Episode 71 of The SitePoint Podcast is now available! This week your hosts are Patrick O’Keefe (@iFroggy), Brad Williams (@williamsba), and Kevin Yank (@sentience).
Listen in your Browser
Play this episode directly in your browser — just click the orange “play” button below:
Download this Episode
You can also download this episode as a standalone MP3 file. Here’s the link:
- SitePoint Podcast #71: The Revolving Internet (MP3, 47.7MB, 52:01)
Subscribe to the Podcast
The SitePoint Podcast is on iTunes! Add the SitePoint Podcast to your iTunes player. Or, if you don’t use iTunes, you can subscribe to the feed directly.
Episode Summary
Here are the topics covered in this episode:
- WordPress vs Thesis: The GPL and WordPress Themes
- Google Monitoring Mouse Movements?
- The New York Times Calls for Government Oversight Over Google Search
- Reddit Questions Analytics Used by Advertisers
- MSNBC Redesign Sacrifices Pageviews For Usability
Browse the full list of links referenced in the show at http://delicious.com/sitepointpodcast/71.
Host Spotlights
- Brad: The Revolving Internet
- Kevin: humans.txt
- Patrick: Crush the Castle 2
Show Transcript
Kevin: July 23rd, 2010. A war of words in the world of WordPress, Google might monitor your mouse, and the search for sensible traffic stats. This is the SitePoint Podcast #71: The Revolving Internet.
And welcome to the SitePoint podcast! A strangely non-technical show this week. I’m usually the one who contributes all the code-y technical subjects, guys, and then I look at them and I go there’s no way we can cover that in a podcast; no one wants to hear me read out code on a podcast. But there’s none of that this week, it’s all this search stuff, there’s analytic stuff, and Brad, there’s a legal issue.
Brad: There’s a storm a brewin’ in the WordPress world as they say, and it’s actually between the WordPress platform and a very popular WordPress theme called Thesis. And basically what it boils down to is WordPress itself is released under the GPL license, version 2 specifically, which essentially states anything that is a derivative work of WordPress also has to be released under GPL, that’s essentially how the GPL license functions. The problem is that Thesis is not licensed under the GPL; it’s actually under a different software license, a proprietary license, so WordPress and more specifically Matt Mullenweg who’s one of the founders of WordPress, one of the original developers, has actually gone on the attack against Chris Pearson who is the developer of Thesis, essentially trying to get him to switch his license over to be GPL compliant.
Kevin: And he’s having none of it.
Brad: Absolutely none of it. And this all started basically with a battle back and forth on Twitter as they normally do these days.
Kevin: This issue of selling themes, it’s something I’ve heard come up now and then. So this isn’t a new issue, it’s been a few years this has been debated, right?
Brad: Yeah, and just to be clear, and this is just reading a lot of the blog posts, and there’s a lot of them about this topic, this isn’t specifically about selling themes or making money from themes or making money off WordPress at all, it’s all about the license. So you can certainly have a GPL compliant theme or plug-in and sell it and be completely within your right under the GPL.
Kevin: You can?
Brad: Yeah, absolutely. The only thing that it states is if you do sell it you have to provide the source code, the original source code, with that like.
Kevin: Ah.
Brad: So then if I were to purchase your theme I could turn around and modify it however I want without worry about breaking any kind of software license, and I could redistribute it if I want. So it’s definitely a tricky topic because there hasn’t been a lot of court cases around the GPL, there’s been a few but no kind of definitive this is the answer. So what it really boils down to is a theme in WordPress, or any other platform, is that considered a derivative work of the WordPress platform itself? And that’s really what the argument is.
Kevin: The two perspectives as I read it is that on Matt Mullenweg’s side he considers a theme to be a piece of software that uses WordPress as a library. And so if you install Thesis on your WordPress blog then someone comes and wants to view a particular blog post, Thesis fires up and says, oh, WordPress I need the text of the post, WordPress I need this, WordPress I need I need that. And every time it asks WordPress for something it’s using a piece of the WordPress functionality, and if you do that sort of thing with a piece of GPL code it is well accepted, if not necessarily tested in court, as you pointed out, but it is well accepted and understood that if you do that, as you say, you have to release the source code of the software that you build. The opposing point of view from Chris Pearson seems to be that themes are not pieces of software that use WordPress; rather WordPress is like a runtime that runs themes. Just like if you buy a record player and you set your vinyl LP on top of it and it starts spinning, that LP is not built on top of the record player, the record player is a tool for playing LPs and therefore if the record player were constructed under GPL you wouldn’t have to release your vinyl record design. I know I’m stretching the metaphor terribly here, but that’s kind of — that’s what he is proposing. And it doesn’t seem like he has the force of the masses on his side with that one. I know I’ve contributed to one or two open source projects over the years, and GPL code, well, if you’re working on a GPL project GPL is taken very seriously. And it seems like it’s taken even more seriously when you’re not working on GPL code. I contributed to a project that was released under a much less restrictive license than GPL, and every time I contributed something you had to sort of sign an electronic waiver that said, yeah, none of the code that is in my contribution is licensed under GPL. Because if it were, they call it a viral license because if you use anything in your project that is GPL licensed your whole project has to be GPL licensed.
