketing dollars in the real world. The internet should make it easier, right? Perhaps, until people learn how to filter the internet. Doubleclick never sees me, because I have
in my hosts file, along with 37,000 other crap sites. I also add "*urchin\.js" to my custom filters in FilterSetG, so AdSense doesn't see me. I suspect other Slashdotters take similar measures.
If a good click through rate on a banner ad is less than 1%, and only about 1% of clicks result in a sale, then the value of that banner to the advertiser is only .01% of it's cost (yes, I know AdSense works differently, but it has its own pitfalls). Pathetic, isn't it?
It makes you wonder how poorly traditional media ads actually perform.
Banner ads, I'm pretty sure, are the first time advertisers have ever been able to measure the returns on ad dollars. Some company spends $20k for a full page ad in a magazine, how much of that came back in sales? No one knows. So just to me sure they don't lose sales, the company continues to buy ads, following some rough percentage of revenues. Demographics is the closest thing marketers have to concrete data... it basically says not to buy ads in Ladies' Home Journal if you're selling vintage car parts. Even then, demographics measures potential returns before the fact, not actual returns after the fact. So, advertising is a wild goose chase based on assumptions, and no one does, or can, really know what's going on.
The internet should be a wake up call for advertisers to the fact that their marketing budgets are being overinflated by... (wait for it) the ad agencies and marketing firms. Sadly no one will realize this, because the foxes are in charge of the henhouse, and claim everyone will fall to ruin otherwise.
Generally, people don't want the crap in the ads, and would rather not even see the ads. Horrible conversion rates prove this. The scariest part of Minority Report, other than the nanny-state concept of "pre-crime", is the level of advertising present everywhere in the film, targeted at individuals with laser-like precision. It got that way because the public allowed it to happen.
The simplest way to fix advertising is to remove all imperative and presumptuous statements from them. No more "Call now!", "You need...", "But wait, there's more!" obnoxious mind games. I'm not calling, I don't need your shit, and I'm not waiting for you to yell at me some more.
Anyway, I took a quick search and found some more ranking sites you might want to look into:
Quantcast (open)
Nielsen (commercial)
Comscore (commercial)
Hitwise (commercial)
Oh, and by the way...
like 80% of users run IE. It doesn't really matter which browser-statistics site you are looking at; it's kind of ~33% IE, ~33% FF, ~33% other. Well, at least I hope it will be this month or the next.
Why, yes, I'm already preparing to celebrate the victory of Firefox in the holy browser wars. Yeehaw!
Aren't you paying per page view anyway? Doesn't that mean that any popularity rating based on page views can be factored out?
-- The act of censorship is always worse than whatever is being censored. Always.
so in that case.... http://data.alexa.com/data?cli=10&dat=snbamz&url=h ttp://www.slashdot.org
might work. Come on and follow the link so Taco gets happy :-) (make sure to use a safe browser please, I cant guarantee that its a safe site)
As a statistician, I can reassure you that the only thing that's worse than no data is flawed data. When you have no data, you know something is wrong and you start correcting that. When you have flawed data, you don't.
This is a huge assumption that I'd say is incorrect more often than not.
Your entire argument as it stands now presupposes that the advertiser doesn't know the data is flawed. But what if he does?
My company buys lots of web ads. We use Alexa as one of our data sources (not the only one) to determine ad buys, both because it's free and because in our experience, its data is no more or less accurate than that of paid vendors like Nielsen. Do we expect 100% accuracy? No. Do we think we can learn anything if, for example, it tells us that two directly competing sites have traffic that's different by about 200% in every metric? Probably.
Buying ads is not an exact science. It doesn't really matter if we get accurate traffic down to the individual click. All we're looking for is relativity - a site's size and reach compared to its competitors. We look at the sites themselves, we look at Alexa and we look at research that we commission and pay for. Usually these sources all agree and we go ahead and buy. In the event that they don't agree, we use our own critical thinking and our own judgment to determine what to believe - that is part of any marketer's job, after all.
It seems to me that this whole article here is missing the point. Alexa's a tool. A free tool. It is useful at what it does, but it is not, nor was it ever intended to be, some sort of accurate measure of site statistics for the entire internet. Nobody who uses it as part of their decision-making process is using it that way.
I think this is a case where somebody looked at Alexa, figured out that it wasn't perfect, and therefore determined that it's utter crap. That's basically what your argument boils down to also. But the point is we don't need perfection, and we don't expect perfection, and this lack of perfection is taken into account in our decision making process. We're not flying to the moon here; we're buying ad space. It's something of an organic process regardless of how good your data is.
If you're talking about somebody using Alexa for their own site, then that's just ridiculous. Even cheap hosting accounts (like I have for my personal site) come with their own log-based stats, and if not, there are plenty of free services like Statcounter out there. I don't think this is what many people use Alexa for, though; it's used more by small to mid-sized companies looking for sites on which to buy ads, or by curiosity seekers who just want to see how big their favorite sites are. I would think most sites would know what their own internal numbers are one way or another, without Alexa.
More Ronald Pink-Pick-Aisle articles, of course!
Oh... Wait... You meant more profit for Slashdot...
Never mind.
the url is (from the firefox toolbar's plain text javascript)
http://data.alexa.com/data?cli=10&dat=ns&url=$url
Don't let the door hit you in the ass on your way out. :)
by Anonymous Coward on Monday July 23, @12:10PM (#19958213)
How would you know? You haven't even come in yet!
The offtopic troll seems to be your new favorite M.O. Here for example. Any particular reason other than to show everyone how proud you are that you've discovered photoshopping?
Alexa is fine. It's flawed, everyone knows it. If it's your job to report Alexa rankings then you should provide that information with a disclaimer. Or find an alternative solution. What's not ok is to avoid using Alexa due to its shortcomings or not educate someone on its shortcomings when given the chance. As they say, there are lies, damned lies and statistics.
A teeny bit OT but
It's no different, really, than stating that populations who do not vote self-select against being represented in government.
I would argue that populations (in the US at least) who do not contribute millions of dollars to the major party candidates self-select against being represented in government.
That great American corporation Sony, for example, can give ten million bucks to the Democrat Senator and nine million bucks to the Republican running against him (or vice versa if the incumbent is a Republican) and be guranteed representation. Sony's lobbyists will be invited to the post-election Senator's office regardless of who wins.
Meanwhile, all you can do (in the words of Mojo Nixon in the song Burn Down the Mall) is to "vote for one fool or another". Good luck getting your sorry ass into Senator Dollargrabber's office to hear what YOU have to say! You'll be lucky to get one of his staffers to read your email.
I see two badly needed reforms that don't have a snowball's chance in hell of ever getting passed.
First, the above mentioned practice of "donating" to more than one candidate in any given race should be a felony punishable by prison time, as these "contributions" are barely disguised bribes. Is any honest person actually FOR legalized bribery?
Second, it should be illegal to donate to anyone you're not eligible to vote for. I live in Illinois. Why should Bill Gates or Larry Ellison have greater access to MY "representatives" than I do? He shouldn't be allowed to contribute to any candidate that isn't on his ballot in Washington state.
1st amendment? Giving money is "speech"? What kind of doublespeak is that???? A donation is NOT speech any more than a 170 year copyright is in any way "limited time", no matter what the Supreme Court says.
I have no representation whatsoever, and neither do most of my fellow citizens. The Bono act is IMO a sad crime against the populace in general and creative persons in particular. The bankrupcy "reform" measure hurts citizens while helping usarious credit card companies. I want drugs, prostitution, and gambling legalized. What candidates support MY views? I have no representation, yet I vote in every single election. If I want to vote for a candidate who has any chance at all of winning, I must vote for a candidate who is guaranteed to vote against my interests.
I've finally come to the conclusion that I should never again waste my vote on a Republicrat. A couple of elections ago I started splitting my vote between the Greens and the Libertarians. Like I said, I want drugs, gambling, and prostitution (when was the last time YOU got laid?) legalized and so do the Libertarians. OTOH I'd like to see a curb on corporations spewing poison into my air and water, and the Greens are with me there. The Republicrats give lip service to liberty while writing laws restricting it, as they give lip service to the environment while giving their corporate sponsors loopholes allowing them to fuck up my world.
They say folks who don't vote are apathetic. Well lets see, the Republican wants marijuana to stay illegal, but the Democrat, on the other hand, would rather marijuana to be against the law. Who to vote for??? Why shouldn't they be apathetic?
Sorry for the kinda OT rant but the shit just pisses me off. This stuff bothers me a lot more than my web site's Page Rank. Which, uh, well, isn't that a fat better indicator of a site's popularity than a tool that you have to be dumb enought to install spyware to use?
-mcgrew
When talking about technological things like page-ranking and Alexa's use on that, yes. Yes they are.
It's not that they are dumb in the wide version of the word, but in the techfield, Digg is arguably "dumber" than Slashdot. Try the same argumentation on Slashdot vs. MySpace.
When I talk about my hobby or profession, I like to single out the 99% that doesn't understand a word from what I'm saying.
Free beer is never free as in speech. Free speech is always free as in beer.
you have to take it with a grain of salt.
where it IS useful, is if you want to compare 2 similiar sites with similiar audiences. For example, using alexa to compare digg to shoutwire would provide an accurate analysis of popularity.
Why the hell would I install a toolbar I don't need, for a crappy search engine I don't use, without benefiting in the process? Would they be willing to pay me to install it? I install what I need to use on my computer. Hell, I like and use Google and I don't bother installing the Google toolbar because I. don't. need. it.
The Christian Right is Neither (Christian nor right). See: Matthew 23, Matthew 25, Ezekiel 16:48-50
"Despite the occasional trolling and flaming that goes on at Slashdot, it still uphold its audience as the most informed and highly intelligent. I can't say that for Digg.com."
Oh Lord. Just because your better than Digg doesn't make Slashdot that much better. Trolling and flaming are the least of this forum's problems.
I saw a teenage goth today. The whole ensemble was only slightly marred by the brightly coloured Noddy backpack. I think she was being ironic or post-modern or something.
I magine someone's already doing this.
It's no different, really, than stating that populations who do not vote self-select against being represented in government.
Don't some (healthy civilized) countries require by law that you vote? I think that's actually a very good idea. We should also declare election day a national holiday. Of course one could make the argument that if a person doesn't want to vote we don't care about their opinion. I offer that philosophically such an argument goes against the very basis of democracy (read read "Wisdom of Crowds") and I can further use our current messed up government (which everyone seems to agree on) as an example that we certainly haven't achieved greatness by excluding those who are apathetic, busy, disenfranchised, etc.
Anyways, just some thoughts.
Surveys are useless. You take part of a target group, figure out what they're doing, and assume everyone else is the exact same way. Alexa is just as useless as those political opinion surveys where they call up 1,607 drunk housewives, basement dwellers, drug dealers and homeless pedophiles and ask who they'd vote for tomorrow. Junk data. GIGO
technical writing / development
The obvious answer being that you don't get to reduce the number of ads, just choose which you don't want to see. The more you mark down, the more obscure the ads that it shows you next time.
If you just followed the link in the article, or typed "alexa" into your search, you'd have known within a few seconds.
I surf the web via telnet. Is there an alexa toolbar for that?
I mean, Billy Joel wrote a song about it, right? http://en.wikipedia.org/wiki/Downeaster_Alexa
Saying your "phone ran out of batteries" is like saying your "car ran out of gas tanks".
A rant about rants. How about that! /. page hits (by reading the ensuing non-discussion) we just propagate the problem.
By replying to these articles and by giving
"but money is the God of Algiers & Mahomet their prophet." - Rich. O'Bryen June 8th 1786
CmdrTaco wrote:
:)
Equally perplexing is the accounting of iframes. Let's look at someone like double click's alexa rating. Now it's hard to say, but I don't think I've ever visited their website. Have you? But according to Alexa, they have nearly a 1% share of the internet. I'd tend not to believe it...
That's not surprising to me at all. I don't think this is because of all the iframes that pop up on pages, or they would have a much higher percentage than 1%. I think it's actual ad clicks. When you click an ad, you go to a doubleclick link which will redirect you to the advertiser's page. If all those ad clicks are counted as actual traffic, 1% is actually a very believable figure.
And I've never heard of Alexa until now
m
Star Trek went for three seasons. Its successors TNG, DS9, and Voyager went for seven seasons each and Enterprise went for four seasons. Stargate SG-1 aired for something like ten or eleven seasons and the spinoff series Stargate Atlantis is still soldiering on. neo-Battlestar Galactica will be entering its fourth season later this year. Seaquest made it for ~2.5 seasons. Babylon 5 lasted for five seasons. Andromeda lasted for four or five seasons.