Brad: And that’s one of the points that after the big blowup through Twitter and Andrew Warner of Mixergy actually had Chris Pearson and Matt Mullenweg on live to kind of battle back and forth almost, all of the core devs of WordPress and some other developers actually dove in the Thesis code, and they found full chunks of WordPress code that was almost line for line pulled directly from WordPress and then used within Thesis.
Kevin: Ouch. Yeah.
Brad: So that’s obviously a clear infringement on the GPL. Now without that is, you know, if that code hadn’t existed, and Pearson states it won’t exist in the next version that he’s working on, but even if it doesn’t exist is it still considered a violation of the GPL if his theme is not a part of the GPL? And this is, I mean this is a really interesting topic especially if it goes to court because there’s a lot of very popular open source platforms like through Drupal and Joomla that are all licensed under the GPL. So this could really set the precedent on what — is a theme part of the GPL; there’s a plug-in or module considered that has to fall into that GPL. So it’s certainly one we’ll keep our eyes on.
Kevin: It’s something that can be blundered into, I think, as a naïve software developer. It’s usually, you know, when you set out there to build your very first WordPress theme I suspect that software licenses are not the first thing on your radar. This is something that Mullenweg has said, that if you want to contribute to a blog platform and the core beliefs of the GPL, the culture of the GPL, is not something you agree with then you should be looking somewhere elsewhere than WordPress. And I kind of agree with that. There are open source projects out there that are in the GPL world and then there are those that aren’t, and there’s kind of a really strong wall between those two. If you want to play in the arena of GPL software like WordPress, if you want to be a developer in that, you have to respect the hundreds and thousands of other developers who have given of their time to build a complicated software product like WordPress, and all of them did that with the understanding that their code was being contributed to something that was protected by the GPL so that no one could benefit from their work without paying it forward, so to speak, without releasing their further work to further benefit the masses. This is the theory of the GPL that if everyone is working agrees with this then everyone is working towards the greater good, but for someone like Chris Pearson to come along, build Thesis and say, no, I’m drawing the line here, my work is mine to benefit from and I don’t care whose shoulders I’m standing on top of, that’s I think what the emotional issue is here and why so many people are upset about this.
Patrick: I’ve been running phpBBhacks.com for ten years, so I’ve seen far, far more open source politics than I care to see in my life.
Kevin: (laughs)
Patrick: And they can get very, very highly charged, very political, very nasty at times. I mean you think open source you think flowers and roses and things, but that’s not exactly how it always works out. I don’t necessarily agree with everything that’s happening with regard to WordPress and how they’re handling this necessarily, I think maybe some of this is a disservice; they’ve been on sort of a crackdown lately. And I really started to talk about it and think about it at WordCamp Raleigh in May just because of different things that had been happening around the WordCamps where they actually said that if you ran a non-GPL theme you could not speak at a WordCamp in the world period. So that was an interesting thing. I think you can go too far; when you’re an open source project you have to be careful how you exert control. I think this is a subject that a lot of people can rally around, but there are a lot of gray area subjects where applying the GPL in letter, in spirit, is maybe not the best thing to do because when you are so open and you are so freely available, when you do exert that control finally you have a group of people who is used to doing whatever they want for the most part and now they see you as controlling them. So you have to be really careful. On this particular issue I do think that — I don’t think Pearson is correct, and this is what we’ve always done at phpBBhacks.com as well, and phpBB is GPL and a very popular GPL project at that, we’ve always treated the code of the style as GPL, graphics and CSS not necessarily, graphics not by default certainly, because graphics obviously work on their own, they don’t appropriate any of the phpBB code, so those have always been treated by us at least to be copyrighted elements, whereas the theme code itself is usually using some form of phpBB code, it kind of has to to function, so that’s always been GPL. We’ve always gone beyond that standard with our database and always asked author permission as well and treated that as very sacred. But code GPL, CSS and graphics not by default I would say, so that’s what I expect to happen here. Now will it take a lawsuit, I think it’s an interesting thing because they have been talking about this for years, as you said, and I was talking to somebody at WordCamp Raleigh, I said at the end of the day talk, I get sick of talk, like are we gonna talk about this forever? Are you guys gonna go in the street and fight, you know, have a fistfight? Are you gonna go to court and file a lawsuit? Just go ahead and do it already because you’ve been talking about this for years and at the end of the day you’re just kind of wasting a lot of people’s time here. So either we’re going to do it or we’re not and we just move on. So I don’t know what the end result is here, but I would like to see an end result.
Kevin: Yeah.