At my last company, we knew exactly how many Alexa users there were per day, because all but 2 or 3 of them worked at our company! We just had to compare the traffic for an admin-only page to our frontpage.
One of us installed it just to see what was up, and a few days later I stuck it on a test box and our ratings increased by 25% the next day. After we realized our market(people who build electronics from scratch) wasn't likely to install Alexa, we just installed it on a dozen of our coworkers' computers and our ratings skyrocketed, especially in our "market group".
As long as Alexa relies on self-selecting panels where the selection criteria is "computer illiterate" and/or "SEO manipulator", it will always be useless to any site that attracts computer literate users.
If the site says "We'll get you N page views per week" for X period of weeks, and the first week comes in at N/4, you cancel the rest. Anyone relying on Alexa rather than demanding a performance guarantee is just being dumb.
Kevin Smith on Prince
" Agent: Alexa
Deny: / "
why not do a a mod_rewrite rule for Alexa, sending them back to alexa.com, or better yet, the DHS. Be nice to see Alexa classified as a "terr'rist". (yeah, I know there are problems with this theory, but we can dream, right :-)
Kevin Smith on Prince
interesting way to put it, but yes!
Seriously, do you want everyone who allready knows about your product to see another anoying comercial? Or would you rather want to reach out to those people who have no idea you exist, and inform them of your wares?
How much is your data worth? Back it up now.
At one time you could install Windows on a freshly formatted drive, then install a decent spyware remover from cdrom and it would detect and recommend you remove Alexa, all without ever connecting to the internet. Alexa was incorporated into Internet Explorer then without the obvious toolbar. I have no idea if this is true for more current versions of Windows and Internet Explorer or not, but it was simply part of my set up procedure to remove it back when I reinstalled Win9x for many who couldn't afford to take it down the road to the nearest computer repair or just trash it and buy a new computer.
"And if you want to help me prove this, please install their toolbar. Of course since most of you are Slashdot readers, most of you won't and that only helps prove my point."
So basically no matter what you do, he takes it as evidence supporting his point of view!
...and who have an above-average impact on what other people will buy....
while I agree with most of your post, that's a questionable statement unless you're referring to inverse impact. While "our" purchasing choices are certainly bellwethers to a class of goods, the actual products that we buy are almost never the ones that see wide commercial success. iPods and Tivos are both excellent examples of herd products that were dismissed as junk by the geeks until the leaders we bought are no longer available in the market. On a broader scale, think laser disc versus VCR.
"Nothing was broken, and it's been fixed." -- Jon Carroll
Yes, but assuming you host the ad content, you can look at your logs to see how many page views you've got (as well as how many clickthroughs). I'm pretty sure - not certain, but pretty sure - that's how the payment structure is set up. Which, again, counts on the ad already being put out on the site, and does nothing for predicting which sites would be good bets to put the ad out on.
Reality has a conservative bias: it conserves mass, energy, momentum...
The problem isn't the data Alexa generates. The problem is what you expect to learn from it.
Alexa is not a list for overall internet popularity - it is a list for internet popularity among those who are stupid enough to install spyware on their computer, as you have said. The purpose is to rank popularity among the ideal consumer.
You have to face it, Slashdot users are not ideal consumers. We aren't swayed by advertising. We try to solve problems with ingenuity before money. We think before we act. We do not buy garbage.
Alexa is important to internet advertisers who are willing to pay money to put their ads on your website. If Alexa provided a 100% accurate model of internet popularity, then advertisers would find it less useful, as they would no longer be able to identify where the most impressionable and uneducated consumers went.
If you really want good Alexa ratings, just put a link to the toolbar at the top of slashdot.org. Soon you'll probably be in the top 20.
The thing about rants is that they're noisy and accomplish nothing. If /. were a car, then rants would be like a loose, rattling door panel or something. That's why we need less rants and more analogies. Analogies are more like the engine of the /. "car" because they drive the discussion forward. The best analogies are about cars, because cars are easy for everyone to understand. Car analogies are like a sweet fine-tuned V-8 engine, and they make /. perform like a supercar.
Sparky. It's called Sparky. http://www.alexa.com/site/download
i find alexa useful to get a rough idea of how much traffic a site is getting. i'd say its rankings have one or two sig figs, but often that's enough... sometimes all you need is an order of magnitude - so while everything CmdrTaco is saying is basically true, the fact that a site has an Alexa rank of 500 or 5000 or 50000 or 500000 says something about how much traffic it gets ...
Wouldn't that make this a meta-rant?
This sig is exempt from disclosure under the privacy Act of 1974.
Go ahead, label it as flamebait.
Somehow, I doubt that that would happen if it had been Bill Gates or the CEO of RealNetworks who was asking people to install spyware. Talk about double standards.
I guess Slashdot still hasn't grown beyond being CmdrTaco's little personality cult.
"Accept that some days you are the pigeon, and some days you are the statue." - David Brent, Wernham Hogg
I agree that Alexa is completely flawed. I did find a way to make it work better, if you load their widget on your site it will calculate your traffic (Example: http://www.lotrofaces.com/ - all the way at the bottom). I know the operator of that site and when I told him about adding that it bumped him up in the ratings. Slashdot or your site might not want to do that but it does make it more accurate for your site. If slashdot did that I'm sure they would jump up to #1 or at least to top 10.
The Netcraft toolbar isn't quite as badly flawed, but it is still flawed.
I found out that having a web site with a 3 second refresh (it was a camera pointing to the back of my house that I was using because someone stole the wheels from my car), I was able to get my website into the 7th top-most position. This was a server that I was the only person going to.
Then, they found out, and removed the server. It's still black-listed on their searches (you can't get any information from it). Somewhere, I've got a screenshot of the Netcraft site which shows there was no 7th place site (it goes 1,2,3,4,5,6,8,9,10...).
They might have changed the toolbar by now...
you can analyse the log files and factor out multiple views per host ip to get the actual number of real views
Sigh. No, no you can't. A hundred different people can see your website without ever making a single loggable request. Or one person can appear to be coming from a dozen IPs. This is the case with common configurations like load-balanced proxy caches, not abuse or pathological cases.
When are people going to realise that HTTP doesn't work in the way that they want? You can't count visitors or page views by looking at traffic. You're as guilty of snake oil marketing as the Alexa stats that Taco complains about.