Brad: Along the same lines Matt Mullenweg actually talked with the Software Freedom Law Center to clarify whether WordPress themes fall under the GPL. And their conclusion was actually that the PHP, like you said Patrick, the PHP files that leverage WordPress’ copyrighted code are protected by the GPL; however, images, CSS, JavaScript are not, so you can actually have a split license on a WordPress theme and still be completely within the GPL and still not have to worry about any type of a lawsuit. So it’s, you’re right Kevin, if you’re getting into WordPress themes I mean this is something most people probably never think about until it’s too late and maybe a lawsuit’s in their face. So you can read about this stuff all day and all night, but we’ll see if something happens, if WordPress and Mullenweg step up and actually take this to court which it looks like the next step because there certainly wasn’t any kind of resolution on the Mixergy interview.
Patrick: And you know the thing about it is the Software Freedom Law Center I don’t know them necessarily, but from their title I’m gonna guess there’s a little bit of bias there maybe toward open source projects, maybe, maybe.
Kevin: Maybe.
Patrick: So everybody has their opinion in this, and like I said, opinions have been heard for two, three years now; I mean something has to come to a head sooner or later. Let’s start suing each other or doing something other than browbeating each other using our communities of thousands of people to attack one another, let’s do something else.
Brad: Sue or get off the pot.
Patrick: Basically, yes, but one other thing I wanted to say was that part of the Thesis thing, and I will say I have a Thesis license, I don’t use it, I won’t use it, I just don’t want to use it at this time, but I have purchased it and I’ve looked at it, and part of the allure of Thesis is that the backend features that it has for easy customization. So those tie right into the WordPress admin area so they of course will most likely be GPL. I mean that’s part of the allure of the theme. It’s not a really image heavy theme. Yes, the CSS is part of it is powerful, whatever, but I mean the main part of Thesis that makes it so great to people is that coding, are those features in the admin area. I mean, yeah, it’s definitely a powerful thing. That’s why people buy it — not for images.
Brad: It’s the framework.
Kevin: Yeah, Thesis is one of these themes that it gives you groundwork, a foundation that you are then expected to skin yourself, right Brad?
Brad: Yeah, absolutely, so it’s a theme framework so like Patrick said, I mean even though you can place your CSS and JavaScript in a separate license, technically Thesis there’s probably maybe five percent of the code base behind Thesis would actually fall under that. I mean the majority of it is PHP code, you know, it’s a framework, it’s made so that there is a ton of backend options that you can easily point and click to change your site around. But it isn’t very graphics heavy, it’s not like you’re buying a stylized, designed theme, you’re buying a framework that allows you to kind of develop the theme however you want. And I think that’s probably why Chris Pearson is a little concerned because he can’t really do a split license, it’s not really gonna benefit him the way he wants it.
Kevin: That ruling, quote/unquote ruling, from the Software Freedom Law Center, dates back to July last year, so that’s how long this has been going on in earnest. And the question really is, is Matt Mullenweg gonna jump or is he going to let it slide, and it feels like he’s been letting it slide for at least six months now, or at least a year now. The question is what effect would it have on the ecosystem for Automattic, the owners of WordPress, to go suing a member, a prominent member, of the WordPress development community. But Patrick I tend to agree with you, the damage that’s being done of the arguing back and forth and the uncertainty seems just as bad to me. So perhaps selfishly I would really like them to go forward with the lawsuit; I can say that because I wouldn’t have to pay the lawyer’s bills.
Patrick: Right, and that’s a question I was wondering; who’s going to file for this exactly? Who pays for the lawsuit? Is it money from the WordPress Foundation I guess? Brad you’re probably well-versed on this than I am, but where’s the money gonna come from?
Brad: Yeah, I mean that’s — I can’t imagine it would come from the Foundation because a lot of that money was via sponsorships to various WordCamps and kind of if a WordCamp has extra money they would put it into the Foundation. It’s still a fairly new entity, it’s only been around — I think it was announced back at WordCamp Boston in January, so it hasn’t been around that long. Yeah, and that’s a good question; I don’t know who would pay for it.
Patrick: Right. You have one person in the community who’s visible who disregards the license, let’s say, let’s leave any allegations; let’s just say it’s a certainty he violated the license, he’s spitting in the face of WordPress, whatever. So is that worth suing over and spending that money and spending the time on there? I mean if we just say okay you know what, he’s doing that, we’ve exposed it, people know it, the people who feel strongly don’t like him, good, we got it, he’s been browbeat into the ground. Now, is there, I guess is there some reason to still go forward with the lawsuit? Is it such a firm principle that, okay, we are offended by this, we’re disrespected, we’re gonna spend $200,000 on a lawsuit, or whatever it is, and that’s a good use of our resources and our time. Rather than just saying okay he’s done that, let’s move on, let’s continue to make good stuff. I mean I could see an argument either way, but I’m just saying there’s a lot of time and resources being spent on this by WordPress developers and by people who are instrumental in the creation and maintenance of WordPress itself, so I don’t know if this is — I don’t know if this sort of thing is the best use of resources when we’re talking about an open source, freely available piece of software. Usually these sorts of things are reserved for when people are actually making money and their money is being affected, then people spend money to defend that money. Don’t get me wrong, there’re a lot of people making money here who are involved in this argument on both sides, but I just don’t know that it’s the best use of resources to go forward with that step as much as there may be people wanting that to happen.
Brad: You know even if a lawsuit doesn’t go forward, even if this is the last we hear of it from Matt Mullenweg’s side, you know a lot of damage has already been done towards Thesis. I mean I’ve seen a lot of users coming out that have been using Thesis for a while and had no idea it was in violation of the WordPress license. They had no idea until Matt Mullenweg publicly started talking about it. And now they know and a lot of them I’ve seen have been jumping ship. Now I’m sure it’s a small percentage because there’re a lot of Thesis users, but a lot of people just had no idea there was a conflict with the licenses, so now that they know they’re jumping ship. So maybe that was the intention just to get the information out there and then let people make the decision whether they want to go with Thesis or not.
Patrick: Yeah, and Mullenweg is definitely exerting his influence here and trumpeting every time someone leaves Thesis. And maybe that’s not a big number, but I mean he’s definitely embarked on a campaign of sorts even saying that he will buy another premium license for you if you switch from Thesis. So, you know, that’s how serious it is that we have a license in the ecosystem, or a style in the ecosystem, that is disrespectful to the license let’s say, and that’s how serious it is that his Twitter stream is full with pages and pages of him talking about it or re-tweeting other people switching, or offering to buy themes for other people. It’s just ugly I think and I don’t know what the end result is, like I said, but I don’t enjoy watching it, and I’ve been involved in some minor conflicts myself with regard to phpBB and it’s just an ugly thing, and there’s not really a whole lot of winners in that sort of situation.
Kevin: Is Google monitoring your mouse movements? There was an email that went around at SitePoint HQ last week — because the search engine optimization blogs it’s not something I usually do with my spare time, I don’t read these a lot — but an email went around going just when you thought you had some sort of grip on your search ranking it looks like Google might be monitoring the mouse movements of people who do a Google search, land at a search results page and then are poring over that list of results. A patent has been granted to Google according to seobythesea.com for doing exactly that. The idea is if you hover your mouse over a particular search result on that results page chances are you are reading it, taking it in, and as a result that can be considered a more valuable search result than the ones that you don’t even spend half a second glancing at. The theory that’s advanced in this article is that you might actually find the information you were looking for in the little snippet that’s displayed on the Google search results page, and so even though you never actually click through to that search result it’s still to be considered a high quality result that could be further prioritized in future searches for the same thing by other users. This all sounds well and good, but I got curious and went looking on the Google search results pages and digging through the code, and I can’t find any evidence that they’re actually doing this. But would it be a good idea?
Patrick: Well, I’m on Google right now doing a search for Kevin Yank and circling my mouse over kevinyank.com repeatedly over and over and over again trying to add to your juice there. No.
Kevin: You do that. Just put your mouse there on my homepage and go away for awhile.
Patrick: Go have dinner.
Brad: That’s what I see these SEO firms just doing vigorously like moving their mouse over their own websites and their client’s websites.
Kevin: (laughs)
Patrick: That’s going to be a job listing we’re going to see explode: Mouse Mover, SEO Corp.
Brad: After I read this article I wasn’t sure actually if I used my mouse when I’m reading results.
Kevin: Yeah.
Brad: So I actually went to Google and did a couple searches just to see if I could try to figure out if I was actually doing that, and I don’t think I am unless I’m subconsciously telling myself not to because I’m trying to do this test. I usually just use my scroll wheel.
Kevin: It’s one of those things you can’t actually test it.
Brad: Yeah, you can’t really test it if you’re thinking about it, and if you’re not thinking about it you’re not testing it, so it’s a tricky one, but it’s –
Kevin: Yeah, yep. No, I think so too; I think I’m a scroll wheel guy now.
Brad: That’s primarily what I did. You know what this reminds me of is I remember hearing about how grocery stores would kind of track people’s eyeball movement as they went up and down the aisles at the store and kind of see what they’re looking at and see if their line of sight– It kind of reminds me of that, like a line of sight thing like what are you actually looking at while you’re viewing a web page, but obviously they don’t know where our eyeballs are at but they certainly can look where our pointers are at, so, it’s weird.
Kevin: There was a developer who used to work at SitePoint and to watch him read a webpage was really interesting. He actually highlights every piece of text as he reads it, and I was talking to him about it once and he said it’s a habit that he got into because he worked in an environment — and SitePoint is an environment like this — where it’s very open plan and you can get interrupted at any moment by someone needing your help, and it’s good to be able to just turn away from your monitor in the middle of reading something safe in the knowledge that when you get to come back to it the piece of text that you were reading will be highlighted there on your screen. And so, yeah, just watching him read any web page he would just highlight it sort of one sentence or one paragraph at a time, read it, and you could actually watch him read; it meant he did a lot of extra clicking, but yeah, I wonder, maybe Google should be monitoring people highlighting text, that would be a really meaningful method.
Patrick: And after that conversation you said ‘cuckoo’ and that’s why he’s no longer there.