The real problem isn't that people want to measure stuff like this. The real problem is that the people selling this snake oil bury their heads in the sand and pretend that if they want the stats to be accurate, they can use the "oh, well it will only be off by a little bit" excuse.
Rant or not, he's absolutely right. I, for one, record many an episode of Futurama and any other number of shows and I constantly wonder if my DVR actually counts towards the ratings for that show. Not that it matters for Futurama at this point (well, not yet), but it's been on my mind.
It all comes down to the fact that eventually someone was going to have to point out the problem. Otherwise, how would the problem be addressed?
And as far as everyone's problems with "news or rant," the root word of editorial is, well... That's an easy one. And isn't Slashdot actually a blog in disguise, anyway?
Your ad here.
Along the lines to of your concern
I wonder will any service be able to ensure that a person actually and intentionally viewed a site/page they wanted. Since beyond technical issues at the core, this is more a matter of a persons morals and standards than anything else.
for example: ( might be urban myth ) I was told of a family that had a Nielsen ratings box, that would often turn the TV to a show and go out for dinner. Since they were expected to maintain a certain number of hours TV viewing and after a while... they just did not care, but wanted to remain in the program.
Do you let yourself be affected by advertising? Surely not, you are one of the elite. BTW, what brand of mp3 player do you own?
Due to circumstances beyond my control, I am master of my fate and captain of my soul.
Untrusting conspiracy theorists and logical thinkers are not who are typically targeted by mass media advertisers. I think many of us are excluded by design.
Cheers
Check out my lame java blog at www.javachopshop.com
If by "put up with" you mean putting up with their mere existence, then I guess I see what you mean. Otherwise, nobody's making you visit them or anything, right?
Please don't use "umm" or "err" or "erm".
that is a great theory ... someone mod this guy up
so what if the advertisement is an email? Do you host an ad server and get a notification of the users clicking/following the link out of thier inbox/# of opened ads etc? How would the advertiser technically speaking get that information?
Nielsen has been recording metrics for PVR users for years. Their software integrates with TiVo, and they provide "Live + 7" ratings that include viewers who view the program timeshifted.
Stop complaining. Recording user metrics is an imperfect science. Alexa cannot ever hope to eliminate all bias in their sampling methodology. They do the best that they can.
As far as the reporting mechanism is concerned, it is incapable of distinguishing between slashdot.org on the Internet and an identically-named, identically-numbered machine on a private network. As the fictional users are randomly changing, along with any registration info, from a pool that was collected earlier, there would be no possible way for the stat collecting firm to identify which users are real (on the real Slashdot) and which are fake (on the fake Slashdot).
What's more, because these are done on spare boxes, there is no overhead placed on the real site and the real servers are not burdened with pointless traffic.
Once you get an idea for how much fake isolated hits you can get, you then sell a hit-booster kit to other companies. They get a similar setup (fake server with real name, and a stress-testing platform with fake Alexa users) at so many dollars per fake increment on their stats. It should be a good money-spinner, if the unit increments are large enough.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
He's got a point, mate. ;)
Erskin
geek.
This is for CmdrTaco and anyone else who wants to read it.
Dude, it's the paradigm that sucks, not Alexa per se. Consider Nielsen ratings: would you or any self-respecting Slashdotter actually be so foolish as to agree to be a "Nielsen family"? I doubt it. It's the same dynamic at play. I blogged about the relative stupidity of Nielsen families in particular a while back; those people are ruining my ability to enjoy quality programming like Firefly, Space: Above and Beyond, Keen Eddie, and countless others because of their mindless plebeian tastes.
These are also the same people who often cause unreasonable pricing for consumer items, because they're too stupid to know when to vote with their dollars and just say "no". "$70 for a set of warmed-over LucasFilm Star Wars films that already turned a profit three times over? No problem, I simply *must* have them!"
As a result, manufacturers set prices based on this same mindless demographic; those of us who are "smart" consumers, who could wrangle a better fairer price, are dragged along for the ride kicking and screaming.
That's kinda what has happened here: you (CmdrTaco) are being dragged along kicking - and screaming - by all the Alexoids, and you don't like that any more than I like having Firefly yanked off the air.
I'm quietly of the suspicion that national and especially online advertising is only a fraction as profitable as corporations think it is. I suspect if someone could do a truly objective cost-benefit analysis of mass advertising, like car commercials on TV, we'd find that it's actually costing money that is never rewarded in equivalent sales, and for which we're all ultimately footing the bill in the form of higher prices to pay down all that pointless advertising.
Solving the "Alexa dilemma" just might require eugenics or some other speciation event.
If the masses can keep you down, you're not the Ubermensch.
This is just a flamebait shameless plug. The poster knows that the best wait to sell to slashdot users is an insane flamebait like this. All I can read is "Install Alexa", "Download Alexa" and "try it out". The only thing he is criticizing is the lack of users Alexa has... wich of course will increase some few thousands after this slashvert.
I hate to break it to you, bud, but unless you're a Nielson family, you won't count whether you DVR it or watch it.
Techies are known for a negative attitude toward advertising. I know I always block all ads on every site. If I see an ad, I automatically right-click and block it. If an advertiser does manage to get my attention after all, I am going to hate him so much for it that I wouldn't buy the product no matter how good it was or how much I wanted it. My guess is that the majority of people here feel the same way. We hate ads. We hate spam. We hate being sold to, period. And this isn't because we don't buy anything; we do. I buy lots of things all the time, but I will never base my decision on an advertisement. I'll do some research, decide what I want, figure out what quality level is acceptable, and then find the lowest price that fits those criteria. As a result, I am not a target demographic for advertisers, because to send me ads is counterproductive. Hence Alexa, which, by its method of measurement, naturally excludes the people like us, is a very useful tool to give advertisers what they want. It's that simple.
I'm in the same boat. Never heard of it until I read this article. Although it's true that I'd be unlikely to install it because of privacy and security concerns, that isn't the reason I haven't installed it. The reason I haven't installed it is that I had no idea that it existed.
Now, I'm not a web professional. Websites appear nowhere in my job title, and no one makes money off of what I do online. But, I'm a lot more knowledgable about general internet and IT topics than just about anyone I know.
So, if a group of young, web-savvy, tech hipsters like us have never heard of it. . . who has?