Kevin: (laughs)
Patrick: No, but you know actually it’s funny you should mention the highlighting because there is a service called Tynt that you may have heard of, I think we have talked about it maybe before, but basically what they do is you sign up with them and hook it on to your webpage, insert some JavaScript I assume, and they will track everything that’s highlighted on your website; articles, pages, I think it’s an option where you can actually have them add a link also to the bottom of any quote so that when someone highlights it and copies it–
Kevin: Ooh, I hate that.
Patrick: — and pastes it they get a link. But I think you can also turn that off and just track highlighting. So in some way that can already sort of be done if you’re interested in that sort of data and you have a developer who does that on your website.
Kevin: I’ve always imagined sort of a social browsing experience where you could see all the other users who were reading the same page you were. And there have been attempts at this in instant messaging clients and things like that before, but imagine you can see little people’s avatar icons moving down paragraph by paragraph through the same article that you were looking at and sort of faded highlights appearing as they were highlighting things. Probably not something you’d want turned on all the time, but man –
Patrick: Sounds like something out of Harry Potter.
Kevin: (laughs) Yeah, it would bring the Web alive. I don’t know if it’s a good use of JavaScript resources, but yes, definitely a strange one. It’s just one of those things that Google got a patent for, they thought yeah it might be a good idea, let’s make sure that we’re the ones who get to do that if we decide to do it. But I don’t think they actually are at this point.
Patrick: So the news media versus Google storyline continues in a way with a report at businessinsider.com by Nick Zane. He says that the New York Times wants the government to start regulating Google’s Search Business. What they mean is that they want the government to watch how Google manages its algorithm so that when they make tweaks to it they are doing so strictly to improve the quality of the search results and not to help Google’s other businesses and their rankings. The Times does say that it would be challenging or impractical to have the government be in charge of approving it every time that there is a tweak, but nonetheless they’d still like the government to be involved with Google’s algorithm changes.
Kevin: And I agree with Business Insider in that this is an insane notion. (Laughs) Business Insider goes on to make a bunch of points about why this is a bad idea, and I think I agree with most if not all of them. But the key one is that at least in what is supposed to be an open market like the United States the government should not be interfering with business decisions of a company like Google, especially if Google does not have a monopoly. And so before, as a first step, I feel like the government would have to decide for itself that Google had a monopoly on search before they would even be able to start looking at the algorithm that Google uses for search. Do you agree Brad?
Brad: Yeah, I mean the last thing we need I think is the government getting involved in Google’s algorithm. There’s got to be a very small percentage of people in the world that would even understand the algorithm, and I’m certainly not one of them, so I can’t imagine there’s that many government types that would be able to even kind of comprehend how it all works, so I certainly don’t think that needs to happen. I agree, if they had 95 percent, 99 percent market share okay then they might have a point, but at this point it’s 63 percent market share, Yahoo! and Bing both creeping up eating into that, you know, if you’re worried about Google don’t use Google. I think it’s just maybe the New York Times is a little bit turned off. You know what’s funny about this article on Business Insider, if I actually click through the link to New York Times it forces me to create an account to view the article; I can’t view any of the article.
So what I did, I took the title of the article put it into Google which promptly found the article for me and then I was able to click right through and it didn’t require me to register. So had Google not found the right result in the first place I wouldn’t even be able to read this article without creating an account; just a little bit of irony I think, but I thought that was kind of funny.
Kevin: This feels like the New York Times didn’t know– This is their editorial so it’s not even attributed to a particular author, in theory this is the voice of the New York Times expressing an opinion, which is weird enough as it is, but it feels like they had a spot to fill for an editorial and they didn’t know what to talk about so they cooked up this half-baked thing. It really isn’t very definitive about anything. To hear it discussed it would sound like the New York Times was saying “For the good of the Internet the government must step in!” But reading the article it’s more like “I think it might be a good idea for the government to have a look at the Google search algorithm.” And it’s so vague and half-hearted you can tell that they’re not even that convinced by their argument. It’s very strange. But the contention that Google would be doing the wrong thing by making a change to their algorithm that would benefit only Google, this is another one of the core points that’s made by Business Insider; Google is a money-making business, that’s the business they’re in. If they can change their search algorithm in a way that makes them more money that’s what Google’s gonna do because they’re out there to make a profit. It just seems like lucky for its users the best way Google has found to make more money is to deliver more relevant search results, at least for now, sponsored results aside and all that.
Patrick: What they should do is put in big red text anything that’s related to the search term that you just put in that they run, so when you search for news at the top there should be a huge red box that takes up 400 pixels by 400 pixels and says Google news, click here now, and then farther down is the rest. But actually right now Google news is fourth for a search for news on Google, so.
Well, speaking of Google, the popular social bookmarking site Reddit put out their Google Analytics numbers on their blog on July 15th. According to the screenshot that they posted, from June 14th to July 14th they had 429 million page views with 36.6 million visits. The reason they did this was because of the inaccuracy, in their opinion, of the numbers provided by some of the stat companies that advertisers sometimes trust when buying ads such as compete.com, Quantcast, Alexa, and Nielsen, to illustrate the disparity between their actual Google Analytics numbers and these numbers on Compete, Quantcast, etcetera. I just gave you the numbers that are in analytics, like I said, 429 million page views, 36 million visits. Well, Compete shows them as getting around 927,000 unique visitors a month; that’s less than a million when analytic shows 36.6 million. Quantcast shows, well they show, let’s see, 13 to 10 million visits a month, that’s visits, so unique visitors most likely. Alexa just shows some weird chart-y thing about daily page views percentage that nobody really understands. And Nielsen finally shows their online market size estimate to be 652,000. This is something that I struggle with myself not being big enough to attract the attention of most of these. I will say with Quantcast you can hook up your site directly to it and be quantified as they call it, so Reddit might want to look into that, I do it myself. But beyond that it seems like a guessing game with what these other sites use to count your tracking numbers.
Kevin: For a while there I was convinced that any one of these services would be fine as long as you just take it as a relative measure against itself. So the fact that you’re number 100 on Alexa doesn’t say anything by itself, but the fact that you’re 100 and your competitor is 90 on Alexa says pretty definitively that they’re ahead of you. That’s what I thought. But over the years my confidence in that has eroded as we’ve heard stories of this week Alexa changed its– I’m not saying it was this week, but a couple of years ago I feel like I heard a story where Alexa changed its algorithm to favor sites with technical content a little more. And that very day you could see Alexa’s graph, the Alexa ranking of sitepoint.com jump, whereas some of its competitors dropped that very same day.
Patrick: But it was a great day!
Kevin: It was a good day for us, we were happy to be on that side of the algorithm change.
Patrick: Drinks all around.
Kevin: But that day I kind of went, oh okay, they’re all just making it up, and it’s a shame.
Patrick: And it’s really based on the toolbar installations, right?
Kevin: Yes. And it’s a shame that advertisers, the data that is available to advertisers, it seems like the more publicly available the data the less reliable it is. According to Reddit, the best numbers they’ve been able to find are through Google analytics which is something that only they have access to.
Patrick: You brought up SitePoint which is a question I was gonna ask without saying too much, I know this isn’t even your area, you’re not really in advertising, but is this something that SitePoint battles with or struggles with, do you know, the rankings in these sites and working to get them to be as accurate as possible?
Kevin: It’s something that when one of these sites isn’t telling a story that agrees with our internal analytics — very much like Reddit, Google Analytics is the final arbiter for us. And our internal traffic numbers are higher than Google Analytics, but by a consistent margin that we put down to browsers with JavaScript disabled that won’t show up in Google Analytics, search engine spiders, RSS feed readers, all of these things that create requests that are services but that Google Analytics doesn’t see. But as long as that margin is fairly consistent, and it tends to be, Google Analytics is what we use, and these other things, the Competes, the Quantcasts, the Alexas, whenever they spit out a favorable number we’ll be sure to include that number in any press package that we’re putting out there, but when we get down to discussing with a potential advertiser, the numbers that we’re talking about are all Google Analytics pre-sale, and post-sale once we’ve signed an agreement to serve up an ad, the numbers that we are following are not, oh, how many page views has Google Analytics told us we’ve gotten this month, it’s all about ad impressions that are measured through our ad delivery software. And that’s something the advertiser can log into and see a report of. And so as the process continues I feel like the numbers that are involved are more and more sensible and more and more true to life. But, you know, it is a shame that a potential advertiser might stumble across SitePoint, see that we have ad space that might appeal to them, and the first thing they might do is go to compete.com, check our Compete ranking against some of our competitors; although the numbers there seem to tell a story, just how meaningful is that story? Looking at Reddit’s numbers it doesn’t seem that meaningful at all.
Patrick: But other thing I noticed was comScore isn’t included in this, and they’re another one, like Jason Calacanis, for example, just railing against them and how they handle their business to verify those numbers. I feel like Quantcast tried to do this, tried to be the one agency that would make it easy for you to verify your numbers because you can sign up with them, you can give them code, you can check my sites and they’re all quantified. I guess is there just no money in this sort of thing for someone to come out and say, okay, we’re going to be the most accurate one, we’re going to work with publishers to get them to sign on to our system and become the definitive source. Because it seems like there might be some room here for someone to come in and be that service; obviously it will take some server power, but beyond that this isn’t the first complaint I’ve heard about these sites. I’ve heard plenty of complaints about these types of services before, but I don’t know, I guess it’s just not an attractive enough business for someone to actually come into and go at it 100 percent right.
Kevin: It’s surprising that Google, you know, if we’re all saying that Google has a really accurate and useful analytics package, it is surprising that Google hasn’t opened up the opportunity to publish some portion of your analytics publicly. If I could tick a box in my Google Analytics package that said release my traffic graph but not with any actual numbers associated with it, release just the relative graph so people could see that, I might choose to do that. What we end up doing and what I guess a lot of people end up doing is once they’re in negotiations with an advertiser, if the advertiser is saying “well Compete is saying this and you’re telling me that,” what you have to end up doing is getting their email address and granting them access to your Google analytics so that they can take a look for themselves and be satisfied. And it’s a terrible situation because that’s not a pleasant thing to have to do.
Patrick: That’s a great point.