(I'll leave aside the question of whether one ough to be able to read a 12 paragraph essay intended for a general audience about a thing without finding out what it is in even the most basic sense. I gather from context that it's some sort of opt-in, client-side, website visit counter. Can't imagine why on earth anyone would choose to install such a thing - all privacy concerns aside, what's in it for the user?)
He was on reddit...
Send email from the afterlife! Write your e-will at Dead Man's Switch.
He claims he can't do what he needs to do without it. Our AV software, NOD32, flags it as spyware, blocks, and tries to delete it. Because he's such a whiny little bitch and he already has enough crapware on his laptop that makes it unresponsive for at least 5 minutes after logon, not exaggerating here, we went ahead and made an exclusion on his install. I don't know exactly what he uses it for; the best I've been able to decipher from his "George W.ing" description is he uses it to look up information. I refuse to install it on my computer and really haven't gotten around to setting up a test system to figure out what vital, apocalyptic-preventing information it gives him. We've told him it's spyware, but he just ignores us like the peons he thinks we are. Funny thing is he gets a stern talking to from his bosses whenever he's a real ass since my boss is the NetOps Director and his boss is the CTO, C-level beats president. One of these days he's really going to fuck up and we'll standing with log files in hand with grins ear to ear, which might be too soon with him doing to non-PCI compliant crap right after we have an audit.
1. Our tastes/needs change with time. While you're shopping for a new car, car ads may be interesting to you but after you but it, do you want to see another loud, screaming guy yelling he has the most cars at the lowest price? But, if you say, "not interested," at that point, you won't be targeted for new car ads again, even though you'll likely return to the car market in 2-4 years. You'll want to see the ads then, maybe, but ... too late.
2. Commercials are often the funniest, most original content on TV these days. When they are new, they are worth watching, regardless of whether you ever plan to buy that product. The real problem IMHO is the darn things just get overplayed until we can't stand to see them again. And there's no sense expecting companies to limit runs of their ads since they cost so much to produce they must need them to show a lot to amortize their production cost.
Not that I am agreeing w/ "the least (recorded) traffic?" bit. You have to have at least some comercial quality traffic. But really, if I must see adds, at least show me something I haven't seen yet. Something new. Instead of ... oh wait, I hosts out my adds... nvm
How much is your data worth? Back it up now.
Not that it isn't game-able or the ultimate solution, but Project Wonderful actually lets people bid in real-time on ad spaces, so the price per page can actually vary (and be surprisingly affordable). So the "pay a set price for x impressions" model isn't necessarily the only way to go.
I'm glad someone's finally calling bullshit on Alexa
DAILY ROTATION
"Also, people like me would just vote every ad down until we didn't have to see anything. If I want to see advertisements for a product I'LL GO LOOK FOR IT."
What product? Oh wait! You didn't even know "The Product" existed.
What the hell is Alexa?
Even it's website is a little shy on answering that question.
(Or have I just made Taco's point?)
If they did as I pointed out, the problem of caches is fixed unless the caching mechanism is broken. Putting parameters after the url is the usual way its implemented, and it tells the caching proxy that this is NOT static content, and to pass through the request instead of using the cached object. If they don't re-fetch your ad, you don't pay, since if they don't fetch it, they never see it.
Also, it doesn't matter if the average user cycles through a dozen IPs, over the course of a day - if they view your ad once, because it will come from just one of those IPs, not multiple IPs. Of course, you can also just look at the different IPs the various requests for different parts of the page are passed through, and build up a map of IPs that all front for the same user.
No solution is perfect, but this sure is better than nothing.
Kevin Smith on Prince
Or serve the image from your own server, and parse your logs for requests.
Or host the text ads yourself.
Or make the text ad a block of javascript that goes to your server and serves up custom ads, like the "big boys" do, and embed it in the html of a web page. There are dozens of ways - just ask any spammer, they have LOTS of experience with this sort of stuff.
Kevin Smith on Prince
While your idealistic approach to advertising may sound logical, there are many reasons why they would never be adopted in today's advertising market for a variety of reasons - at the top of that list would be trust and laziness.
." This rarely happens on the web because (a) publishers don't deal with advertisers on a one-to-one basis and (b) advertisers don't handle marketing campaigns on a site-by-site basis. Both parties rely upon middlemen such as DoubleClick to manage their advertising media. Publishers do so because it allows them to maintain a manageable, consistent, and trustworthy relationship with one vendor to handle their advertising placements on their sites. Advertisers do so because it allows them to reach many publishers without having to establish each relationship and it allows them to host advertising media in a reliable and secure environment. Most of these middlemen offer very detailed aggregations of data on the performance of these ads in each placement that relies in no part on Alexa or any such tools that are being discussed in this thread. I work for one of these middlemen, so I know a lot about how these things work and where everyone is making their money.
Take suggestion #1 for instance: "1. as an advertiser, host the ad on your own server, and just look in your logs
Why I'm considered an expert is very confusing
I'm rather puzzled by this, too.
Your rant might be impressive Taco - if it contained at it's core an argument stronger than "the cool kids won't let us play with them!".
In that vein, had I (a Firefox user since way back) been in charge of Alexa, I wouldn't have worried or hurried to get a Firefox version out. It wasn't until fairly recently that Firefox numbered a significant portion of the overall internet userbase. The same things goes with your arguement about some random website outscoring Slashdot - Slashgeeks are an infitesmal slice of internet demographics.
Have you met my penis bird?
Humpty Dumpty was pushed.
I agree with both posts above.
./ often. Why? Because I learn a lot from it as often other postings have nuggets I treasure. Errors are pointed out and corrected and so I am confident that I am well informed.
./s rating site (and this discussion thread), to show why they are a good site to advertise on. It is not the number of clicks that counts, it is the influence that the site has.
:) Slashdot is still an a site that is very influential.
The marketeers would like to track us. Why? So them can tell which sites we visit and
and so place ads in our path, and reward the site that we visit with higher rates.
As already said, we avoid spyware, with good reason, and value out privacy.
But we can help the marketeer and our fellow customers and the sites, and ourselves by identifying which sites we use in collecting knowledge. Such a scheme will ensure that our loved sites stay around.
I read
So let's us help identify the sites we visit. Some sites I only visit when I want to buy certain items e.g. Steve's Digicams Digital camera reviews www.steves-digicams.com/
and Digital Photography Review www.dpreview.com. Both have helped me buy cameras.