Kevin: Yeah, I think you’re right. It’s a business opportunity, Patrick, it’s a business opportunity.
Brad: I’m just glad I don’t have any sites that get real traffic (laughter).
Patrick: This is not a concern for you.
Brad: Yeah, but on the other hand I do have clients that get a lot of traffic, and I can tell you that pretty much every client we have that is getting 5,000 or more hits a day, they typically run multiple analytics package for this very reason just to kind of compare two or three different tracking packages and try to get a good average that they can use in their marketing efforts to bring in advertisers. So it’s certainly a challenge, and you’re right, it would be nice to see somebody kind of come out and say this is — these are the definitive numbers and this is how we’re getting them. But hopefully that will happen.
Patrick: You know it’s possible that Google could go in this direction because on the last episode of the podcast, on episode 69, we mentioned the story of the top 1,000 websites shared by DoubleClick Ad Planner by Google, and essentially it’s Alexa top 1,000 but just done by Google. So, you know, maybe they’ll make some headway into that market, I think they really could.
Kevin: I read a related story to this late in June, and I’ve put it on the list for today because I think it’s really interesting. This is from Mike Industries, and this is a blog that has been speaking out against page views as a metric of traffic. And he’s spoken before about the fact that everyone looks at page views, page views, page views, especially the advertising industry wants to know how many page views you have, and as a result users suffer. And he has written one or two scathing essays on this subject before, but now he is applauding msnbc.com for their recent article page design because essentially it is all Ajax; if you go and view a story at msnbc.com you get to this page and the entire story is on that one page. Not only that, but related content when you click on it in that page it loads in place inside that existing page. Effectively you can spend your whole afternoon reading and viewing content at msnbc.com and only ever make a single page view. And he’s saying that this is a landmark change, it represents the fact that MSNBC has seen fit to throw out page views as a meaningful metric of their traffic, and he wants to see more people do this sort of thing. I’m not sure it’s that big a deal because as he points out the ads are also loaded dynamically as you visit the page, so as you’re scrolling down additional ads are appearing, as you access other content related ads are pulled in and displayed alongside them. The fact that it’s all done with Ajax and the address in your browser bar never changes isn’t that big a difference. You would have to get your advertisers on side by saying yes we have much fewer page views than our competition but look at our ad impression numbers.
Brad: I’d love it. I mean I think it’s an interesting design. You certainly can’t say they don’t make good use of white space, you know, it’s very clean. I think it’s great. There’s nothing more frustrating for me than you see maybe a top ten list on Digg or Reddit or whatever, and it looks interesting and you open it up and it’s every item on the top ten list is a separate page you have to load. I mean it’s so annoying, it slows down reading it; anytime I see that I immediately close the list, I’m not going to click through ten different pages just to see this top ten list. And you know it’s exactly why they do that just to pump up the page view counts when it’s really unnecessary, so it’s nice to see a huge player kind of — whether they’re taking a stand or not it’s nice to see them go a different direction. I really love the image galleries. I’m just looking at this image gallery of the oil spill and there’re 63 images and it’s neat, you hover over the little icon, it shows you a little thumbnail, you click on it, it scrolls to the image, it’s very fluid, I mean I think it’s a nice design.
Kevin: I don’t think it’s that controversial. I do think it’s more work to implement something like this, but even Google Analytics would let you measure interactions with a page like this as page views. You could say that opening the image gallery, count that as another page view; you can ask Google Analytics to do that for you and they will. And so you can even generate page view graphs that look just like you would’ve had before using a design like this. Where you’re going to suffer is in those Alexa’s, those Compete graphs, those Quantcasts, those ones that we were just talking about. And I guess MSNBC decides that they’re a big enough name that if someone is interested in advertising with them the fact that they have a low Alexa score is not going to impact them that negatively, especially if they can go yeah well whatever our Alexa graph is we can guarantee you this many impressions in the next month. And that’s generally how it goes.
Patrick: I agree with everything you’ve said, Kevin, I don’t really see this as a huge deal, I think it’s cool, it’s a nice looking site, and the user experience is nice I’m sure, but right now it works like my blog essentially; you’re on the page, you click to an article, guess what, it’s all there, there’s no pages, any content is embedded. So I don’t have any pages or any of that. We all know sites that go to the extreme like Brad talked about, top ten lists, or I’ve experienced like top thirty or fifty lists where each page is one item, and sometimes I might stick around, sometimes I’ll leave. I don’t know, I don’t think it’s that big a deal; I think there are plenty of sites that do this responsibly, and also what’s good for MSNBC might not be what’s good for the small publisher or the smaller publishers as well. MSNBC might be able to more reasonably afford to do something like this than a smaller publisher would where maybe cutting their page views, even if they’re already being responsible with pagination, and they’re already handling it in a tasteful manner. If they cut their page views by 33 percent how does that affect them selling ads and being able to provide for their staffers and their writers and to have a feasible operation? So as much as down with page views is I don’t really care that much about page views; that’s fine, I’d be happy with that metric going away. It’s something we have to live with and smaller publishers are less equipped to deal with a drop in page views. So I can see it from both sides, and if MSNBC does well and they change the perception that advertisers have then great and I think others will follow. But right now you have to look to those leaders to do those sorts of things.
Kevin: Let’s wrap up this show with our host spotlights guys. Brad, what have you got for us?
Brad: My spotlight is definitely a top contender for the strangest site on the Internet this year, and it is called therevolvinginternet.com. It’s a very strange site because as soon as it loads, and maybe we can bring this up to hear the audio that plays, as soon as it loads a song starts playing and the page actually starts rotating just like you would expect revolving. And it lands on Google and you can actually search, it’s fully functional, you can search Google, you can click through, and the entire time you’re on this website the Internet is revolving so it’s kind of a fun, weird little site, but for some reason I got a good laugh out of it.
Kevin: I’ve got a fun one too, it’s called humans.txt, and this is a little practical joke by a friend of the show and jQuery: Novice to Ninja co-author Earle Castledine, who’s also known as Mr. Speaker. He has decided that it is time for these CAPTCHAs that check that you’re human to come to an end. Obviously we are moving from the Web 2.0 world into the Web 3.0 world where websites are no longer designed for human beings to consume them, they’re designed for marauding software agents to read them and build interesting information and plots against humanity. And so he’s decided that what we need is not a test to verify that you’re human, but a test to verify that you’re a computer, and he has built exactly that. If you go to his humans.txt page which is linked in the show notes, it gives you a fairly easy looking math problem to solve, 1 + 1, and if I type in 2 and then click Start, it starts a counter. I now have 5,000 milliseconds in order to answer 150 further math problems. The first one being what is the bitwise and of 73 and 12 (laughs), and I’ve just ran out of time. It has decided that I am ‘one of them’, one of the filthy bandwidth-wasting humans and it is going to kick me off the site. If you want to see what the machines see there’s an ‘act like a machine’ checkbox that you can check and your browser will automatically solve all the math problems for you, and then it will say ‘one of us’. So if you’re building your next site for computers be sure to include humans.txt to keep those nasty humans away.
Patrick, what have you got?
Patrick: My spotlight is a Flash game. It’s called Crush The Castle 2, it’s on armorgames.com, it was programmed by Joey Betts with art by Chris Condon of Con Artist Games, and it’s a sequel to Crush The Castle, it was a very popular, still is, on armorgames.com, a very popular game where you try to, of course, crush castles with stones, boulders, and different objects. The second game has more castles, more objects, and it’s just a really fun game; a good way to take a break, let’s say, from programming or something along those lines. So, yeah, check it out if you enjoy Flash-based games you’re definitely — it’s a couple of hours I would say to beat it from start to finish, but it’s a lot of fun.
Kevin: I haven’t seen this game. Is this a top down view or is it a side-on view of the castles?
Patrick: It is, it’s side-on, and you basically have a catapult on one side and a castle on the other, and you start with a crude object like a piece of wood and you move all the way up to something that creates a vortex that sucks everything in; that’s the last thing you get after you beat all castles because that demolishes everything in sight. And it’s just a lot of fun, a lot of fun options in it, and you can even create your own castles and play other people’s castles online in your browser. So you can definitely get a lot of replay value out of it as well.
Kevin: That’s great. I’m still waiting for someone to make Rampart, does anyone remember Rampart?
Brad: I remember Rampart, that’s been awhile.
Kevin: I love that game. I love that game, where you and your friend would each design a castle and then you’d have to bomb the crap out of each other basically.
Brad: It was real; it wasn’t a video game, it was a board game, right?
Kevin: It was a board game!
Brad: You’re like flinging boulders at each other across the room.
Patrick: See, I didn’t even know what that was until I Googled it and I got Rampart the arcade game by Atari.
Kevin: Yeah, yeah, that’s what I remember. We’re gonna have to play the real thing, Brad.
Patrick: If we ever get together, the four of us, we’re breaking out Rampart. (laughter) And streaming it live.
Kevin: Excellent. I think that’s the end of the show guys. Who are you and where are you?
Brad: I’m Brad Williams from Webdev Studios, my blog is strangework.com, and I’m Twitter @williamsba.
Kevin: That blog isn’t powered by Thesis, is it?
Brad: No, I can promise you that.
Patrick: And I am Patrick O’Keefe of the iFroggy Network. I blog at managingcommunities.com; find me on Twitter @iFroggy.
Kevin: Your Alexa ranking is looking especially good this week, Patrick.
And I am @sentience on Twitter, and you can follow SitePoint on Twitter @sitepointdotcom. Visit the SitePoint Podcast at http://www.sitepoint.com/podcast, leave your comment on this show, you know, let us know what you do for traffic numbers. And be sure to subscribe to receive every show automatically.
This episode of the SitePoint Podcast is produced by Carl Longnecker, and I’m Kevin Yank. Thanks for listening. Bye, bye.
Theme music by Mike Mella.
Thanks for listening! Feel free to let us know how we’re doing, or to continue the discussion, using the comments field below.
Related Posts
Comments