My suggestion to Slashdot: Make a site to which we add our recommendations and comments. Selected by product, service etc. This way the sites can refer to
In spite of being REALLY OLD
The calibre of the posts is one reason why it is successful -- great discussions!!
Hope this helps Thank you slashdotters. Now! Improve on this idea!
I just wanted to throw in my 2 cents and say this was the best content I've ever read on /. in coming here since it was launched.
I'd love to see more exclusive content like this, whether by CmdrTaco or other intelligent editors.
And having a similar job as you, CmdrTaco, I couldn't agree more.
And also, I have a site that is ONLY included in 302 redirects (a "bounce" domain for stats tracking) and it says we have 4x more reach than Slashdot.
The more I use Amazon's recommendation system (and I've been using it ever since it came out), the less and less I like it.
And then if I click on 'not interested' (in the lightweights) it merely moves the nearly endless spew of cheap crap cookbooks further up the recommendation list, and eventually stops recommending food and cooking books altogether.
When did rate the duplicates highly, they so choked my recommendation that literally none of the first 150 recommendations on my list was anything but connected to those two authors!
But self voting on ad content would be flawed too by savvy consumers. I'd vote no on all the toy ads on saturday morning tv and I skip most of the ads on stuff targeted at me as I almost never watch live tv anyway.
The point is that Alexa is flawed, without a doubt. But it seems more flawed from the point of view of a group which deliberately makes itself all but impossible to measure. And frankly, if we're not willing to provide the information necessary for advertisers to make informed choices, we're going to continue to be on the web and on television.
Well said. But who cares?
Really, what difference does it make? That I don't see relevant ads? That's somehow important to me?! I see almost no ads at all: I use a well-equipped copy of Firefox, and I subscribe to Netflix instead of cable or satellite. Neilsen hasn't been relevant to anything I care about for years. Advertisers? Give me a break. I buy my music from eMusic.com or independent CD resellers. I haven't bought more than a couple CD's in a brick-and-mortar store since Tower closed its doors, and probably hadn't bought more than 20% of them from major retailers in the past 8 years anyway.
There are few people I care about less than advertisers. They have no need of me (it seems) and I certainly have no need of them. If they can't make money off of me that's somehow _my_ fault, and _my_ worry? I'm sorry, I've got more than enough to do and worry about. It's a new world... adapt or die.
You are in a maze of twisty little passages, all alike.
At W3C we log almost everything as well, and we end up with way too much data as a result.
But we use the logs to detect and prevent certain classes of abuse as well (e.g. too many requests in a short time interval or re-requesting the same resources over and over), and we also want to be able to track trends over time, so we have been reluctant to just throw that data away.
I have a plan that I have yet to implement, which is to log only 0.001% of the requests for certain very popular resources (e.g. HTML DTDs and valid-HTML icons), which would allow us to monitor trends without logging tens of gigs of data per day; we'd just need to compensate for it when calculating stats later.
Then I planned to monitor for abuse by also logging every request to a script that watches for abusive traffic patterns, an easy adaptation from the current script that wakes up and skims the logs every 10 mins.
(in your journal entry, when you say you are MD5ing IP addresses for privacy reasons, are you adding a random bit of data to the IP address before calcuating the MD5? If not it's pretty easy to find out which IP address corresponds to a given MD5 sum.)
With such a system in place, I would probably end up seeing only ads for Head On (Apply directly to the forehead!!!) during commercial breaks, which would cause me to laugh uproariously until I went completely insane.
"And give everyone the power to moderate and you have people burying others' comments because they disagree with them."
Uh, this happens all the time on slashdot. I have all the "negative" mods set to +6 because there's more insightful and interesting posts that get buried by malicious mods than there are actual trolls and flaming. And because of the default "ignore AC" settings, some of the best posts ever made on slashdot have sat unread at 0. I don't think any other site has a moderation system as bad as slashdots.
Are you crazy I am NOT going to install Alexia or any other spyware on my computer.
I was a Neilson Basic Peon for a while .
They still OWE ME MONEY for the equipment they Fusked Up. (Little pieces of tinfoil sticking out of all the televisions, I kid you NOT.)
No way!
Not me, nor my children will ever allow something like that again, ever!
No problem. You can use my goldy meter or my bronzy meter. They're just like your irony meter, only they're made of gold and bronze.
John
Putting parameters after the url is the usual way its implemented, and it tells the caching proxy that this is NOT static content
In many situations, yes. But this is not the case when the response includes expiration times, which is very common for content like images.
If they don't re-fetch your ad, you don't pay, since if they don't fetch it, they never see it.
That's an incorrect assumption on two counts. Firstly, caching proxies can still serve cached resources that aren't fresh in some circumstances. Secondly, they can revalidate the content without fetching it. The second situation would be registered in your server logs, so it's merely imprecision on your part, not an important part of the argument.
Also, it doesn't matter if the average user cycles through a dozen IPs, over the course of a day - if they view your ad once, because it will come from just one of those IPs, not multiple IPs.
Your original claim was that "you can analyse the log files and factor out multiple views per host ip to get the actual number of real views". Multiple views per host IP can correspond to one user or many.
Of course, you can also just look at the different IPs the various requests for different parts of the page are passed through, and build up a map of IPs that all front for the same user.
You do realise more than one user is typically behind a proxy, right?
No solution is perfect, but this sure is better than nothing.
The idea that something is better than nothing is very dangerous with statistics. Wrong, misleading data is worse than nothing. At least with nothing you know you don't have accurate data. To what degree are you accurate? Do you know? Do you have any way of measuring? Are you 5% out? 50% out?
Now it might be that for the purpose of fraud detection, these kinds of heuristics work well enough to raise suspicions for further investigation. But as a general tool to count how many visitors or page views you get, no, it's not even close to being good enough. No data is better.
An alternative is to use Quantcast's rating service. You can place their javascript tag on your pages to get accurate 3rd party measurement.
"And frankly, if we're not willing to provide the information necessary for advertisers to make informed choices, we're going to continue to be ignored, both on the web and on television"
I think the real problem is we know the threat of "big brother" does not really come from government in my opinion but private industry, greed outweighs any sense of privacy. Companies want privacy for themselves to hide or obfuscate their operations and their goals, but when it comes to the bottom line privacy be damned.
Adam Smith strongly disliked both governments and corporations. He viewed government primarily as an instrument for extracting taxes to subsidize elites and intervening in the market to protect corporate monopolies. In his words, "Civil government, so far as it is instituted for the security of property, is in reality instituted for the defense of the rich against the poor, or of those who have some property against those who have none at all.'' Smith never suggested that government should not intervene to set and enforce minimum social, health, worker safety, and environmental standards in the common interest or to protect the poor and nature from the rich. Given that most governments of his day were monarchies, the possibility probably never occurred to him.
The theory of market economics, in contrast to free-market ideology, specifies a number of basic conditions needed for a market to set prices efficiently in the public interest. The greater the deviation from these conditions, the less socially efficient the market system becomes
The Compete graph says that Slashdot gets a tenth of the number of users that Dig gets every month, since an upward trend staring last October. That is presumably when digg found Compete and began to game (out-compete?) Compete. Alexa says that Slashdot and digg have similar trends; which is really all Alexa can make claims about.
Irony used in correct sense shocker!
By that metric, you're dumber than just about any drunk-ass bum. 'Cause hey, you won't understand a word they say. Dumbass.
Really? In the same paragraph we're told that IE users are only 25% of
So where the hell does the notion come from? Has anybody actually measured recently? (Need I say, "With some real attempt to eliminate sources of bias.")
New mod option wanted: -1 DrunkenRambling
As has been no doubt said, web metrics is as imprecise an art as TV metrics, and in the end it only serves as a wet finger in the air to provide rough data to another industry that thrives on educated guesses. I haven't used Alexa for years (I've been Linux or Mac and Firefox/Mozilla/whatever for a long time) but I recall back then that the toolbar didn't just do whatever it did, but was crammed with third party apps which could only be described as spyware and which on occasion could pull down a machine while the various bits called home but even now it seems that there is no alternative to a degree of spying either on the desktop or on the server side, and while it's a service that advertisers think they need, it's one that they will pay for and for which web developers will put their hands in the stream to catch some of the money. Also, to be fair, Alexa helped found The Wayback Machine so they have returned something to the web.
Also, people like me would just vote every ad down until we didn't have to see anything
That is not what will happen, you will be just as likely to see all adds, as opposed to those ones that suit you. It is in your best interest, as it were, to only vote the ones down that will never apply to you.
The thing I do not like about Amazon's recommendations is that when you buy an 'X', where X is something that you are likely to only want one of (such as digital camera, computer monitor, PVR etc), it recommends lots of other X's. Even with books, CDs and DVDs if you buy one version/edition it will often recommend other copies of the same title. It is good for recommending book, CD & DVD titles which might be of interest, but for other goods I find it is poor.
To avoid things like this, you can tweak your settings. Go to your recommendations and it will show you the thing you purchased; you can then de-select "use this for recommendations" and it will forget that you ever purchased it, and not use that as a criteria for recommending things. I also make sure to do this when I buy a gift for someone, which is nothing I would ever want for myself.
As others have mentioned, the system isn't perfect. But if something similar were applied to TV ads, I think it would be better than being inundated with ads that I am completely not interested in. Both for me, because it wouldn't be wasting my time with products and services that I will never want or need; and for the advertiser, because I'm less likely to feel the need to skip every ad. They would get a targeted audience, which would be much more valuable to them.
If it is made a legal requirement to vote then it should also be a requirement that every ballot paper have a 'none of the above' option. A win for 'none of the above' would both make all the defeated candidates ineligible to stand again and require that parties to rethink their policies/manifestos.
When he says I am routinely asked questions by a variety of people that lead inevitably to Alexa, I'm sure he means advertisers. And in particular, advertisers who don't want to pay as much for an ad here as on Digg or, for that matter, PerezHilton.com. I suspect the point of this rant is to have a convenient url handy for nitwit advertisers to look at who have been conned into thinking Alexa actually matters. So in that sense, this rant has a major point, one this is likely worth hundreds of thousands of dollars annually to Slashdot.
I agree they need to know that two separate editions are really the "same book". They should definitly have a button for "this book is really the same as that"
No, that was because she was a Current 93/David Tibet fan.
Yes, I am also a goth, yes, I am also a fan. Might be one of the most interesting products of the 80ies goth scene. Listen in, should be appealing to geeks.
And try to remember what Crowley said about the blue and the gold. Theres more to goth than black clothes.
Regards,
RHK
Nobody writes jokes in base 13. - DNA
Insofar as ads provide the income stream of resources you/we value (slashdot, for example), it makes all the difference in the world.
Insofar as you consume no ad-supported goods, services, or media, then it doesn't matter to you in the slightest. In current American society, however, being a consumer who foots the entire bill for everything you consume (as opposed to having your consumption subsidized by advertising), is the exception rather than the norm.
So you may well be right, it doesn't affect you at all. There is a large group of people in the demographic to which I'm referring, however, that it does affect. Such as the group of people who would have liked to see more Firefly, as an example. Or the group of people who enjoy reading slashdot and do not bother to subscribe.
Reality has a conservative bias: it conserves mass, energy, momentum...
I said "if the proxy isn't broken" - caching proxies that serve up urls with parameters that differ for each request are clearly broken. If you're generating the parameters dynamically/pseudo-randomly (which you should be doing) they shouldn't be cached.
If you use a javascript xmlhttprequest to fetch the ad content from your server(a 2-line paste on the web page, to include your source script, which generates the pseudorrandom parameters for each request, and less than 20 lines on the host), then caching proxies don't even come into it, and you get an accurate-enough count.
Kevin Smith on Prince
We did it at the last place I worked at. I didn't trust their figures, and going through the logs, it was obvious they were grossly inflating their claimed click-through results - by almost 2000%. When confronted with the hard evidence, they backed down, apologized, gave us a free month, etc. There were days that they claimed we had received plenty of click-throughs, but there wasn't a single log entry. In other words, they weren't accurately parsing their log files (they were running their servers on Windows, so who knows what they were doing ...) or they were committing out-and-out fraud. Our test clicks showed up in the log files as being relayed via their server, so they couldn't say that there was something wrong with our setup.
2 years later, they're out of business. I guess others cottoned on to the same thing ... or (more likely) my old boss started bragging about how he caught them fudging their click-through rates and page views, and others started putting their own checks into place.
Click fraud is a problem, but so is site fraud.
Kevin Smith on Prince
http://www.evilindenial.com/
Whether or not anybody here agrees with how Alexa works, the fact of the matter is that it /does/ work. It isn't because it gets a good concensus of the Internet as a whole, but because it gets a good concensus of the primary target for businesses and advertising.
I'm not sure that many major businesses are going to be targeting more advanced and knowledgeable crowds (dealing with physical products right now, not with the Nielsen ratings and DVRs and such). I will admit that my knowledge here is quite limited, but from what I've seen, the three major demographics in the case of online advertising and businesses (the casual user, the average Joe and the advanced/knowledgeable user) are all on a different field for the advertisers to make money from.
The average Joe users tend to be relatively easy to make money from, along with the casual user, and are more likely to take advantage of Alexa. Those average Joe's will generally pick up something new relatively quickly in comparison with the casual users, but their opinions will trickle down to the casual users later on, giving businesses who would use Alexa an early and later source of revenue.
As has been mentioned, though, the early adopters tend to most often be knowledgeable users, as well as long-term sources of income. Except for when something isn't quite up to par.
For example, if a company is advertising, say, a 4GB flash drive for $30, the average Joe and some casual users would probably jump on the deal, and while the knowledgeable users might be interested enough to buy one, they'll probably be the ones to do the research to find out that said drive only has single-digit read/write speed. That might be enough to turn some of the advanced users from buying the drive, opting for a smaller or more expensive drive with better performance.
Along the same lines, a company selling a high-capacity HDD at a very low price would likely get in a large number of average users and some casual users, while the more advanced users might recognize the company as one who tends to have low-quality products with weak warranties.
What does all this matter? While most companies (at least in the States, I apologize for not being knowledgeable about non-U.S. businesses but I can't imagine they would be much different) are indeed concerned about the future, many are more interested in there here-and-now more than they are the future. If the company is losing profit over several years, investors will tend to abandon that company in favor of one who is turning a profit right away, even if the first company, the one taking a loss now, is building up a solid reputation and loyalty from the advanced users who will make the company far more money over the long-term than the second business who is catering only to the casual and average users. Of course, once the first company starts turning a solid profit, investors will be interested again, but if the company was abandoned badly enough in the beginning, it may never get the chance to turn a profit.
Simply put, Alexa works because it gives information on the people who will turn a profit early on and continue to turn a profit over time, regardless of whether or not that profit is as high as it could be.
Anyway, sorry if this isn't all that comprehensible. I'm not all that great with putting my thoughts into words.
Calling a sword by a pretty name is no more than adding perfume to poison.
caching proxies that serve up urls with parameters that differ for each request are clearly broken.
You were really unclear on this. What you said was:
Putting parameters after the url is the usual way its implemented, and it tells the caching proxy that this is NOT static content, and to pass through the request instead of using the cached object.
The presence of a query string, without an expiry date, means that an object shouldn't be considered fresh. That's what I thought you were referring to.
"URLs with parameters that differ for each request" is a totally different thing. That doesn't tell the caching proxy not to use the cached object. Why? Because it's not the same object at all.
If you use a javascript xmlhttprequest...
you get an accurate-enough count.
How have you measured "accurate enough"? "Gives me numbers I like" doesn't count.
Some caching proxies are broken in that they WILL serve up the same file if the parameters after the URL are the same; that's why I advocate giving different parameters each time. Now, if the web browser doesn't support javascript, they don't pull the ad. If they don't pull the ad, it doesn't count for an ad view, any more than a browser running adblock+, which also won't pull, say, banner ads. No advertiser should have to pay for ads that aren't requested by the user's browser, right? After all, if the page is served, but the ad isn't, then its time to either move to a different technique, or a different site.
The javascript method is simple to do, and it provides a nice log. The way I've implemented it at a couple of sites is that, for each request, it logs the query string and date/time, etc. into a log file. If the site has N page views, but only N-100 requests get to the javascript ad, then only N-100 count. Fair to everyone. If the request for the ad gets filtered out, why should the advertiser pay for it? It was never delivered. Ditto for people who disable javascript - advertisers shouldn't pay for that particular "page view" since the advertiser's ad was never delivered.
Fairness to all parties is what we should be striving for ... and until something better comes along, this is about as good as its going to get. Of course, if you have a better idea ...
Kevin Smith on Prince
I haven't much faith in compete.com's statistics. It shows my friend's site has 33,761 people a month when he really gets over 80,000 visitors a day.
Some caching proxies are broken in that they WILL serve up the same file if the parameters after the URL are the same
That's not broken. Read the RFC. The proxy should not consider the object to be fresh, but it is permissible in some circumstances to serve stale responses.
Now, if the web browser doesn't support javascript, they don't pull the ad.
JavaScript isn't a black and white issue. You said that you were using XMLHttpRequest. Are you aware that there are plenty of situations where normal JavaScript works but XMLHttpRequest doesn't? Older browsers are like this. Internet Explorer 6 and below with ActiveX switched off are like this. Many companies have group policies that cause this. Some corporate proxies only partially filter out JavaScript, which can cause this. The end result? Even if your adverts are being shown via JavaScript, that in no way proves that your JavaScript that reports back will function.
Now, how are you accounting for these situations? How have you measured how often this is the case? I keep asking how you are establishing how accurate your numbers are, but you keep ducking the question. Do you have any idea how accurate your numbers are? At all?
and until something better comes along, this is about as good as its going to get.
See my point about no information being less harmful than wrong information.
Why is this funny? I searched, and found no instance of digg on the post, yet it was 100% aimed at the whole 'slashdot is no longer relevant'.
Well, you know what? it isn't. News is news. We want fast filtering of information for us. Digg doesn't even do this. Nor reddit. There are rss tools to help you do this, but having people, a democracy, SAY what is good will always help.
Then if you can trust people.
The sad thing is, SLASHDOT GOT THIS RIGHT WITH ITS COMMENTS.
I have yet to see slashdot's comment system beat. Sad that digg doesn't use this. We need a trust network for our own sanity and productivity.
Now I just log in rarely when I find posts like this.
"Even if your adverts are being shown via JavaScript, that in no way proves that your JavaScript that reports back will function."
If the ad is being served via an xmlhttprequest (which is the way I wrote one adserver), if the browser doesn't support xmlhttprequest, or the request is blocked, the ad doesn't get shown, the ad starts aren't incremented, and there is no log entry.
Its a solution that works. No big deal :-)
Kevin Smith on Prince